SpatiotemporalVariabilityintheHydrometeorological Time...
Transcript of SpatiotemporalVariabilityintheHydrometeorological Time...
Research ArticleSpatiotemporal Variability in the HydrometeorologicalTime-Series over Upper Indus River Basin of Pakistan
Muhammad Yaseen1 Ijaz Ahmad 2 Jiali Guo 3 Muhammad Imran Azam3 andYasir Latif4
1Centre for Integrated Mountain Research (CIMR) University of the Punjab Lahore 54590 Pakistan2Centre of Excellence in Water Resources Engineering University of Engineering and Technology Lahore 54890 Pakistan3College of Hydraulic and Environmental Engineering China ree Gorges University Yichang 443002 China4Key Laboratory of Tibetan Environment Changes and Land Surface Processes Institute of Tibetan Plateau ResearchUniversity of Chinese Academy of Sciences Beijing 100101 China
Correspondence should be addressed to Jiali Guo jialiguoctgueducn
Received 2 November 2019 Revised 1 January 2020 Accepted 10 January 2020 Published 30 April 2020
Academic Editor Anthony R Lupo
Copyright copy 2020Muhammad Yaseen et al(is is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited
(is paper investigates the spatiotemporal variability in hydrometeorological time-series to evaluate the current and futurescenarios of water resources availability from upper Indus basin (UIB) MannndashKendall and Senrsquos slope estimator tests were used toanalyze the variability in the temperature precipitation and streamflow time-series data at 27 meteorological stations and 34hydrological stations for the period of 1963 to 2014 (e time-series data of entire study period were divided into two equalsubseries of 26 years each (1963ndash1988 and 1989ndash2014) to assess the overlapping aspect of climate change acceleration over UIB(e results showed a warming pattern at low altitude stations while a cooling tendency was detected at high-altitude stations Anincrease in streamflow was detected during winter and spring seasons at all hydrological stations whereas the streamflow insummer and autumn seasons exhibited decreasing trends (e annual precipitation showed a significant decreasing trend at tenstations while a significant increasing trend was observed at Kohat station during second subseries of the study period (e mostsignificant winter drying trends were observed at Gupis Chitral Garidopatta and Naran stations of magnitude of 47 13 25and 18 respectively during the second subseries(e annual runoff exhibited significant deceasing trends over Jhelum subbasinat Azad Pattan Chinari Domel Kohala Muzaffarabad and Palote while within Indus basin at Chahan Gurriala KhairabadKarora and Kalam in the second time-series It is believed that the results of this study will be helpful for the decision-makers todevelop strategies for planning and development of future water resources projects
1 Introduction
Tibetan Plateau comprises three major mountainous areas ofAsia that is Hindukush Karakoram and Himalaya (HKH)also known as the ldquothird polerdquo or ldquoroof of the worldrdquo becauseof the massive volumes of recurrent snow and glacial icestorage in its high-altitude basins [1 2] (e upper Indusbasin (UIB) encompasses a huge constituency of hilly areasand water resources originating from this region are ofdecisive significance to the interests of Pakistan (e upperIndus river system is of high importance to sustainable watersupply for large populations located in the lower reaches ofIndus river in Pakistan Being an agricultural country with
heavy population growth there is a great stress on waterresources to meet the food and fiber requirement for thepeople (e satisfaction of irrigation system domesticconsumption and hydropower needs are dependent onwater resources from the UIB and its tributaries (e hugeriver basins of south and southeast Asia rely on summermonsoonal wet regimes downstream but the UIB is de-pendent on melted water from glaciers and snow-fedcatchments [3] If anthropogenic interventions causinggreenhouse gas emissions into the atmosphere persist theaverage global surface temperature may rise from 02 to05degCdecade in the next few decades over UIB [4] (istemperature upsurge in the Himalayan region is greater than
HindawiAdvances in MeteorologyVolume 2020 Article ID 5852760 18 pageshttpsdoiorg10115520205852760
the average global rise of 074degC over the last century [5]Large-scale warming of the earth surface over the last tendecades or even more is also indicated by several researchers[6ndash8] Not only did such high-scale warming affect the globalcirculation patterns but also direct affects occur in localclimatic settings with changes in distribution and charac-teristics of precipitation and temperature [9] Changes varyin space and time domains as affected by local climatic andtopographic settings [10ndash13] (ese spatiotemporal varia-tions in the climatic variables have motivated this study inwhich we aim to assess possible acceleration of climatechanges and related hydrological impacts over UIB
Numerous climate change studies have been carried outover the UIB For instance significant decreasing temper-ature trends were detected during the monsoon season andwarming during the premonsoon season [14ndash16] Moreoverwarming winter and cooling summer trends were foundover UIB however these trends lack a definite pattern ofprecipitation [17 18] However Bocchiola and Diolaiuti [16]argued that winter warming and summer cooling trendswere overstated in earlier studies and were only restricted toGilgit and Bunji stations respectively (ey reported in-creasing (insignificant) precipitation trends over the Chitraland Northwest Karakoram regions and a drying pattern overthe UIB Some recent studies have reported general in-creasing temperature trends during the premonsoon insteadduring DecemberndashFebruary Latif et al [19] stated that thedecreasing precipitation trends dominated over the UIBspatially and temporally
Archer [20] and Lutz et al [21] observed that a rise of 1degCin mean summer temperature over UIB has directly affectedthe glacier melt and resulted in an increase of 17 insummer runoff for the river Shyok (basin area 65025 km2)and a 16 increase for the river Hunza (basin area13925 km2) respectively (e mean temperature in the UIBhas increased over the last century however long-termpersistent trends (gt100 years) have not been detected[15 16] Also the statistically significant (plt 005) increasesin winter maximum temperature of 027 055 and 051degCper decade were observed at Gilgit Skardu and Dir in theUIB (e climate change impacts are projected in a semiaridclimate for the upper Indus basin rises of annual meantemperature up to 48degC and 18 increase in annual pre-cipitation by the end of the 21st century [22] (ese findingsindicated that warming is more pronounced in the Astore(54degC) and Gilgit (54degC) river basins compared to theSkardu (49degC) river basin in the summer season Projectedchanges in precipitation in the Hunza (+19) and Gilgit(+21) river basins are quite similar but changes in theAstore (113) river basin are comparatively large More-over long-term precipitation and temperature series in theUIB exhibited a complex season-dependent spatial corre-lation structure indicating large differences in climatechange as affected by local climatic and topographic settings[13 23 24]
Glacial melt from the Karakoram region dominated theflows into the main Indus river system [25] and thus changesin precipitation and temperature directly affect streamflowdischarges Archer and Fowler [15 23] showed falling river
flows from the highest central Karakoram watersheds overthe past two decades and suggested that much winter pre-cipitation is going into long-term storage by glacier surgesKhattak et al [17] showed increasing trends in the winterflows with the increase of mean maximum winter tem-perature and a negative trend in mean summer flows Sharifet al [26] observed some effect of timing and magnitude offlows by climate change in the high elevated areas of the UIBFlows are affected by initial snowpack conditions by winterprecipitation at the beginning of the spring snow melt de-rived streamflow and trends in temperature that differseasonally It is noted that findings of these studies generallyrely on time-series consisting of several decades thus pre-venting assessment of aspects of acceleration of climatechanges and related hydrological impacts It is expected thatthe global climate change may have significant impacts onthe regime of hydrologic extremes Hence the design andmanagement of water resources systems should be adoptedto changing hydrologic extremes However most of thestudies conducted over UIB were restricted to limitednumber of stations and data lengths (erefore the numberof hydrometeorological stations was extended depending onthe available data and coverage of entire Indus Kabul andJhelum river basins was incorporated to investigate thevariations in the hydrometeorological time-series
(is study investigates the spatiotemporal variability inthe hydrometeorological time-series data and hydrologicalimpacts over UIB by using MannndashKendall and Senrsquos slopeestimator tests Moreover the impacts of local topographicalsetting and altitude variations on runoff contributions fromglacier melt snow melt and monsoon precipitation werealso evaluated
2 Description of Study Area
(e Indus river basin is one of the worldrsquos largest trans-boundary river basins with a total drainage area of about108times106 km2 [27 28] (e UIB contributes to half of thesurface water disposal in Pakistan depending upon themelted water resources from Hindukush-Karakoram-Hi-malayan (HKH) region (e URB is located within theterrestrial ranges of 33deg 40ʹ to 37deg 12ʹ N and 70deg 30ʹ to 77deg 30ʹE in the mountainous ranges of Hindu-Kush KarakoramHimalaya and Tibetan Plateau [12 17] (ese ranges jointlyhost 11000 glaciers [29] which make it one of the worldrsquosmost glaciated areas with roughly 22000 km2 of glaciersurface area [30](e altitude in the UIRB varies from 200mto 8500masl with an average elevation of 3750m aslcovering a catchment area of 286000 km2 (e research areais considered as the prime source of fresh water for Pakistanand plays a vibrant role in the sustainable economic de-velopment of the country (e location of climatic andstreamflow stations is presented in Figure 1 (e majorsubbasins of the UIB are discussed in the followingparagraphs
21 Shyok and Shigar Subbasins of UIB (e eastern andcentral part of the Karakoram is covered by the Shyok and
2 Advances in Meteorology
Shigar basins respectively About 24 of the area of theShyok river basin is covered with snow [31] Similarly one-third of the area of Shigar basin is covered by glaciers in-cluding the worldrsquos largest glaciers and ice masses and al-most 25 to 90 of the area is covered with snow (e basicsource of precipitation of these two basins is westerly dis-turbances during winter and spring seasons followed by thesummer monsoon intruding during various intervals[3 19 32 33] (is study used the discharge data of Shyokriver at Yugo and Indus river at Shigar for the trend analysisof flows within these two basins
22 Astore and Hunza Subbasins of UIB (e Astore basin islocated in the western Himalayan and Hunza basin is in thewestern Karakoram ranges (e glacial coverage within thesebasins is less than snow coverage as compared to the Shyokand Shigar basins(e glaciers and permanent ice cover withinHunza basin is 28 and 14 which exhibits almost 21 and3 of the total UIB glacial coverage within Hunza and Astorebasins respectively [3 31] (ree high-altitude stations areinstalled within Hunza basin that is Khunjrab Naltar andZiarat while river discharge is measured at Danyior bridge atHunza river (ere is only one climatic station (Astore) in-stalled in Astore basin measuring temperature and precipi-tation data in this basin Discharge data of Astore river atDoyian are used in this study for the trend analysis installed byWater and Power Development Authority (WAPDA)
23 Gilgit Subbasin of UIB (e Gilgit subbasin ranges be-tween 358 and 37 E and 725 to 744 N comprehends easternpart of Hindukush range and drains towards southeast tojoin Indus river (e discharge of Gilgit river is measured atGilgit hydrometric station and at the confluence of Hunzaand Gilgit river which is called Alam bridge (e drainagearea of this basin incorporates 12000 km2 with an elevationrange from 1481 to 7134m masl Four climatic stations areinstalled in this area that is Gilgit Gupis Yasin andUshkore by the Pakistan Meteorological Department(PMD) and WAPDA In this study data at two stations(Gilgit and Gupis) were used from 1960 to 2014 Hassonet al [3] reported that the Gilgit basin receives maximumprecipitation at Ushkore (3151m) and minimum precipi-tation at Gilgit (1460m) due to the westerly disturbances andsummer monsoon
24 Jehlum Subbasin of UIB (eMangla basin is located onthe southern slope of the Himalayas with elevation rangingfrom 300m to 6282mmasl and has basin area of around33425 km2 at Mangla dam (is dam serves hydropowergeneration and regulates the flow from Mangla reservoirAbout 55 of the area lies in Indian held Kashmir and 45lies in Pakistan including Azad Kashmir (ere are fivesubcatchments that is Jhelum Poonch Kanshi NeelumKishanganga and Kunhar which drain water to Manglareservoir
71deg0prime0PrimeE 72deg0prime0PrimeE 73deg0prime0PrimeE 74deg0prime0PrimeE 75deg0prime0PrimeE 76deg0prime0PrimeE 77deg0prime0PrimeE
71deg0prime0PrimeE 72deg0prime0PrimeE 73deg0prime0PrimeE 74deg0prime0PrimeE 75deg0prime0PrimeE 76deg0prime0PrimeE 77deg0prime0PrimeE
37deg0prime0PrimeN
36deg0prime0PrimeN
35deg0prime0PrimeN
34deg0prime0PrimeN
33deg0prime0PrimeN
37deg0prime0PrimeN
36deg0prime0PrimeN
35deg0prime0PrimeN
34deg0prime0PrimeN
33deg0prime0PrimeN
Stream_gauges
Climatic stations
Drainage line
Catchment
198-1000
1000-2000
2000-3000
3000-4000
4000-5000
5000-6000
6000-7000
7000-8000
gt8000
Elevationamsl (m)
Figure 1(e upper Indus basin confined in Pakistan boundary showing rivers elevation streamflow gauges (Table 1) and climatic stations(Table 2)
Advances in Meteorology 3
25 Kabul Subbasin of UIB Kabul river in the easternAfghanistan and northwestern Pakistan is 700 km long ofwhich 560 km lies in Afghanistan It originates in theSanglakh ranges located 72 km west of Kabul city It flowseast through Kabul and Jalalabad north of the Khyber Passinto Pakistan (e river has four major tributaries theLowgar the Panjsher the Konar (Kunar) and the AlingarMost of area of this catchment lies in Afghanistan Due tounavailability of data from Afghanistan the study area wasconfined to the catchment falling within Pakistan boundary(e Kabul river a major western flank tributary joins withIndus near Attock
3 Methodology
31 Data and Methods In this study hydrological time-series data of 34 stream gauges and meteorological data of27 stations for the period of 1963 to 2014 were collectedfrom WAPDA and PMD (e information regarding thelocation of each stream gauge station area of subbasinsand mean annual streamflow is presented in Table 1
Moreover information of mean annual maximum andminimum temperature (Tmax Tmin) and precipitation (P) of27 climatic stations is shown in Table 2 Streamflowmeasurements in the UIB are carried out by WAPDA withthe earliest records commencing from 1960 (e streamgauges have a broad range of drainage area from 262 km2 to286000 km2 (e study area contained three major basinsnamely Jhelum Indus and Kabul (e locations of dif-ferent hydrological and meteorological stations are pre-sented in Figure 1
(e hydrometeorological time-series data of entire studyperiod (1963ndash2014) were divided into two equal subseriesthat is 1963 to 1988 and 1989 to 2014 to analyze the aspectsof acceleration of climate change Mean monthly seasonaland annual values of Tmax Tmin P and Qst were dereivedfrom the daily time-series data To analyze the seasonalvariations in the hydrometeorological time-series data fourseasons were defined as winter (December January andFebruary (DJF)) spring (March April and May (MAM))summer (June July and August (JJA)) and autumn (Sep-tember October and November (SON)) seasons
Table 1 List of stream gauges used in the present study and their characteristics (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Latitude (dd) Longitude (dd) Area (Km2)Mean annual
streamflow (m3s)1963ndash1988 1989ndash2014
1 Naran 349 737 1036 477 4562 Garhi Habibullah 344 734 2355 100 10553 Muzaffarabad 344 735 7275 3423 32194 Chinari 342 738 13598 2987 2895 Domel 344 735 14504 3273 32236 Kohala 341 735 24890 776 78057 Azad Pattan 337 736 26485 11507 124188 Kotli 335 739 3238 1239 12739 Palote 332 734 1111 6 5310 Kharmong 352 759 67858 4627 46511 Yogo 352 761 33670 3412 368812 Shigar 354 757 6610 1946 220513 Kachura 355 754 112665 962 1159614 Gilgit 359 743 12095 2772 333715 Dainyor Br 359 744 13157 3654 29516 Alam Br 358 746 26159 6618 619317 Bunji 357 746 142709 1706 1875318 Doyain 355 747 4040 1183 149219 Shatial Br 355 736 150220 19389 2110620 Karora 349 728 635 204 17521 Besham Qila 349 729 162393 23502 2436822 Daggar 345 725 598 54 5923 Phulra 343 731 1057 186 20524 Kalam 355 726 2020 857 86225 Chakdara 346 72 5776 1691 207126 Chitral 359 718 11396 2644 285427 Nowshera 34 72 88578 849 824228 Gurriala 337 723 3056 269 24829 Khairabad 339 722 252525 32227 2834430 (al 334 715 5543 277 22631 Chirah 337 733 326 57 432 Chahan 334 729 241 17 1333 Dhok Pathan 331 723 6475 44 38434 Massan 33 717 286000 35272 38095
4 Advances in Meteorology
32 Change Detection To detect climate impacts and pos-sible climate change acceleration over the past 52 yearsStudentrsquos t-test was selected for sample means whereas theldquoF-testrdquo was selected to assess aspects of variability on time-series (e nonparametric MannndashWhitney U test was usedto assess aspects of the distribution of observations
321 Studentrsquos t-Test and F-Test (e application of Stu-dentrsquos t-test aims to detect the change in the mean values oftwo 26-year periods for statistical significance F-test wasused to detect the variability in the time-series Both testswere performed at the 90 confidence interval (e t-sta-tistic is calculated by using equation (1) when the variancesof both time-series have similar values
t X1 minus X2
SP
1n1( 1113857 + 1n2( 1113857
1113969 (1)
x1 and x2 are mean values of two subseries n1 and n2indicate number of observations s1 and s2 are the standarddeviations subscripts 1 and 2 indicate the periods 1963ndash1988and 1989ndash2014 and Sp is the standard deviation which isgiven as
SP
n1 minus 1( 1113857S21 + n2 minus 1( 1113857S22
n1 + n2 minus 2
1113971
(2)
If the variances for the two periods are different then thet-statistic is used and given in the following equation
t X1 minus X2
S21n1) + S22n2(
1113969 (3)
322 MannndashWhitney U-Test (e nonparametric Man-nndashWhitney U test [34] is selected to detect the shift in thetemperature precipitation and streamflow time-series data(eMannndashWhitney (MW)U test statistic [35 36] is given inthe following equation
U min U1 U21113858 1113859 (4)
We have
U1 n1n2 +n1 n1 + 1( 1113857
2minus R1
U2 n1n2 +n2 n2 + 1( 1113857
2minus R2
(5)
where U1 and U2 are the total count of samples 1 and 2 andR1 and R2 are the rank sums of sample 1 and sample 2respectively When the null hypothesis H0 is true and n1and n2 are both larger than 8 U is considered approximatelynormally distributed with mean of E(U) and variance ofV(U) given as
Table 2 List of climatic stations in upper Indus basin (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Elevation (m)Max temp (degC) Min temp (degC) Precipitation (mm)
1963ndash1988 1989ndash2014 1963ndash1988 1989ndash2014 1963ndash1988 1989ndash20141 Astore 2168 154 158 4 41 39 422 Bagh 1067 254 199 4 46 13 133 Balakot 9955 154 159 145 142 50 504 Bunji 1372 24 237 14 143 15 175 Cherat 1372 219 211 89 83 33 386 Chilas 1250 266 262 83 78 130 1237 Chitral 14978 228 238 111 114 44 498 Dir 1375 225 233 79 74 11 119 Drosh 14639 238 243 71 61 7 1410 Garidopatta 8135 154 159 93 102 117 11611 Gilgit 1460 235 243 171 17 45 4512 Gujar Khan 457 281 29 94 69 65 6713 Gupis 2156 187 189 158 164 36 3714 Kakul 1308 227 233 93 102 117 11615 Kohat 1440 289 303 171 17 45 4516 Kotli 610 284 284 166 149 1272 118317 Mangla 282 304 309 176 17 35 3918 Murree 2206 163 18 89 84 1765 173419 Muzaffarabad 702 154 159 176 17 35 3920 Naran 2363 141 105 87 96 42 4421 Palandri 1402 154 159 176 17 35 3922 Parachinar 1725 211 213 94 69 65 6723 Peshawar 320 293 297 158 164 36 3724 Rawalakot 1677 20 211 87 96 44 4625 Risalpur 575 295 299 146 142 55 5426 Saidu Sharif 961 256 263 123 119 90 9027 Skardu 2317 18 192 51 47 17 20
Advances in Meteorology 5
E(U) n1n2
2
V(U) n1n2 n1 + n2 + 1( 1113857
12
(6)
323 Relative Changes (e relative change () in theannual and seasonal temperature precipitation andstreamflow was assessed by using the following equation
relative change mean of 2nd period minus mean of 1st period
mean of 1st period
(7)
33 Trend Analysis For detection of trends we (i) pre-whitened time-series to eliminate effect of serial correlationof observations (ii) applied MannndashKendall trend analysisto identify if trends are significant and (iii) assessed thetrend slope line by means of Senrsquos estimator Analysis iscommon and reference is made to applications in[1 2 37ndash40]
331 MannndashKendall Test A nonparametric rank-basedMannndashKendall (MK) trend analysis test was used toevaluate the variations in the hydrometeorological time-series data over UIB [41 42] (e main advantage of theMK test is that there are no assumptions about the sta-tistical distribution of the sample data Since the method isrank-based extreme data points in the hydrometeoro-logical time-series will not largely affect the results(eMKtest statistic (S) is given by
S 1113944nminus1
k11113944
n
jk+1sig Xj minus Xk1113872 1113873
sgn Xj minus Xk1113872 1113873
if Xj minus Xk1113872 1113873lt 0 then minus 1
if Xj minus Xk1113872 1113873 0 then 0
if Xj minus Xk1113872 1113873gt 0 then 1
⎧⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎭
(8)
where ldquonrdquo denotes the length of a dataset and Xj and Xk are thesequential data values at times j and k Positive value of Sindicates an increasing (upward) trend and negative value of Sreveals a decreasing (downward) trend in the time-series data
Var(S) n(n minus 1)(2n + 5) minus 1113936
p
k1 tk tk minus 1( 1113857 2tk + 5( 1113857
18
(9)
where tk is the number of tied values in the qth group and thesign ldquo1113936rdquo represents the summation of all the tied groupsHowever if there are no tied groups in the data then thismay be ignored After calculating the variance Var(S) fromequation (4) the standardized test statistic (Zmk) value iscalculated by using the following equation
Zmk
if Slt 0 thenS + 1
VAR(S)
1113968
if xk minus xj1113872 1113873 0 then 0
if xk minus xj1113872 1113873gt 0 thenS minus 1
VAR(S)
1113968
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎭
(10)
A positive value of Zmk indicates an upward trend (ieincreasing) whereas a negative value indicates a downwardtrend (ie decreasing) (e test statistic (S) follows thestandard normal distribution where probability of ob-serving a value higher than the test statistic Zmk is testedunder the null hypothesis H0 that there is no trend forchosen α-level of significance H0 is rejected if the absolutevalue of ZmkgtZ1minusa2 at the α-level is significant
332 Senrsquos Estimator of Slope Senrsquos nonparametric method[43] was used to estimate the magnitude of trends in thetime-series data (e slope of ldquonrdquo pairs of data can be firstestimated by using the following equation
Qi Xj minus Xk
j minus k1113890 1113891 ifjgt k (11)
Senrsquos estimator is the median Qmed of the N pairs of QiIn the procedure N values of Qi are ranked from smallest tolargest and Senrsquos estimator is determined by using the fol-lowing equation
Senrsquos Estimator
Q[(N+1)2] if Nwas odd
12
QN2 + Q[(N+2)2]1113872 1113873 if Nwas even
⎧⎪⎪⎪⎨
⎪⎪⎪⎩
⎫⎪⎪⎪⎬
⎪⎪⎪⎭
(12)
Q med is tested by a two-sided test at the 100(1minus α)confidence interval and the true slope may be obtained bythe nonparametric test Data were processed using an Excelmacro named MAKESENS [44]
4 Results and Discussions
41 Variability in Temperature Studentrsquos t-test F-test andMannndashWhitney U test were used to detect the percentagechange between two consecutive (26-year each) hydrome-teorological time-series Table 3 showed the results ofmaximum and minimum temperatures (Tmax and Tmin) andprecipitation variables at seasonal and annual scale (eresults of testing are combined with assessment of relativechanges as indicated by percent change (e results showedinconsistent variations in all variables at different stationsand seasons (e annual Tmax during winter and springseason decreased (minus22 and minus27) between two subseries atBagh and Naran stations respectively whereas at Murreeand Peshawar stations it increased by 10 and 20 re-spectively It was noted that the changes in Tmax at seasonalscale were quite high in magnitude as compared to Tminduring the same seasons For Tmax most significant
6 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
the average global rise of 074degC over the last century [5]Large-scale warming of the earth surface over the last tendecades or even more is also indicated by several researchers[6ndash8] Not only did such high-scale warming affect the globalcirculation patterns but also direct affects occur in localclimatic settings with changes in distribution and charac-teristics of precipitation and temperature [9] Changes varyin space and time domains as affected by local climatic andtopographic settings [10ndash13] (ese spatiotemporal varia-tions in the climatic variables have motivated this study inwhich we aim to assess possible acceleration of climatechanges and related hydrological impacts over UIB
Numerous climate change studies have been carried outover the UIB For instance significant decreasing temper-ature trends were detected during the monsoon season andwarming during the premonsoon season [14ndash16] Moreoverwarming winter and cooling summer trends were foundover UIB however these trends lack a definite pattern ofprecipitation [17 18] However Bocchiola and Diolaiuti [16]argued that winter warming and summer cooling trendswere overstated in earlier studies and were only restricted toGilgit and Bunji stations respectively (ey reported in-creasing (insignificant) precipitation trends over the Chitraland Northwest Karakoram regions and a drying pattern overthe UIB Some recent studies have reported general in-creasing temperature trends during the premonsoon insteadduring DecemberndashFebruary Latif et al [19] stated that thedecreasing precipitation trends dominated over the UIBspatially and temporally
Archer [20] and Lutz et al [21] observed that a rise of 1degCin mean summer temperature over UIB has directly affectedthe glacier melt and resulted in an increase of 17 insummer runoff for the river Shyok (basin area 65025 km2)and a 16 increase for the river Hunza (basin area13925 km2) respectively (e mean temperature in the UIBhas increased over the last century however long-termpersistent trends (gt100 years) have not been detected[15 16] Also the statistically significant (plt 005) increasesin winter maximum temperature of 027 055 and 051degCper decade were observed at Gilgit Skardu and Dir in theUIB (e climate change impacts are projected in a semiaridclimate for the upper Indus basin rises of annual meantemperature up to 48degC and 18 increase in annual pre-cipitation by the end of the 21st century [22] (ese findingsindicated that warming is more pronounced in the Astore(54degC) and Gilgit (54degC) river basins compared to theSkardu (49degC) river basin in the summer season Projectedchanges in precipitation in the Hunza (+19) and Gilgit(+21) river basins are quite similar but changes in theAstore (113) river basin are comparatively large More-over long-term precipitation and temperature series in theUIB exhibited a complex season-dependent spatial corre-lation structure indicating large differences in climatechange as affected by local climatic and topographic settings[13 23 24]
Glacial melt from the Karakoram region dominated theflows into the main Indus river system [25] and thus changesin precipitation and temperature directly affect streamflowdischarges Archer and Fowler [15 23] showed falling river
flows from the highest central Karakoram watersheds overthe past two decades and suggested that much winter pre-cipitation is going into long-term storage by glacier surgesKhattak et al [17] showed increasing trends in the winterflows with the increase of mean maximum winter tem-perature and a negative trend in mean summer flows Sharifet al [26] observed some effect of timing and magnitude offlows by climate change in the high elevated areas of the UIBFlows are affected by initial snowpack conditions by winterprecipitation at the beginning of the spring snow melt de-rived streamflow and trends in temperature that differseasonally It is noted that findings of these studies generallyrely on time-series consisting of several decades thus pre-venting assessment of aspects of acceleration of climatechanges and related hydrological impacts It is expected thatthe global climate change may have significant impacts onthe regime of hydrologic extremes Hence the design andmanagement of water resources systems should be adoptedto changing hydrologic extremes However most of thestudies conducted over UIB were restricted to limitednumber of stations and data lengths (erefore the numberof hydrometeorological stations was extended depending onthe available data and coverage of entire Indus Kabul andJhelum river basins was incorporated to investigate thevariations in the hydrometeorological time-series
(is study investigates the spatiotemporal variability inthe hydrometeorological time-series data and hydrologicalimpacts over UIB by using MannndashKendall and Senrsquos slopeestimator tests Moreover the impacts of local topographicalsetting and altitude variations on runoff contributions fromglacier melt snow melt and monsoon precipitation werealso evaluated
2 Description of Study Area
(e Indus river basin is one of the worldrsquos largest trans-boundary river basins with a total drainage area of about108times106 km2 [27 28] (e UIB contributes to half of thesurface water disposal in Pakistan depending upon themelted water resources from Hindukush-Karakoram-Hi-malayan (HKH) region (e URB is located within theterrestrial ranges of 33deg 40ʹ to 37deg 12ʹ N and 70deg 30ʹ to 77deg 30ʹE in the mountainous ranges of Hindu-Kush KarakoramHimalaya and Tibetan Plateau [12 17] (ese ranges jointlyhost 11000 glaciers [29] which make it one of the worldrsquosmost glaciated areas with roughly 22000 km2 of glaciersurface area [30](e altitude in the UIRB varies from 200mto 8500masl with an average elevation of 3750m aslcovering a catchment area of 286000 km2 (e research areais considered as the prime source of fresh water for Pakistanand plays a vibrant role in the sustainable economic de-velopment of the country (e location of climatic andstreamflow stations is presented in Figure 1 (e majorsubbasins of the UIB are discussed in the followingparagraphs
21 Shyok and Shigar Subbasins of UIB (e eastern andcentral part of the Karakoram is covered by the Shyok and
2 Advances in Meteorology
Shigar basins respectively About 24 of the area of theShyok river basin is covered with snow [31] Similarly one-third of the area of Shigar basin is covered by glaciers in-cluding the worldrsquos largest glaciers and ice masses and al-most 25 to 90 of the area is covered with snow (e basicsource of precipitation of these two basins is westerly dis-turbances during winter and spring seasons followed by thesummer monsoon intruding during various intervals[3 19 32 33] (is study used the discharge data of Shyokriver at Yugo and Indus river at Shigar for the trend analysisof flows within these two basins
22 Astore and Hunza Subbasins of UIB (e Astore basin islocated in the western Himalayan and Hunza basin is in thewestern Karakoram ranges (e glacial coverage within thesebasins is less than snow coverage as compared to the Shyokand Shigar basins(e glaciers and permanent ice cover withinHunza basin is 28 and 14 which exhibits almost 21 and3 of the total UIB glacial coverage within Hunza and Astorebasins respectively [3 31] (ree high-altitude stations areinstalled within Hunza basin that is Khunjrab Naltar andZiarat while river discharge is measured at Danyior bridge atHunza river (ere is only one climatic station (Astore) in-stalled in Astore basin measuring temperature and precipi-tation data in this basin Discharge data of Astore river atDoyian are used in this study for the trend analysis installed byWater and Power Development Authority (WAPDA)
23 Gilgit Subbasin of UIB (e Gilgit subbasin ranges be-tween 358 and 37 E and 725 to 744 N comprehends easternpart of Hindukush range and drains towards southeast tojoin Indus river (e discharge of Gilgit river is measured atGilgit hydrometric station and at the confluence of Hunzaand Gilgit river which is called Alam bridge (e drainagearea of this basin incorporates 12000 km2 with an elevationrange from 1481 to 7134m masl Four climatic stations areinstalled in this area that is Gilgit Gupis Yasin andUshkore by the Pakistan Meteorological Department(PMD) and WAPDA In this study data at two stations(Gilgit and Gupis) were used from 1960 to 2014 Hassonet al [3] reported that the Gilgit basin receives maximumprecipitation at Ushkore (3151m) and minimum precipi-tation at Gilgit (1460m) due to the westerly disturbances andsummer monsoon
24 Jehlum Subbasin of UIB (eMangla basin is located onthe southern slope of the Himalayas with elevation rangingfrom 300m to 6282mmasl and has basin area of around33425 km2 at Mangla dam (is dam serves hydropowergeneration and regulates the flow from Mangla reservoirAbout 55 of the area lies in Indian held Kashmir and 45lies in Pakistan including Azad Kashmir (ere are fivesubcatchments that is Jhelum Poonch Kanshi NeelumKishanganga and Kunhar which drain water to Manglareservoir
71deg0prime0PrimeE 72deg0prime0PrimeE 73deg0prime0PrimeE 74deg0prime0PrimeE 75deg0prime0PrimeE 76deg0prime0PrimeE 77deg0prime0PrimeE
71deg0prime0PrimeE 72deg0prime0PrimeE 73deg0prime0PrimeE 74deg0prime0PrimeE 75deg0prime0PrimeE 76deg0prime0PrimeE 77deg0prime0PrimeE
37deg0prime0PrimeN
36deg0prime0PrimeN
35deg0prime0PrimeN
34deg0prime0PrimeN
33deg0prime0PrimeN
37deg0prime0PrimeN
36deg0prime0PrimeN
35deg0prime0PrimeN
34deg0prime0PrimeN
33deg0prime0PrimeN
Stream_gauges
Climatic stations
Drainage line
Catchment
198-1000
1000-2000
2000-3000
3000-4000
4000-5000
5000-6000
6000-7000
7000-8000
gt8000
Elevationamsl (m)
Figure 1(e upper Indus basin confined in Pakistan boundary showing rivers elevation streamflow gauges (Table 1) and climatic stations(Table 2)
Advances in Meteorology 3
25 Kabul Subbasin of UIB Kabul river in the easternAfghanistan and northwestern Pakistan is 700 km long ofwhich 560 km lies in Afghanistan It originates in theSanglakh ranges located 72 km west of Kabul city It flowseast through Kabul and Jalalabad north of the Khyber Passinto Pakistan (e river has four major tributaries theLowgar the Panjsher the Konar (Kunar) and the AlingarMost of area of this catchment lies in Afghanistan Due tounavailability of data from Afghanistan the study area wasconfined to the catchment falling within Pakistan boundary(e Kabul river a major western flank tributary joins withIndus near Attock
3 Methodology
31 Data and Methods In this study hydrological time-series data of 34 stream gauges and meteorological data of27 stations for the period of 1963 to 2014 were collectedfrom WAPDA and PMD (e information regarding thelocation of each stream gauge station area of subbasinsand mean annual streamflow is presented in Table 1
Moreover information of mean annual maximum andminimum temperature (Tmax Tmin) and precipitation (P) of27 climatic stations is shown in Table 2 Streamflowmeasurements in the UIB are carried out by WAPDA withthe earliest records commencing from 1960 (e streamgauges have a broad range of drainage area from 262 km2 to286000 km2 (e study area contained three major basinsnamely Jhelum Indus and Kabul (e locations of dif-ferent hydrological and meteorological stations are pre-sented in Figure 1
(e hydrometeorological time-series data of entire studyperiod (1963ndash2014) were divided into two equal subseriesthat is 1963 to 1988 and 1989 to 2014 to analyze the aspectsof acceleration of climate change Mean monthly seasonaland annual values of Tmax Tmin P and Qst were dereivedfrom the daily time-series data To analyze the seasonalvariations in the hydrometeorological time-series data fourseasons were defined as winter (December January andFebruary (DJF)) spring (March April and May (MAM))summer (June July and August (JJA)) and autumn (Sep-tember October and November (SON)) seasons
Table 1 List of stream gauges used in the present study and their characteristics (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Latitude (dd) Longitude (dd) Area (Km2)Mean annual
streamflow (m3s)1963ndash1988 1989ndash2014
1 Naran 349 737 1036 477 4562 Garhi Habibullah 344 734 2355 100 10553 Muzaffarabad 344 735 7275 3423 32194 Chinari 342 738 13598 2987 2895 Domel 344 735 14504 3273 32236 Kohala 341 735 24890 776 78057 Azad Pattan 337 736 26485 11507 124188 Kotli 335 739 3238 1239 12739 Palote 332 734 1111 6 5310 Kharmong 352 759 67858 4627 46511 Yogo 352 761 33670 3412 368812 Shigar 354 757 6610 1946 220513 Kachura 355 754 112665 962 1159614 Gilgit 359 743 12095 2772 333715 Dainyor Br 359 744 13157 3654 29516 Alam Br 358 746 26159 6618 619317 Bunji 357 746 142709 1706 1875318 Doyain 355 747 4040 1183 149219 Shatial Br 355 736 150220 19389 2110620 Karora 349 728 635 204 17521 Besham Qila 349 729 162393 23502 2436822 Daggar 345 725 598 54 5923 Phulra 343 731 1057 186 20524 Kalam 355 726 2020 857 86225 Chakdara 346 72 5776 1691 207126 Chitral 359 718 11396 2644 285427 Nowshera 34 72 88578 849 824228 Gurriala 337 723 3056 269 24829 Khairabad 339 722 252525 32227 2834430 (al 334 715 5543 277 22631 Chirah 337 733 326 57 432 Chahan 334 729 241 17 1333 Dhok Pathan 331 723 6475 44 38434 Massan 33 717 286000 35272 38095
4 Advances in Meteorology
32 Change Detection To detect climate impacts and pos-sible climate change acceleration over the past 52 yearsStudentrsquos t-test was selected for sample means whereas theldquoF-testrdquo was selected to assess aspects of variability on time-series (e nonparametric MannndashWhitney U test was usedto assess aspects of the distribution of observations
321 Studentrsquos t-Test and F-Test (e application of Stu-dentrsquos t-test aims to detect the change in the mean values oftwo 26-year periods for statistical significance F-test wasused to detect the variability in the time-series Both testswere performed at the 90 confidence interval (e t-sta-tistic is calculated by using equation (1) when the variancesof both time-series have similar values
t X1 minus X2
SP
1n1( 1113857 + 1n2( 1113857
1113969 (1)
x1 and x2 are mean values of two subseries n1 and n2indicate number of observations s1 and s2 are the standarddeviations subscripts 1 and 2 indicate the periods 1963ndash1988and 1989ndash2014 and Sp is the standard deviation which isgiven as
SP
n1 minus 1( 1113857S21 + n2 minus 1( 1113857S22
n1 + n2 minus 2
1113971
(2)
If the variances for the two periods are different then thet-statistic is used and given in the following equation
t X1 minus X2
S21n1) + S22n2(
1113969 (3)
322 MannndashWhitney U-Test (e nonparametric Man-nndashWhitney U test [34] is selected to detect the shift in thetemperature precipitation and streamflow time-series data(eMannndashWhitney (MW)U test statistic [35 36] is given inthe following equation
U min U1 U21113858 1113859 (4)
We have
U1 n1n2 +n1 n1 + 1( 1113857
2minus R1
U2 n1n2 +n2 n2 + 1( 1113857
2minus R2
(5)
where U1 and U2 are the total count of samples 1 and 2 andR1 and R2 are the rank sums of sample 1 and sample 2respectively When the null hypothesis H0 is true and n1and n2 are both larger than 8 U is considered approximatelynormally distributed with mean of E(U) and variance ofV(U) given as
Table 2 List of climatic stations in upper Indus basin (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Elevation (m)Max temp (degC) Min temp (degC) Precipitation (mm)
1963ndash1988 1989ndash2014 1963ndash1988 1989ndash2014 1963ndash1988 1989ndash20141 Astore 2168 154 158 4 41 39 422 Bagh 1067 254 199 4 46 13 133 Balakot 9955 154 159 145 142 50 504 Bunji 1372 24 237 14 143 15 175 Cherat 1372 219 211 89 83 33 386 Chilas 1250 266 262 83 78 130 1237 Chitral 14978 228 238 111 114 44 498 Dir 1375 225 233 79 74 11 119 Drosh 14639 238 243 71 61 7 1410 Garidopatta 8135 154 159 93 102 117 11611 Gilgit 1460 235 243 171 17 45 4512 Gujar Khan 457 281 29 94 69 65 6713 Gupis 2156 187 189 158 164 36 3714 Kakul 1308 227 233 93 102 117 11615 Kohat 1440 289 303 171 17 45 4516 Kotli 610 284 284 166 149 1272 118317 Mangla 282 304 309 176 17 35 3918 Murree 2206 163 18 89 84 1765 173419 Muzaffarabad 702 154 159 176 17 35 3920 Naran 2363 141 105 87 96 42 4421 Palandri 1402 154 159 176 17 35 3922 Parachinar 1725 211 213 94 69 65 6723 Peshawar 320 293 297 158 164 36 3724 Rawalakot 1677 20 211 87 96 44 4625 Risalpur 575 295 299 146 142 55 5426 Saidu Sharif 961 256 263 123 119 90 9027 Skardu 2317 18 192 51 47 17 20
Advances in Meteorology 5
E(U) n1n2
2
V(U) n1n2 n1 + n2 + 1( 1113857
12
(6)
323 Relative Changes (e relative change () in theannual and seasonal temperature precipitation andstreamflow was assessed by using the following equation
relative change mean of 2nd period minus mean of 1st period
mean of 1st period
(7)
33 Trend Analysis For detection of trends we (i) pre-whitened time-series to eliminate effect of serial correlationof observations (ii) applied MannndashKendall trend analysisto identify if trends are significant and (iii) assessed thetrend slope line by means of Senrsquos estimator Analysis iscommon and reference is made to applications in[1 2 37ndash40]
331 MannndashKendall Test A nonparametric rank-basedMannndashKendall (MK) trend analysis test was used toevaluate the variations in the hydrometeorological time-series data over UIB [41 42] (e main advantage of theMK test is that there are no assumptions about the sta-tistical distribution of the sample data Since the method isrank-based extreme data points in the hydrometeoro-logical time-series will not largely affect the results(eMKtest statistic (S) is given by
S 1113944nminus1
k11113944
n
jk+1sig Xj minus Xk1113872 1113873
sgn Xj minus Xk1113872 1113873
if Xj minus Xk1113872 1113873lt 0 then minus 1
if Xj minus Xk1113872 1113873 0 then 0
if Xj minus Xk1113872 1113873gt 0 then 1
⎧⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎭
(8)
where ldquonrdquo denotes the length of a dataset and Xj and Xk are thesequential data values at times j and k Positive value of Sindicates an increasing (upward) trend and negative value of Sreveals a decreasing (downward) trend in the time-series data
Var(S) n(n minus 1)(2n + 5) minus 1113936
p
k1 tk tk minus 1( 1113857 2tk + 5( 1113857
18
(9)
where tk is the number of tied values in the qth group and thesign ldquo1113936rdquo represents the summation of all the tied groupsHowever if there are no tied groups in the data then thismay be ignored After calculating the variance Var(S) fromequation (4) the standardized test statistic (Zmk) value iscalculated by using the following equation
Zmk
if Slt 0 thenS + 1
VAR(S)
1113968
if xk minus xj1113872 1113873 0 then 0
if xk minus xj1113872 1113873gt 0 thenS minus 1
VAR(S)
1113968
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎭
(10)
A positive value of Zmk indicates an upward trend (ieincreasing) whereas a negative value indicates a downwardtrend (ie decreasing) (e test statistic (S) follows thestandard normal distribution where probability of ob-serving a value higher than the test statistic Zmk is testedunder the null hypothesis H0 that there is no trend forchosen α-level of significance H0 is rejected if the absolutevalue of ZmkgtZ1minusa2 at the α-level is significant
332 Senrsquos Estimator of Slope Senrsquos nonparametric method[43] was used to estimate the magnitude of trends in thetime-series data (e slope of ldquonrdquo pairs of data can be firstestimated by using the following equation
Qi Xj minus Xk
j minus k1113890 1113891 ifjgt k (11)
Senrsquos estimator is the median Qmed of the N pairs of QiIn the procedure N values of Qi are ranked from smallest tolargest and Senrsquos estimator is determined by using the fol-lowing equation
Senrsquos Estimator
Q[(N+1)2] if Nwas odd
12
QN2 + Q[(N+2)2]1113872 1113873 if Nwas even
⎧⎪⎪⎪⎨
⎪⎪⎪⎩
⎫⎪⎪⎪⎬
⎪⎪⎪⎭
(12)
Q med is tested by a two-sided test at the 100(1minus α)confidence interval and the true slope may be obtained bythe nonparametric test Data were processed using an Excelmacro named MAKESENS [44]
4 Results and Discussions
41 Variability in Temperature Studentrsquos t-test F-test andMannndashWhitney U test were used to detect the percentagechange between two consecutive (26-year each) hydrome-teorological time-series Table 3 showed the results ofmaximum and minimum temperatures (Tmax and Tmin) andprecipitation variables at seasonal and annual scale (eresults of testing are combined with assessment of relativechanges as indicated by percent change (e results showedinconsistent variations in all variables at different stationsand seasons (e annual Tmax during winter and springseason decreased (minus22 and minus27) between two subseries atBagh and Naran stations respectively whereas at Murreeand Peshawar stations it increased by 10 and 20 re-spectively It was noted that the changes in Tmax at seasonalscale were quite high in magnitude as compared to Tminduring the same seasons For Tmax most significant
6 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
Shigar basins respectively About 24 of the area of theShyok river basin is covered with snow [31] Similarly one-third of the area of Shigar basin is covered by glaciers in-cluding the worldrsquos largest glaciers and ice masses and al-most 25 to 90 of the area is covered with snow (e basicsource of precipitation of these two basins is westerly dis-turbances during winter and spring seasons followed by thesummer monsoon intruding during various intervals[3 19 32 33] (is study used the discharge data of Shyokriver at Yugo and Indus river at Shigar for the trend analysisof flows within these two basins
22 Astore and Hunza Subbasins of UIB (e Astore basin islocated in the western Himalayan and Hunza basin is in thewestern Karakoram ranges (e glacial coverage within thesebasins is less than snow coverage as compared to the Shyokand Shigar basins(e glaciers and permanent ice cover withinHunza basin is 28 and 14 which exhibits almost 21 and3 of the total UIB glacial coverage within Hunza and Astorebasins respectively [3 31] (ree high-altitude stations areinstalled within Hunza basin that is Khunjrab Naltar andZiarat while river discharge is measured at Danyior bridge atHunza river (ere is only one climatic station (Astore) in-stalled in Astore basin measuring temperature and precipi-tation data in this basin Discharge data of Astore river atDoyian are used in this study for the trend analysis installed byWater and Power Development Authority (WAPDA)
23 Gilgit Subbasin of UIB (e Gilgit subbasin ranges be-tween 358 and 37 E and 725 to 744 N comprehends easternpart of Hindukush range and drains towards southeast tojoin Indus river (e discharge of Gilgit river is measured atGilgit hydrometric station and at the confluence of Hunzaand Gilgit river which is called Alam bridge (e drainagearea of this basin incorporates 12000 km2 with an elevationrange from 1481 to 7134m masl Four climatic stations areinstalled in this area that is Gilgit Gupis Yasin andUshkore by the Pakistan Meteorological Department(PMD) and WAPDA In this study data at two stations(Gilgit and Gupis) were used from 1960 to 2014 Hassonet al [3] reported that the Gilgit basin receives maximumprecipitation at Ushkore (3151m) and minimum precipi-tation at Gilgit (1460m) due to the westerly disturbances andsummer monsoon
24 Jehlum Subbasin of UIB (eMangla basin is located onthe southern slope of the Himalayas with elevation rangingfrom 300m to 6282mmasl and has basin area of around33425 km2 at Mangla dam (is dam serves hydropowergeneration and regulates the flow from Mangla reservoirAbout 55 of the area lies in Indian held Kashmir and 45lies in Pakistan including Azad Kashmir (ere are fivesubcatchments that is Jhelum Poonch Kanshi NeelumKishanganga and Kunhar which drain water to Manglareservoir
71deg0prime0PrimeE 72deg0prime0PrimeE 73deg0prime0PrimeE 74deg0prime0PrimeE 75deg0prime0PrimeE 76deg0prime0PrimeE 77deg0prime0PrimeE
71deg0prime0PrimeE 72deg0prime0PrimeE 73deg0prime0PrimeE 74deg0prime0PrimeE 75deg0prime0PrimeE 76deg0prime0PrimeE 77deg0prime0PrimeE
37deg0prime0PrimeN
36deg0prime0PrimeN
35deg0prime0PrimeN
34deg0prime0PrimeN
33deg0prime0PrimeN
37deg0prime0PrimeN
36deg0prime0PrimeN
35deg0prime0PrimeN
34deg0prime0PrimeN
33deg0prime0PrimeN
Stream_gauges
Climatic stations
Drainage line
Catchment
198-1000
1000-2000
2000-3000
3000-4000
4000-5000
5000-6000
6000-7000
7000-8000
gt8000
Elevationamsl (m)
Figure 1(e upper Indus basin confined in Pakistan boundary showing rivers elevation streamflow gauges (Table 1) and climatic stations(Table 2)
Advances in Meteorology 3
25 Kabul Subbasin of UIB Kabul river in the easternAfghanistan and northwestern Pakistan is 700 km long ofwhich 560 km lies in Afghanistan It originates in theSanglakh ranges located 72 km west of Kabul city It flowseast through Kabul and Jalalabad north of the Khyber Passinto Pakistan (e river has four major tributaries theLowgar the Panjsher the Konar (Kunar) and the AlingarMost of area of this catchment lies in Afghanistan Due tounavailability of data from Afghanistan the study area wasconfined to the catchment falling within Pakistan boundary(e Kabul river a major western flank tributary joins withIndus near Attock
3 Methodology
31 Data and Methods In this study hydrological time-series data of 34 stream gauges and meteorological data of27 stations for the period of 1963 to 2014 were collectedfrom WAPDA and PMD (e information regarding thelocation of each stream gauge station area of subbasinsand mean annual streamflow is presented in Table 1
Moreover information of mean annual maximum andminimum temperature (Tmax Tmin) and precipitation (P) of27 climatic stations is shown in Table 2 Streamflowmeasurements in the UIB are carried out by WAPDA withthe earliest records commencing from 1960 (e streamgauges have a broad range of drainage area from 262 km2 to286000 km2 (e study area contained three major basinsnamely Jhelum Indus and Kabul (e locations of dif-ferent hydrological and meteorological stations are pre-sented in Figure 1
(e hydrometeorological time-series data of entire studyperiod (1963ndash2014) were divided into two equal subseriesthat is 1963 to 1988 and 1989 to 2014 to analyze the aspectsof acceleration of climate change Mean monthly seasonaland annual values of Tmax Tmin P and Qst were dereivedfrom the daily time-series data To analyze the seasonalvariations in the hydrometeorological time-series data fourseasons were defined as winter (December January andFebruary (DJF)) spring (March April and May (MAM))summer (June July and August (JJA)) and autumn (Sep-tember October and November (SON)) seasons
Table 1 List of stream gauges used in the present study and their characteristics (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Latitude (dd) Longitude (dd) Area (Km2)Mean annual
streamflow (m3s)1963ndash1988 1989ndash2014
1 Naran 349 737 1036 477 4562 Garhi Habibullah 344 734 2355 100 10553 Muzaffarabad 344 735 7275 3423 32194 Chinari 342 738 13598 2987 2895 Domel 344 735 14504 3273 32236 Kohala 341 735 24890 776 78057 Azad Pattan 337 736 26485 11507 124188 Kotli 335 739 3238 1239 12739 Palote 332 734 1111 6 5310 Kharmong 352 759 67858 4627 46511 Yogo 352 761 33670 3412 368812 Shigar 354 757 6610 1946 220513 Kachura 355 754 112665 962 1159614 Gilgit 359 743 12095 2772 333715 Dainyor Br 359 744 13157 3654 29516 Alam Br 358 746 26159 6618 619317 Bunji 357 746 142709 1706 1875318 Doyain 355 747 4040 1183 149219 Shatial Br 355 736 150220 19389 2110620 Karora 349 728 635 204 17521 Besham Qila 349 729 162393 23502 2436822 Daggar 345 725 598 54 5923 Phulra 343 731 1057 186 20524 Kalam 355 726 2020 857 86225 Chakdara 346 72 5776 1691 207126 Chitral 359 718 11396 2644 285427 Nowshera 34 72 88578 849 824228 Gurriala 337 723 3056 269 24829 Khairabad 339 722 252525 32227 2834430 (al 334 715 5543 277 22631 Chirah 337 733 326 57 432 Chahan 334 729 241 17 1333 Dhok Pathan 331 723 6475 44 38434 Massan 33 717 286000 35272 38095
4 Advances in Meteorology
32 Change Detection To detect climate impacts and pos-sible climate change acceleration over the past 52 yearsStudentrsquos t-test was selected for sample means whereas theldquoF-testrdquo was selected to assess aspects of variability on time-series (e nonparametric MannndashWhitney U test was usedto assess aspects of the distribution of observations
321 Studentrsquos t-Test and F-Test (e application of Stu-dentrsquos t-test aims to detect the change in the mean values oftwo 26-year periods for statistical significance F-test wasused to detect the variability in the time-series Both testswere performed at the 90 confidence interval (e t-sta-tistic is calculated by using equation (1) when the variancesof both time-series have similar values
t X1 minus X2
SP
1n1( 1113857 + 1n2( 1113857
1113969 (1)
x1 and x2 are mean values of two subseries n1 and n2indicate number of observations s1 and s2 are the standarddeviations subscripts 1 and 2 indicate the periods 1963ndash1988and 1989ndash2014 and Sp is the standard deviation which isgiven as
SP
n1 minus 1( 1113857S21 + n2 minus 1( 1113857S22
n1 + n2 minus 2
1113971
(2)
If the variances for the two periods are different then thet-statistic is used and given in the following equation
t X1 minus X2
S21n1) + S22n2(
1113969 (3)
322 MannndashWhitney U-Test (e nonparametric Man-nndashWhitney U test [34] is selected to detect the shift in thetemperature precipitation and streamflow time-series data(eMannndashWhitney (MW)U test statistic [35 36] is given inthe following equation
U min U1 U21113858 1113859 (4)
We have
U1 n1n2 +n1 n1 + 1( 1113857
2minus R1
U2 n1n2 +n2 n2 + 1( 1113857
2minus R2
(5)
where U1 and U2 are the total count of samples 1 and 2 andR1 and R2 are the rank sums of sample 1 and sample 2respectively When the null hypothesis H0 is true and n1and n2 are both larger than 8 U is considered approximatelynormally distributed with mean of E(U) and variance ofV(U) given as
Table 2 List of climatic stations in upper Indus basin (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Elevation (m)Max temp (degC) Min temp (degC) Precipitation (mm)
1963ndash1988 1989ndash2014 1963ndash1988 1989ndash2014 1963ndash1988 1989ndash20141 Astore 2168 154 158 4 41 39 422 Bagh 1067 254 199 4 46 13 133 Balakot 9955 154 159 145 142 50 504 Bunji 1372 24 237 14 143 15 175 Cherat 1372 219 211 89 83 33 386 Chilas 1250 266 262 83 78 130 1237 Chitral 14978 228 238 111 114 44 498 Dir 1375 225 233 79 74 11 119 Drosh 14639 238 243 71 61 7 1410 Garidopatta 8135 154 159 93 102 117 11611 Gilgit 1460 235 243 171 17 45 4512 Gujar Khan 457 281 29 94 69 65 6713 Gupis 2156 187 189 158 164 36 3714 Kakul 1308 227 233 93 102 117 11615 Kohat 1440 289 303 171 17 45 4516 Kotli 610 284 284 166 149 1272 118317 Mangla 282 304 309 176 17 35 3918 Murree 2206 163 18 89 84 1765 173419 Muzaffarabad 702 154 159 176 17 35 3920 Naran 2363 141 105 87 96 42 4421 Palandri 1402 154 159 176 17 35 3922 Parachinar 1725 211 213 94 69 65 6723 Peshawar 320 293 297 158 164 36 3724 Rawalakot 1677 20 211 87 96 44 4625 Risalpur 575 295 299 146 142 55 5426 Saidu Sharif 961 256 263 123 119 90 9027 Skardu 2317 18 192 51 47 17 20
Advances in Meteorology 5
E(U) n1n2
2
V(U) n1n2 n1 + n2 + 1( 1113857
12
(6)
323 Relative Changes (e relative change () in theannual and seasonal temperature precipitation andstreamflow was assessed by using the following equation
relative change mean of 2nd period minus mean of 1st period
mean of 1st period
(7)
33 Trend Analysis For detection of trends we (i) pre-whitened time-series to eliminate effect of serial correlationof observations (ii) applied MannndashKendall trend analysisto identify if trends are significant and (iii) assessed thetrend slope line by means of Senrsquos estimator Analysis iscommon and reference is made to applications in[1 2 37ndash40]
331 MannndashKendall Test A nonparametric rank-basedMannndashKendall (MK) trend analysis test was used toevaluate the variations in the hydrometeorological time-series data over UIB [41 42] (e main advantage of theMK test is that there are no assumptions about the sta-tistical distribution of the sample data Since the method isrank-based extreme data points in the hydrometeoro-logical time-series will not largely affect the results(eMKtest statistic (S) is given by
S 1113944nminus1
k11113944
n
jk+1sig Xj minus Xk1113872 1113873
sgn Xj minus Xk1113872 1113873
if Xj minus Xk1113872 1113873lt 0 then minus 1
if Xj minus Xk1113872 1113873 0 then 0
if Xj minus Xk1113872 1113873gt 0 then 1
⎧⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎭
(8)
where ldquonrdquo denotes the length of a dataset and Xj and Xk are thesequential data values at times j and k Positive value of Sindicates an increasing (upward) trend and negative value of Sreveals a decreasing (downward) trend in the time-series data
Var(S) n(n minus 1)(2n + 5) minus 1113936
p
k1 tk tk minus 1( 1113857 2tk + 5( 1113857
18
(9)
where tk is the number of tied values in the qth group and thesign ldquo1113936rdquo represents the summation of all the tied groupsHowever if there are no tied groups in the data then thismay be ignored After calculating the variance Var(S) fromequation (4) the standardized test statistic (Zmk) value iscalculated by using the following equation
Zmk
if Slt 0 thenS + 1
VAR(S)
1113968
if xk minus xj1113872 1113873 0 then 0
if xk minus xj1113872 1113873gt 0 thenS minus 1
VAR(S)
1113968
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎭
(10)
A positive value of Zmk indicates an upward trend (ieincreasing) whereas a negative value indicates a downwardtrend (ie decreasing) (e test statistic (S) follows thestandard normal distribution where probability of ob-serving a value higher than the test statistic Zmk is testedunder the null hypothesis H0 that there is no trend forchosen α-level of significance H0 is rejected if the absolutevalue of ZmkgtZ1minusa2 at the α-level is significant
332 Senrsquos Estimator of Slope Senrsquos nonparametric method[43] was used to estimate the magnitude of trends in thetime-series data (e slope of ldquonrdquo pairs of data can be firstestimated by using the following equation
Qi Xj minus Xk
j minus k1113890 1113891 ifjgt k (11)
Senrsquos estimator is the median Qmed of the N pairs of QiIn the procedure N values of Qi are ranked from smallest tolargest and Senrsquos estimator is determined by using the fol-lowing equation
Senrsquos Estimator
Q[(N+1)2] if Nwas odd
12
QN2 + Q[(N+2)2]1113872 1113873 if Nwas even
⎧⎪⎪⎪⎨
⎪⎪⎪⎩
⎫⎪⎪⎪⎬
⎪⎪⎪⎭
(12)
Q med is tested by a two-sided test at the 100(1minus α)confidence interval and the true slope may be obtained bythe nonparametric test Data were processed using an Excelmacro named MAKESENS [44]
4 Results and Discussions
41 Variability in Temperature Studentrsquos t-test F-test andMannndashWhitney U test were used to detect the percentagechange between two consecutive (26-year each) hydrome-teorological time-series Table 3 showed the results ofmaximum and minimum temperatures (Tmax and Tmin) andprecipitation variables at seasonal and annual scale (eresults of testing are combined with assessment of relativechanges as indicated by percent change (e results showedinconsistent variations in all variables at different stationsand seasons (e annual Tmax during winter and springseason decreased (minus22 and minus27) between two subseries atBagh and Naran stations respectively whereas at Murreeand Peshawar stations it increased by 10 and 20 re-spectively It was noted that the changes in Tmax at seasonalscale were quite high in magnitude as compared to Tminduring the same seasons For Tmax most significant
6 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
25 Kabul Subbasin of UIB Kabul river in the easternAfghanistan and northwestern Pakistan is 700 km long ofwhich 560 km lies in Afghanistan It originates in theSanglakh ranges located 72 km west of Kabul city It flowseast through Kabul and Jalalabad north of the Khyber Passinto Pakistan (e river has four major tributaries theLowgar the Panjsher the Konar (Kunar) and the AlingarMost of area of this catchment lies in Afghanistan Due tounavailability of data from Afghanistan the study area wasconfined to the catchment falling within Pakistan boundary(e Kabul river a major western flank tributary joins withIndus near Attock
3 Methodology
31 Data and Methods In this study hydrological time-series data of 34 stream gauges and meteorological data of27 stations for the period of 1963 to 2014 were collectedfrom WAPDA and PMD (e information regarding thelocation of each stream gauge station area of subbasinsand mean annual streamflow is presented in Table 1
Moreover information of mean annual maximum andminimum temperature (Tmax Tmin) and precipitation (P) of27 climatic stations is shown in Table 2 Streamflowmeasurements in the UIB are carried out by WAPDA withthe earliest records commencing from 1960 (e streamgauges have a broad range of drainage area from 262 km2 to286000 km2 (e study area contained three major basinsnamely Jhelum Indus and Kabul (e locations of dif-ferent hydrological and meteorological stations are pre-sented in Figure 1
(e hydrometeorological time-series data of entire studyperiod (1963ndash2014) were divided into two equal subseriesthat is 1963 to 1988 and 1989 to 2014 to analyze the aspectsof acceleration of climate change Mean monthly seasonaland annual values of Tmax Tmin P and Qst were dereivedfrom the daily time-series data To analyze the seasonalvariations in the hydrometeorological time-series data fourseasons were defined as winter (December January andFebruary (DJF)) spring (March April and May (MAM))summer (June July and August (JJA)) and autumn (Sep-tember October and November (SON)) seasons
Table 1 List of stream gauges used in the present study and their characteristics (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Latitude (dd) Longitude (dd) Area (Km2)Mean annual
streamflow (m3s)1963ndash1988 1989ndash2014
1 Naran 349 737 1036 477 4562 Garhi Habibullah 344 734 2355 100 10553 Muzaffarabad 344 735 7275 3423 32194 Chinari 342 738 13598 2987 2895 Domel 344 735 14504 3273 32236 Kohala 341 735 24890 776 78057 Azad Pattan 337 736 26485 11507 124188 Kotli 335 739 3238 1239 12739 Palote 332 734 1111 6 5310 Kharmong 352 759 67858 4627 46511 Yogo 352 761 33670 3412 368812 Shigar 354 757 6610 1946 220513 Kachura 355 754 112665 962 1159614 Gilgit 359 743 12095 2772 333715 Dainyor Br 359 744 13157 3654 29516 Alam Br 358 746 26159 6618 619317 Bunji 357 746 142709 1706 1875318 Doyain 355 747 4040 1183 149219 Shatial Br 355 736 150220 19389 2110620 Karora 349 728 635 204 17521 Besham Qila 349 729 162393 23502 2436822 Daggar 345 725 598 54 5923 Phulra 343 731 1057 186 20524 Kalam 355 726 2020 857 86225 Chakdara 346 72 5776 1691 207126 Chitral 359 718 11396 2644 285427 Nowshera 34 72 88578 849 824228 Gurriala 337 723 3056 269 24829 Khairabad 339 722 252525 32227 2834430 (al 334 715 5543 277 22631 Chirah 337 733 326 57 432 Chahan 334 729 241 17 1333 Dhok Pathan 331 723 6475 44 38434 Massan 33 717 286000 35272 38095
4 Advances in Meteorology
32 Change Detection To detect climate impacts and pos-sible climate change acceleration over the past 52 yearsStudentrsquos t-test was selected for sample means whereas theldquoF-testrdquo was selected to assess aspects of variability on time-series (e nonparametric MannndashWhitney U test was usedto assess aspects of the distribution of observations
321 Studentrsquos t-Test and F-Test (e application of Stu-dentrsquos t-test aims to detect the change in the mean values oftwo 26-year periods for statistical significance F-test wasused to detect the variability in the time-series Both testswere performed at the 90 confidence interval (e t-sta-tistic is calculated by using equation (1) when the variancesof both time-series have similar values
t X1 minus X2
SP
1n1( 1113857 + 1n2( 1113857
1113969 (1)
x1 and x2 are mean values of two subseries n1 and n2indicate number of observations s1 and s2 are the standarddeviations subscripts 1 and 2 indicate the periods 1963ndash1988and 1989ndash2014 and Sp is the standard deviation which isgiven as
SP
n1 minus 1( 1113857S21 + n2 minus 1( 1113857S22
n1 + n2 minus 2
1113971
(2)
If the variances for the two periods are different then thet-statistic is used and given in the following equation
t X1 minus X2
S21n1) + S22n2(
1113969 (3)
322 MannndashWhitney U-Test (e nonparametric Man-nndashWhitney U test [34] is selected to detect the shift in thetemperature precipitation and streamflow time-series data(eMannndashWhitney (MW)U test statistic [35 36] is given inthe following equation
U min U1 U21113858 1113859 (4)
We have
U1 n1n2 +n1 n1 + 1( 1113857
2minus R1
U2 n1n2 +n2 n2 + 1( 1113857
2minus R2
(5)
where U1 and U2 are the total count of samples 1 and 2 andR1 and R2 are the rank sums of sample 1 and sample 2respectively When the null hypothesis H0 is true and n1and n2 are both larger than 8 U is considered approximatelynormally distributed with mean of E(U) and variance ofV(U) given as
Table 2 List of climatic stations in upper Indus basin (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Elevation (m)Max temp (degC) Min temp (degC) Precipitation (mm)
1963ndash1988 1989ndash2014 1963ndash1988 1989ndash2014 1963ndash1988 1989ndash20141 Astore 2168 154 158 4 41 39 422 Bagh 1067 254 199 4 46 13 133 Balakot 9955 154 159 145 142 50 504 Bunji 1372 24 237 14 143 15 175 Cherat 1372 219 211 89 83 33 386 Chilas 1250 266 262 83 78 130 1237 Chitral 14978 228 238 111 114 44 498 Dir 1375 225 233 79 74 11 119 Drosh 14639 238 243 71 61 7 1410 Garidopatta 8135 154 159 93 102 117 11611 Gilgit 1460 235 243 171 17 45 4512 Gujar Khan 457 281 29 94 69 65 6713 Gupis 2156 187 189 158 164 36 3714 Kakul 1308 227 233 93 102 117 11615 Kohat 1440 289 303 171 17 45 4516 Kotli 610 284 284 166 149 1272 118317 Mangla 282 304 309 176 17 35 3918 Murree 2206 163 18 89 84 1765 173419 Muzaffarabad 702 154 159 176 17 35 3920 Naran 2363 141 105 87 96 42 4421 Palandri 1402 154 159 176 17 35 3922 Parachinar 1725 211 213 94 69 65 6723 Peshawar 320 293 297 158 164 36 3724 Rawalakot 1677 20 211 87 96 44 4625 Risalpur 575 295 299 146 142 55 5426 Saidu Sharif 961 256 263 123 119 90 9027 Skardu 2317 18 192 51 47 17 20
Advances in Meteorology 5
E(U) n1n2
2
V(U) n1n2 n1 + n2 + 1( 1113857
12
(6)
323 Relative Changes (e relative change () in theannual and seasonal temperature precipitation andstreamflow was assessed by using the following equation
relative change mean of 2nd period minus mean of 1st period
mean of 1st period
(7)
33 Trend Analysis For detection of trends we (i) pre-whitened time-series to eliminate effect of serial correlationof observations (ii) applied MannndashKendall trend analysisto identify if trends are significant and (iii) assessed thetrend slope line by means of Senrsquos estimator Analysis iscommon and reference is made to applications in[1 2 37ndash40]
331 MannndashKendall Test A nonparametric rank-basedMannndashKendall (MK) trend analysis test was used toevaluate the variations in the hydrometeorological time-series data over UIB [41 42] (e main advantage of theMK test is that there are no assumptions about the sta-tistical distribution of the sample data Since the method isrank-based extreme data points in the hydrometeoro-logical time-series will not largely affect the results(eMKtest statistic (S) is given by
S 1113944nminus1
k11113944
n
jk+1sig Xj minus Xk1113872 1113873
sgn Xj minus Xk1113872 1113873
if Xj minus Xk1113872 1113873lt 0 then minus 1
if Xj minus Xk1113872 1113873 0 then 0
if Xj minus Xk1113872 1113873gt 0 then 1
⎧⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎭
(8)
where ldquonrdquo denotes the length of a dataset and Xj and Xk are thesequential data values at times j and k Positive value of Sindicates an increasing (upward) trend and negative value of Sreveals a decreasing (downward) trend in the time-series data
Var(S) n(n minus 1)(2n + 5) minus 1113936
p
k1 tk tk minus 1( 1113857 2tk + 5( 1113857
18
(9)
where tk is the number of tied values in the qth group and thesign ldquo1113936rdquo represents the summation of all the tied groupsHowever if there are no tied groups in the data then thismay be ignored After calculating the variance Var(S) fromequation (4) the standardized test statistic (Zmk) value iscalculated by using the following equation
Zmk
if Slt 0 thenS + 1
VAR(S)
1113968
if xk minus xj1113872 1113873 0 then 0
if xk minus xj1113872 1113873gt 0 thenS minus 1
VAR(S)
1113968
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎭
(10)
A positive value of Zmk indicates an upward trend (ieincreasing) whereas a negative value indicates a downwardtrend (ie decreasing) (e test statistic (S) follows thestandard normal distribution where probability of ob-serving a value higher than the test statistic Zmk is testedunder the null hypothesis H0 that there is no trend forchosen α-level of significance H0 is rejected if the absolutevalue of ZmkgtZ1minusa2 at the α-level is significant
332 Senrsquos Estimator of Slope Senrsquos nonparametric method[43] was used to estimate the magnitude of trends in thetime-series data (e slope of ldquonrdquo pairs of data can be firstestimated by using the following equation
Qi Xj minus Xk
j minus k1113890 1113891 ifjgt k (11)
Senrsquos estimator is the median Qmed of the N pairs of QiIn the procedure N values of Qi are ranked from smallest tolargest and Senrsquos estimator is determined by using the fol-lowing equation
Senrsquos Estimator
Q[(N+1)2] if Nwas odd
12
QN2 + Q[(N+2)2]1113872 1113873 if Nwas even
⎧⎪⎪⎪⎨
⎪⎪⎪⎩
⎫⎪⎪⎪⎬
⎪⎪⎪⎭
(12)
Q med is tested by a two-sided test at the 100(1minus α)confidence interval and the true slope may be obtained bythe nonparametric test Data were processed using an Excelmacro named MAKESENS [44]
4 Results and Discussions
41 Variability in Temperature Studentrsquos t-test F-test andMannndashWhitney U test were used to detect the percentagechange between two consecutive (26-year each) hydrome-teorological time-series Table 3 showed the results ofmaximum and minimum temperatures (Tmax and Tmin) andprecipitation variables at seasonal and annual scale (eresults of testing are combined with assessment of relativechanges as indicated by percent change (e results showedinconsistent variations in all variables at different stationsand seasons (e annual Tmax during winter and springseason decreased (minus22 and minus27) between two subseries atBagh and Naran stations respectively whereas at Murreeand Peshawar stations it increased by 10 and 20 re-spectively It was noted that the changes in Tmax at seasonalscale were quite high in magnitude as compared to Tminduring the same seasons For Tmax most significant
6 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
32 Change Detection To detect climate impacts and pos-sible climate change acceleration over the past 52 yearsStudentrsquos t-test was selected for sample means whereas theldquoF-testrdquo was selected to assess aspects of variability on time-series (e nonparametric MannndashWhitney U test was usedto assess aspects of the distribution of observations
321 Studentrsquos t-Test and F-Test (e application of Stu-dentrsquos t-test aims to detect the change in the mean values oftwo 26-year periods for statistical significance F-test wasused to detect the variability in the time-series Both testswere performed at the 90 confidence interval (e t-sta-tistic is calculated by using equation (1) when the variancesof both time-series have similar values
t X1 minus X2
SP
1n1( 1113857 + 1n2( 1113857
1113969 (1)
x1 and x2 are mean values of two subseries n1 and n2indicate number of observations s1 and s2 are the standarddeviations subscripts 1 and 2 indicate the periods 1963ndash1988and 1989ndash2014 and Sp is the standard deviation which isgiven as
SP
n1 minus 1( 1113857S21 + n2 minus 1( 1113857S22
n1 + n2 minus 2
1113971
(2)
If the variances for the two periods are different then thet-statistic is used and given in the following equation
t X1 minus X2
S21n1) + S22n2(
1113969 (3)
322 MannndashWhitney U-Test (e nonparametric Man-nndashWhitney U test [34] is selected to detect the shift in thetemperature precipitation and streamflow time-series data(eMannndashWhitney (MW)U test statistic [35 36] is given inthe following equation
U min U1 U21113858 1113859 (4)
We have
U1 n1n2 +n1 n1 + 1( 1113857
2minus R1
U2 n1n2 +n2 n2 + 1( 1113857
2minus R2
(5)
where U1 and U2 are the total count of samples 1 and 2 andR1 and R2 are the rank sums of sample 1 and sample 2respectively When the null hypothesis H0 is true and n1and n2 are both larger than 8 U is considered approximatelynormally distributed with mean of E(U) and variance ofV(U) given as
Table 2 List of climatic stations in upper Indus basin (period 1 1963ndash1988 period 2 1989ndash2014)
Sr no Station Elevation (m)Max temp (degC) Min temp (degC) Precipitation (mm)
1963ndash1988 1989ndash2014 1963ndash1988 1989ndash2014 1963ndash1988 1989ndash20141 Astore 2168 154 158 4 41 39 422 Bagh 1067 254 199 4 46 13 133 Balakot 9955 154 159 145 142 50 504 Bunji 1372 24 237 14 143 15 175 Cherat 1372 219 211 89 83 33 386 Chilas 1250 266 262 83 78 130 1237 Chitral 14978 228 238 111 114 44 498 Dir 1375 225 233 79 74 11 119 Drosh 14639 238 243 71 61 7 1410 Garidopatta 8135 154 159 93 102 117 11611 Gilgit 1460 235 243 171 17 45 4512 Gujar Khan 457 281 29 94 69 65 6713 Gupis 2156 187 189 158 164 36 3714 Kakul 1308 227 233 93 102 117 11615 Kohat 1440 289 303 171 17 45 4516 Kotli 610 284 284 166 149 1272 118317 Mangla 282 304 309 176 17 35 3918 Murree 2206 163 18 89 84 1765 173419 Muzaffarabad 702 154 159 176 17 35 3920 Naran 2363 141 105 87 96 42 4421 Palandri 1402 154 159 176 17 35 3922 Parachinar 1725 211 213 94 69 65 6723 Peshawar 320 293 297 158 164 36 3724 Rawalakot 1677 20 211 87 96 44 4625 Risalpur 575 295 299 146 142 55 5426 Saidu Sharif 961 256 263 123 119 90 9027 Skardu 2317 18 192 51 47 17 20
Advances in Meteorology 5
E(U) n1n2
2
V(U) n1n2 n1 + n2 + 1( 1113857
12
(6)
323 Relative Changes (e relative change () in theannual and seasonal temperature precipitation andstreamflow was assessed by using the following equation
relative change mean of 2nd period minus mean of 1st period
mean of 1st period
(7)
33 Trend Analysis For detection of trends we (i) pre-whitened time-series to eliminate effect of serial correlationof observations (ii) applied MannndashKendall trend analysisto identify if trends are significant and (iii) assessed thetrend slope line by means of Senrsquos estimator Analysis iscommon and reference is made to applications in[1 2 37ndash40]
331 MannndashKendall Test A nonparametric rank-basedMannndashKendall (MK) trend analysis test was used toevaluate the variations in the hydrometeorological time-series data over UIB [41 42] (e main advantage of theMK test is that there are no assumptions about the sta-tistical distribution of the sample data Since the method isrank-based extreme data points in the hydrometeoro-logical time-series will not largely affect the results(eMKtest statistic (S) is given by
S 1113944nminus1
k11113944
n
jk+1sig Xj minus Xk1113872 1113873
sgn Xj minus Xk1113872 1113873
if Xj minus Xk1113872 1113873lt 0 then minus 1
if Xj minus Xk1113872 1113873 0 then 0
if Xj minus Xk1113872 1113873gt 0 then 1
⎧⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎭
(8)
where ldquonrdquo denotes the length of a dataset and Xj and Xk are thesequential data values at times j and k Positive value of Sindicates an increasing (upward) trend and negative value of Sreveals a decreasing (downward) trend in the time-series data
Var(S) n(n minus 1)(2n + 5) minus 1113936
p
k1 tk tk minus 1( 1113857 2tk + 5( 1113857
18
(9)
where tk is the number of tied values in the qth group and thesign ldquo1113936rdquo represents the summation of all the tied groupsHowever if there are no tied groups in the data then thismay be ignored After calculating the variance Var(S) fromequation (4) the standardized test statistic (Zmk) value iscalculated by using the following equation
Zmk
if Slt 0 thenS + 1
VAR(S)
1113968
if xk minus xj1113872 1113873 0 then 0
if xk minus xj1113872 1113873gt 0 thenS minus 1
VAR(S)
1113968
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎭
(10)
A positive value of Zmk indicates an upward trend (ieincreasing) whereas a negative value indicates a downwardtrend (ie decreasing) (e test statistic (S) follows thestandard normal distribution where probability of ob-serving a value higher than the test statistic Zmk is testedunder the null hypothesis H0 that there is no trend forchosen α-level of significance H0 is rejected if the absolutevalue of ZmkgtZ1minusa2 at the α-level is significant
332 Senrsquos Estimator of Slope Senrsquos nonparametric method[43] was used to estimate the magnitude of trends in thetime-series data (e slope of ldquonrdquo pairs of data can be firstestimated by using the following equation
Qi Xj minus Xk
j minus k1113890 1113891 ifjgt k (11)
Senrsquos estimator is the median Qmed of the N pairs of QiIn the procedure N values of Qi are ranked from smallest tolargest and Senrsquos estimator is determined by using the fol-lowing equation
Senrsquos Estimator
Q[(N+1)2] if Nwas odd
12
QN2 + Q[(N+2)2]1113872 1113873 if Nwas even
⎧⎪⎪⎪⎨
⎪⎪⎪⎩
⎫⎪⎪⎪⎬
⎪⎪⎪⎭
(12)
Q med is tested by a two-sided test at the 100(1minus α)confidence interval and the true slope may be obtained bythe nonparametric test Data were processed using an Excelmacro named MAKESENS [44]
4 Results and Discussions
41 Variability in Temperature Studentrsquos t-test F-test andMannndashWhitney U test were used to detect the percentagechange between two consecutive (26-year each) hydrome-teorological time-series Table 3 showed the results ofmaximum and minimum temperatures (Tmax and Tmin) andprecipitation variables at seasonal and annual scale (eresults of testing are combined with assessment of relativechanges as indicated by percent change (e results showedinconsistent variations in all variables at different stationsand seasons (e annual Tmax during winter and springseason decreased (minus22 and minus27) between two subseries atBagh and Naran stations respectively whereas at Murreeand Peshawar stations it increased by 10 and 20 re-spectively It was noted that the changes in Tmax at seasonalscale were quite high in magnitude as compared to Tminduring the same seasons For Tmax most significant
6 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
E(U) n1n2
2
V(U) n1n2 n1 + n2 + 1( 1113857
12
(6)
323 Relative Changes (e relative change () in theannual and seasonal temperature precipitation andstreamflow was assessed by using the following equation
relative change mean of 2nd period minus mean of 1st period
mean of 1st period
(7)
33 Trend Analysis For detection of trends we (i) pre-whitened time-series to eliminate effect of serial correlationof observations (ii) applied MannndashKendall trend analysisto identify if trends are significant and (iii) assessed thetrend slope line by means of Senrsquos estimator Analysis iscommon and reference is made to applications in[1 2 37ndash40]
331 MannndashKendall Test A nonparametric rank-basedMannndashKendall (MK) trend analysis test was used toevaluate the variations in the hydrometeorological time-series data over UIB [41 42] (e main advantage of theMK test is that there are no assumptions about the sta-tistical distribution of the sample data Since the method isrank-based extreme data points in the hydrometeoro-logical time-series will not largely affect the results(eMKtest statistic (S) is given by
S 1113944nminus1
k11113944
n
jk+1sig Xj minus Xk1113872 1113873
sgn Xj minus Xk1113872 1113873
if Xj minus Xk1113872 1113873lt 0 then minus 1
if Xj minus Xk1113872 1113873 0 then 0
if Xj minus Xk1113872 1113873gt 0 then 1
⎧⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎭
(8)
where ldquonrdquo denotes the length of a dataset and Xj and Xk are thesequential data values at times j and k Positive value of Sindicates an increasing (upward) trend and negative value of Sreveals a decreasing (downward) trend in the time-series data
Var(S) n(n minus 1)(2n + 5) minus 1113936
p
k1 tk tk minus 1( 1113857 2tk + 5( 1113857
18
(9)
where tk is the number of tied values in the qth group and thesign ldquo1113936rdquo represents the summation of all the tied groupsHowever if there are no tied groups in the data then thismay be ignored After calculating the variance Var(S) fromequation (4) the standardized test statistic (Zmk) value iscalculated by using the following equation
Zmk
if Slt 0 thenS + 1
VAR(S)
1113968
if xk minus xj1113872 1113873 0 then 0
if xk minus xj1113872 1113873gt 0 thenS minus 1
VAR(S)
1113968
⎧⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩
⎫⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎬
⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎭
(10)
A positive value of Zmk indicates an upward trend (ieincreasing) whereas a negative value indicates a downwardtrend (ie decreasing) (e test statistic (S) follows thestandard normal distribution where probability of ob-serving a value higher than the test statistic Zmk is testedunder the null hypothesis H0 that there is no trend forchosen α-level of significance H0 is rejected if the absolutevalue of ZmkgtZ1minusa2 at the α-level is significant
332 Senrsquos Estimator of Slope Senrsquos nonparametric method[43] was used to estimate the magnitude of trends in thetime-series data (e slope of ldquonrdquo pairs of data can be firstestimated by using the following equation
Qi Xj minus Xk
j minus k1113890 1113891 ifjgt k (11)
Senrsquos estimator is the median Qmed of the N pairs of QiIn the procedure N values of Qi are ranked from smallest tolargest and Senrsquos estimator is determined by using the fol-lowing equation
Senrsquos Estimator
Q[(N+1)2] if Nwas odd
12
QN2 + Q[(N+2)2]1113872 1113873 if Nwas even
⎧⎪⎪⎪⎨
⎪⎪⎪⎩
⎫⎪⎪⎪⎬
⎪⎪⎪⎭
(12)
Q med is tested by a two-sided test at the 100(1minus α)confidence interval and the true slope may be obtained bythe nonparametric test Data were processed using an Excelmacro named MAKESENS [44]
4 Results and Discussions
41 Variability in Temperature Studentrsquos t-test F-test andMannndashWhitney U test were used to detect the percentagechange between two consecutive (26-year each) hydrome-teorological time-series Table 3 showed the results ofmaximum and minimum temperatures (Tmax and Tmin) andprecipitation variables at seasonal and annual scale (eresults of testing are combined with assessment of relativechanges as indicated by percent change (e results showedinconsistent variations in all variables at different stationsand seasons (e annual Tmax during winter and springseason decreased (minus22 and minus27) between two subseries atBagh and Naran stations respectively whereas at Murreeand Peshawar stations it increased by 10 and 20 re-spectively It was noted that the changes in Tmax at seasonalscale were quite high in magnitude as compared to Tminduring the same seasons For Tmax most significant
6 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
Tabl
e3
Relativ
echange
()in
annu
alandseason
altemperature
andprecipita
tionin
2ndperiod
(1989ndash
2014)with
respectto
1stperiod
(1963ndash1988)(bold
underlinea
ndlowastshow
edsig
nificanttrend
with
Stud
entrsquos
t-testF
-testandMannndash
Whitney
Utestrespectively
at95
confi
dencelevel)
Srn
oClim
atic
stations
Maxim
umtemperature
Minim
umtemperature
Precipita
tion
Ann
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
nAnn
ual
Winter
Spring
Summer
Autum
n1
Astore
3lowast14lowast
4minus2
5lowast1
minus6
15lowast
minus6lowast
14
23minus12
36lowast
72
Bagh
minus22lowast
minus25
minus24lowast
minus15lowast
minus24lowast
minus8lowast
7minus13lowast
minus6lowast
minus8
230lowast
8minus9
minus3
3Ba
lako
tminus1lowast
00lowast
minus1lowast
minus2lowast
minus1
minus15
6lowast7
minus16
minus6
minus1
minus15lowast
minus6
84
Bunji
minus1
6lowast1
minus5lowast
minus1
128lowast
minus12
8lowast68lowast
46lowast
95lowast
3166
minus17
11lowast
235
Cherat
minus4lowast
minus7lowast
0minus3lowast
minus6lowast
minus3
0minus1
minus6lowast
0minus13
minus3
minus24
minus10
minus13
6Chilas
minus1
00
minus2lowast
minus1
255lowast
2minus1lowast
137lowast
46lowast
591lowast
907
Chitral
4lowast16lowast
7lowast0
4lowastminus6lowast
minus36lowast
minus4lowast
minus7lowast
minus7lowast
14lowast
24minus5
7466lowast
8Dir
3lowast3lowast
5lowast1lowast
2lowastminus8lowast
minus8
minus8lowast
minus6lowast
minus7lowast
minus3
minus3
minus5
minus36
9Drosh
2lowast7lowast
40
1minus47lowast
minus52lowast
minus48lowast
minus33lowast
minus48lowast
minus2
14minus12
911
10Garidop
atta
4lowast10lowast
5lowast2lowast
4lowast1
minus2
0minus3
11minus1
3lowast13
minus23lowast
minus14
minus25lowast
11Gilgit
3lowast12lowast
5lowastminus1
4lowastminus6lowast
minus20lowast
minus2
minus8lowast
minus10lowast
1635
minus2
3042
12Gujar
Khan
3lowast8lowast
21
3lowastminus6lowast
minus16lowast
minus8lowast
minus5lowast
minus2
minus3
3minus11
minus4
613
Gup
is1
144
minus2lowast
0minus14lowast
16minus3
minus12lowast
minus12lowast
19lowast
19lowast
16lowast
15lowast
1614
Kakul
3lowast6lowast
31
2lowastminus11lowast
minus36lowast
minus10lowast
minus6lowast
minus15lowast
413
11
815
Koh
at5lowast
6lowast7lowast
4lowast3lowast
minus1
minus2
minus1
0minus1
40lowast
75lowast
minus2
73lowast
1816
Kotli
01
2minus1
minus2
minus10lowast
minus3
minus21lowast
minus13lowast
minus2lowast
minus7
4minus8
minus7
minus16lowast
17Mangla
2lowast3lowast
3lowast1
minus36
minus3lowast
minus10lowast
minus3lowast
minus1
60lowast
minus5
minus8
minus22
14513
18Murree
10lowast
14lowast
49lowast
4lowast9lowast
minus5
13lowast
64minus1
5minus2
6minus9
0minus5
19Muzaff
arabad
3lowast5lowast
41
2lowast0
7lowast2
minus1
minus1
7lowast25lowast
36
220
Naran
minus27lowast
minus65lowast
minus58lowast
minus7lowast
minus4
9lowast78lowast
10lowast
110lowast
80lowast
80lowast
90lowast
72lowast
55lowast
21Pa
land
ri3lowast
15lowast
4minus1
5lowastminus8lowast
7minus13lowast
minus6lowast
minus8
minus17lowast
1minus20
minus22lowast
minus10
22Pa
rachinar
12
40
0minus27lowast
38lowast
minus23lowast
minus14lowast
minus25lowast
minus3
8minus8
minus5
223
Peshaw
ar20lowast
48lowast
21
60lowast
4lowast20lowast
5lowast0
3lowast20lowast
48lowast
21
50lowast
24Ra
walakot
6lowast20lowast
4lowast2lowast
6lowast9lowast
78lowast
10lowast
110lowast
minus7
7minus10
minus11
minus16lowast
25Risalpur
1lowast4lowast
3lowast0
minus1lowast
minus1
22lowast
minus3lowast
minus2lowast
minus2
minus11
minus9minus13
minus18
22lowast
26SaiduSh
arif
3lowast9lowast
5lowast1lowast
0lowastminus3lowast
30
minus3lowast
minus6lowast
1927lowast
726
2527
Skardu
7lowast37lowast
8lowast1
7lowastminus8lowast
minus11
minus2
minus7lowast
minus19lowast
30lowast
60lowast
1230
20
Advances in Meteorology 7
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
all trends significant trends
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
1st Period (1961ndash1986) 2nd Period (1987ndash2012)
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer AutumnAnnual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Annual Winter Spring Summer Autumn
Max
tem
pera
ture
Min
tem
pera
ture
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Prec
ipita
tion
Stre
amflo
w
Max
tem
pera
ture
Min
tem
pera
ture
Prec
ipita
tion
Stre
amflo
w 25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
25
50
75
100
ndash100
ndash75
ndash50
ndash25
0
Figure 2 Percent number of stations with positive (upward) and negative (downward) trends in annual and seasonal time-series fordifferent periods and number of stations with significant trends by MannndashKendall test at α 005
8 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
decreases (minus165 and minus58) were revealed at Naran stationfor winter and spring seasons respectively Bagh stationexhibited a relative change of minus25 minus24 minus15 and minus24during winter spring summer and autumn seasons re-spectively At Murree station 14 and 49 relative changewas observed in winter and spring seasons while in summerand autumn percent change values were quite smallerPeshawar station showed largest increases of 48 and 110in Tmax during winter and autumn seasons respectivelywhile percent change values for spring and summer were
negligible Minimum temperature exhibited both positiveand negative trends at different stations in different seasonsFor instance Bunji Peshawar and Rawalakot stationsshowed positive changes whereas at most of the stationsthese variations are negative Largest relative positive changeby the t-test F-test and U test of 128 for the Bunji stationwas observed in Tmin at 90 confidence level (e highestnegative percent change values for the second period (minus47and 27) at Drosh and Parachinar stations were detectedrespectively In all four seasons varied trends were observed
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 3 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 9
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
in Tmin by using Studentrsquos t-test Results of the F-test and Utest also suggested that the climate for the 2nd period wasquite different from the 1st period with most changes that arestatistically significant at 90 confidence level Significantchanges in variance are indicated at most of the stations overUIB in Tmin(Figure 2)
(e summary of the trend analyses and the spatialvariation in annual winter spring (premonsoon) summer(monsoon) and autumn (postmonsoon) maximum andminimum temperature are presented in Figures 3 and 4Most of the stations exhibit increasing trends in annualmaximum temperature for the 1st period (e increasingtrends were found at 56 stations out of which only 4were significant Similarly decreasing trends were found at44 stations (19 significant) Cherat Gujar Khan andKakul have the highest decreasing rate (12 08 and 06degC perdecade at 999 99 and 95 significant level respec-tively) (e stations show warming trends at a magnitude of01 to 05degCmiddotdecademinus1 In the 2nd period increasing trendswere observed at 85 stations out of which 44 stationsexhibit significant trends (e decreasing trends were foundat Bagh Gujar Khan Naran and Parachinar at the rates of73 03 03 and 07degC per decade respectively but only Baghand Parachinar exhibited significant trends at 999 and95 level of confidence interval respectively At annualscale almost all of the stations exhibited warming trendsHowever these findings needed further validation to analyzethe clear scenarios of climate change acceleration in thestudy area (erefore for more detailed trends of climatechange three-month seasonal analysis was carried out andsimilar warming patterns for winter spring and autumnwere found during the 1st period (1963ndash1988) MK testdetected significant trends at 25 24 and 22 stations at 90ndash999 significant level in winter spring and autumn tem-perature time-series as shown in Figures 3 and 4 Incon-sistent trends were detected during the summer season (emaximum temperature has decreased at 70 stations out ofwhich 33 stations exhibit significant trends in summerseason (e spring season showed the highest rate ofwarming as compared to other seasons More increasingtrends were observed during the 1st period as compared to
the 2nd period Increasing trends were also revealed at 8189 67 and 84 (15 70 19 and 11 significant)during winter spring summer and autumn seasons re-spectively Negative trends were observed at 19 11 33and 26 (11 4 15 and 15 significant) Strong in-dications of climate change acceleration during secondperiod were observed as compared to the first period In the1st period trend analysis proposed the existence of de-creasing trends in annual minimum temperature at 22stations Most of the stations exhibited decreasing trends inthe annual minimum temperature at 59 stations (41significant) Only three stations showed significant in-creasing trends Bunji Chilas and Peshawar Bunji stationshowed highest warming trend of 15degC per decade In the2nd period trend analysis with the MK test displayed trendexistence at 25 stations More increasing trends were foundas compared to the first period(ese increasing trends werefound at 56 (19 significant) and the decreasing trendswere found at 44 (7 significant) Trend analysis in sea-sonal minimum temperature during the 1st period revealedthat winter and spring seasons exhibited more increasingtrends at 63 and 67 stations (30 and 22 significant)respectively Meanwhile in summer and autumn there was apattern of decreasing trends at 74 and 67 stations (41and 41 significant) respectively As compared to the 1stperiod winter and summer seasons showed decreasingminimum temperature while spring and autumn seasonsindicated increasing minimum temperature in the 2nd pe-riod (1989ndash2014) Figures 5 and 6 revealed that at 67stations (26 significant) and 52 stations (11) warmingtrends were found for the winter and autumn seasons Forsummer season 74 stations (67 significant) exhibitedcooling trends (e cooling rates varied from 01degCmiddotdecademinus1
to 19degCmiddotdecademinus1
42 Variability in Precipitation Significant differences wereobserved at Naran and Gupis stations for all seasons but adifferent pattern was revealed during winter and summer Itwas observed that percent change values that are statisticallysignificant are relatively large at few stations with values in
(a) (b)
Figure 4 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual maximum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
10 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
the range of +25 to minus25 Highest increase in percentchanges of precipitation was detected at Gupis and Naranstations during all seasons but these changes became neg-ative and quite lower at low-altitude stations
(e results of analysis by applying MannndashKendall testand Senrsquos slope estimator methods in the annual precipi-tation time-series were summarized for two consecutive 26-year periods that is 1963ndash1988 and 1989ndash2014 (e annualprecipitation increased significantly at five stations while it
decreased at four stations during the first period It wasnoted that the Gupis station exhibited significant increasingprecipitation at the rate of 32 per year with 99 level ofconfidence In the 2nd period at two stations the annualprecipitation has increased significantly but decreased at tenstations (Table 3)(e highest increasing trend was observedwith a magnitude of 47 per year with 999 level of sig-nificance at Kohat station while the highest decreasing trendwas revealed with a magnitude of 26 at 95 confidence
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 5 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal minimum temperatureshowing change in degC decademinus1 (upward and downward arrows show positive and negative trends respectively bold (blue) arrow showssignificant trend at α 005)
Advances in Meteorology 11
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
interval at Risalpur station as shown in Figures 7 and 8 Inthe 1st period the MK nonparametric test showed negativetrends (Figures 7 and 8) in precipitation time-series duringwinter and autumn seasons at 59 staions (15 significant)and 59 stations (11 significant) and positive trends at74 stations (11 significant) and 74 stations (30 sig-nificant) during spring and summer seasons respectively(e most significant winter drying patterns were revealed atGupis Chitral Garidopatta and Naran stations at rates of47 13 25 and 18 respectively during the secondperiod Spring and summer seasons showed decreasingtrends at 93 stations (48 significant) and 78 stations(22 significant) respectively during the 2nd period It wasobserved that 63 stations (11 significant) exhibited in-creasing trends In autumn seasons 63 insignificant sta-tions showed decreasing trends as shown in Figures 7 and 8
43 Variability in Streamflow (e annual runoff in KurramSoan and Indus subbasins decreased by 18 13 and 12respectively however the runoff variations are found to bestatistically significant in Indus subbasin (e winter seasonshowed the largest variations compared to other seasonsMoreover all subbasins showed positive variations duringwinter season except for Kurram river subbasin as shown inTable 4 (e summer flows have been decreased in all riversCombined change detection results for Qst at annual baseonly show small relative changes (minus29 to 11) at most ofthe stations (ese results suggest that time-series have notnotably changed over time In Swat river a relative change of+22 was observed which was found to be significant usingStudentrsquos t-test F-test and U test At seasonal scale most ofthe changes are positive in winter season but negative insummer In winter season changes are largest (up to 69)and statistically significant whereas in summer seasonchanges are negative and statistically insignificant Chakdarastation of Kabul river basin showed most substantial relativechange of 69 in winter season which was significant for allthree tests In summary analyses on change detection ingeneral indicate acceleration of climate change
(e results of annual mean streamflow at 34 stationsusing MK test of two consecutive 26-year periods are pre-sented in Figures 9 and 10 During the 1st period increasingtrends were observed at 56 stations (11 significant) anddecreasing trends at 44 stations (11 significant) How-ever only seven stations revealed significant decreasingtrends (e highest decreasing trends were revealed at JhansiPost at a rate of 43 during the 1st period that is 1963ndash1988(e decreasing trends in annual mean streamflows werefound at 77 stations (43 significant) and the increasingtrends at 23 stations which are statistically insignificantAll nine tributaries of Jhelum basin (Naran Garhi Hab-ibullah Muzaffarabad Chinari Domail Kohala Azad Pa-than Kotli and Palote) exhibited significant decreasingtrends (e five tributaries of Indus basin (Karora GurrialaKhairabad Chahan and Massan) have also shown signifi-cant decreasing trends At Kabul basin only Kalam showedsignificant decreasing trends (e highest decreasing sig-nificant trends were found at Palote station at magnitude of43 during the whole study period Winter mean flows havesignificantly increased at four stations and decreased atseven stations (e highest significant increasing trend wasobserved at Massan station of Indus river while decreasingtrend was revealed at Jhansi Post station of Kabul river atrates of 15 and 38 for the first and second periods re-spectively All three major rivers exhibited increasing trendsat Azad Pattan in Jhelum Besham in Indus and Nowsherain Kabul however significant trends were detected in theIndus river During spring season significant streamflowtrends were detected over 10 stations (5 increasing and fivedecreasing) (e Brandu river at Daggar showed significantincreasing trend at a rate of 18 whereas the Bara riverexhibited decreasing trend at Jhansi Post station of 39Most of decreasing trends were observed in summer andautumn seasons as shown in Figures 9 and 10 In summerand autumn seasons 57 stations (9 significant) and 60stations (31 significant) exhibited decreasing trends re-spectively In the second period for winter season meanflows have increased at rate of 54 (14 significant) anddecreased at rate of 46 (11 significant) of the data period
(a) (b)
Figure 6 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual minimum temperatureshowing change in degCmiddotdecademinus1 (upward and downward arrows show positive and negative trends respectively blue arrow shows significanttrend at α 005 and green arrow shows insignificant trend)
12 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
average for the period of 1989ndash2014 (e highest significantincreasing trend was found at Chakdara station of Swat riverand decreasing trends were found at Jhansi Post station ofKabul river of 32 and 48 during the 2nd period(1989ndash2014) respectively All three major rivers have shown
increasing trends at Azad Pattan in Jhelum Besham inIndus and Nowshera in Kabul however only the Indusriver exhibited significant trends During spring seasonsixteen stations exhibited significant trends (4 increasingand 12 decreasing) (e Gilgit river at Gilgit and Alam Br
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 7 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 13
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
(a) (b)
Figure 8 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual precipitation showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Table 4 Relative change () in annual and seasonal streamflow during the 2nd period (1989ndash2014) with respect to the 1st period (1963ndash1988)(bold underline and lowast showed significant trend with Studentrsquos t-test F-test and MannndashWhitney U test respectively at 95 confidencelevel)
Stream gauge Annual Winter Spring Summer AutumnNaran minus4 minus6 minus3 minus10 24Garhi Habibullah 5 17 19lowast minus5 21Muzaffarabad minus6 13 3 minus16lowast 6Chinari minus3 6 0 minus7 minus5Domel minus2 13 2 minus8 0Kohala 1 21lowast 6 minus8 8Azad Pattan 8 37lowast 12 0 13Kotli 3 36lowast 6 minus10 10Palote minus12 27 minus27 minus14 minus17Kharmong 1 7lowast 7 minus7 1Yogo 8lowast 4 19 6lowast 19lowastShigar 13lowast 2lowast 5lowast 12lowast 3Kachura 21lowast 18lowast 24lowast 19lowast 26lowastGilgit 20 14lowast 43lowast 16 26lowastDainyor Br minus19lowast 8 5 minus25lowast minus6Alam Br minus6 12lowast 21lowast minus13lowast 3Bunji 10 12lowast 35lowast 5lowast 15lowastDoyain 26lowast 34lowast 28lowast 18lowast 39lowastShatial Br 9lowast 11lowast 19lowast 7 7lowastKarora minus14 19lowast minus20lowast minus28lowast 14Besham Qila 4 18lowast 14lowast minus1 13lowastDaggar 9 39lowast 21 minus6 2Phulra 10 38lowast 13 0 5Kalam 1 9lowast 20lowast minus5 1Chakdara 22lowast 69lowast 36lowast 8 37lowastChitral 8lowast 5lowast 15lowast 6lowast 12lowastJhansi post minus23 minus21lowast minus35lowast minus1 minus30lowastNowshera minus3 8 5 minus9 1Gurriala minus8 24 5 minus18lowast minus11Khairabad minus12lowast minus17 minus18lowast minus15 minus19(al minus18lowast minus24lowast minus31lowast minus1 minus17lowastChirah minus29lowast minus11 minus29lowast minus35lowast minus16lowastChahan minus21 0 minus18 minus30 6lowastDhok Pathan minus13 18 minus3 minus25lowast 15Massan 8lowast 28lowast 7lowast 2 18lowast
14 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
exhibited significant increasing trends at rates of 22 and10 respectively (e Indus river at Shatial Br has shownincreasing trends of 14 whereas lower parts of Indus basinand Jhelum basin exhibited decreasing significant trendsMost of the decreasing trends were observed during summerand autumn flows as shown in Figures 9 and 10 Summer
and autumn seasons exhibited increasing trends at 74 and66 stations out of which 40 and 37 stations showedsignificant decreasing trends respectively All three basinsexhibited significant decreasing trends at Azad PattanBesham and Nowshera All the subbasins of Jhelum riverexhibited decreasing trends Kunhar Neelum and Kanshi
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(a)
Aut
umn
(SO
N)
Sum
mer
(JJA
)Sp
ring
(MA
M)
Win
ter (
DJF
)
(b)
Figure 9 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in seasonal streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
Advances in Meteorology 15
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
basins revealed the significant deceasing trends at the rates of15 23 and 46 respectively Insignificant decreasingtrends were found only at Poonch river at the rate of 18
5 Discussions and Conclusions
UIB is a region that is famous for conflicting signals ofclimate and contrasting hydrological regime [15] (e basicreason for this anomalous behavior is the difference betweenaccumulation patterns of this region as reported by variousresearchers on the basis of geodetic mass balance and remotesensing data acquisition studies [12 24] (e results of thisstudy predicted that the climate change and accelerationover UIB may seriously affect streamflow in Indus river(ere are primarily three sources of streamflow in UIB ieglacier melt (Hunza Shigar and Shyok subbasins) followedby snowmelt (Astore and Gilgit subbasins) and precipita-tion (e hydrology of the Karakoram and Himalyan rangesis different as reported by various authors (e basic dif-ference is between the accumulation patterns of these tworegions Karakoram mountain receives precipitation by thewesterly disturbances and Himalyan range is controlled bythe summer monsoon [23] Climatic characteristics varyacross the Indus basin by large topographic variations fromvery high elevated to low elevated areas facing Himalayamountainous slopes Himalayans cover is one of the mostdynamic and complex mountain ranges in the world and isalso vulnerable to global warming and increasing humanactivities Uncertainties in the rate and magnitude of climatechange and potential impacts prevail but there is no doubtthat it is gradually and powerfully changing the ecologicaland socioeconomic landscape in the Himalayan regionparticularly in streamflows (e impact of warmer climateover snow fed basins is opposite to the impact on the glacierfed basins snow fed basins are more sensitive regardingreduction in the availability of water due to a compoundeffect of increase in evaporation and decrease in snow melt(e present study investigated the existence of trends andrelative changes in the annual and seasonal maximum andminimum temperature precipitation and streamflow fortwo consecutive 26-year periods (e results of this study
revealed that climate change is occurring remarkably withwarming trends in the lower part of Mangla catchmentwhereas cooling trends were observed at the higher elevationregions (e prevailing trends caused by climate changeinfluencing the flows should be considered by the watermanagers for better water management in a water-scarcecountry like Pakistan
Most of the river gauges during winter (DJF)) showedthe significant increasing river flows during the first-timeseries Mukhopadhyay and Khan [45] reported that pre-cipitation is controlled by elevation precipitation is almostnegligible below 2500m and there is little water yield be-tween 1000 and 2500m(ey defined ldquomid-altitude meltrdquo aswater that generates due to seasonal snows during intensewinter between elevations of 2500 and 3500m (ese in-creased river flows are more associated with increasedprecipitation than temperature during this period (eseresults are partially consistent with the findings of Khattaket al [17] who also reported increased winter flows due toincreased precipitation and temperature causing earlymelting of snow But we find a significant decrease in riverflows during the second third and fourth data periodswhich is consistent with decreased mean temperature duringthe same periods
We observed significant changes in the second period ascompared to the first period within UIB which are con-sistent and in agreement with the global warming trendsreported by Hasson et al [32] Our results indicated highvariability of trends in annual and seasonal minimumtemperature but these trends appeared to be more signif-icant and higher in magnitude particularly during thesecond period Similarly we observed significant wettingpattern of trends in annual and seasonal precipitation athigher altitude region but significant drying trends wererevealed by lower- or mid-altitude stations Most of thestations within UIB exhibited significant drying trendswhich are consistent with the recent studies carried out inthis region [19] (e annual runoff has been decreasedsignificantly within Jhelum river basin at Azad PattanChinari Domel Kohala Muzaffarabad and Palote andwithin Indus at Chahan Gurriala Khairabad Karora and
(a) (b)
Figure 10 Spatial distribution of trends detected by MannndashKendall test and estimated by Senrsquos method in annual streamflow showingchange in of data period averages (upward and downward arrows show positive and negative trends respectively blue arrow showssignificant trend at α 005 and green arrow shows insignificant trend)
16 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
Kalam during the second data period Similarly we observedthat seasonal runoff has been decreased significantly in allseasons except winter during second data series (e in-creasing trends of winter runoff are more associated withwesterly precipitation as compared to concurrent meltingtemperature because hydrology of the UIB is dominated bywinter precipitation (westerly disturbances) as compared tosummer monsoon offshoots Moreover the decreasingtrends of summer runoff are attributed to decreased meltingrate consistent with summer cooling reported by variousauthors [46 47] (e decreasing trends in summer dischargeshow least melting rates in summer resulting in stability ofglaciers and consequently positive basin storage Climatechanges occur most noticeably in terms of temperature andprecipitation over the UIB according to various authorsMoreover this study found the spring season to be quite drysupporting the idea of declining precipitation (reported bynumerous studies carried out earlier in this area) Down-stream areas in the lower portions of the drainage basin(where most of the population depends on the agriculture)are being affected by decreasing rainfall and its impacts oncrop sowing and harvesting times (ere will be more stresson available water resources (which are already scarce) ifprecipitation does not show any significant upsurge in-creased dryness could further stress agricultural productionTo avoid this potentially distressing situation from gettingworse water resources management must play an importantrole to ensure the best utilization of available resources forexample flood control building dams and reservoirs liningof canals and water courses and conservative surface irri-gation (trickle and sprinkler irrigation)
Data Availability
(e hydrometeorological time-series data used to supportthe findings of this study are available from the corre-sponding author upon request
Conflicts of Interest
(e authors declare that they have no conflicts of interest
Acknowledgments
(is study was supported by the National Natural ScienceFoundation of China (nos 51509141 and 51809150)
References
[1] Q You J Min and S Kang ldquoRapid warming in the TibetanPlateau from observations and CMIP5 models in recentdecadesrdquo International Journal of Climatology vol 36 no 6pp 2660ndash2670 2016
[2] T Yao L (ompson W Yang et al ldquoDifferent glacier statuswith atmospheric circulations in Tibetan Plateau and sur-roundingsrdquoNature Climate Change vol 2 no 9 pp 663ndash6672012
[3] S Hasson V Lucarini S Pascale and J Bohner ldquoSeasonalityof the hydrological cycle in major south and southeast Asianriver basins as simulated by PCMDICMIP3 experimentsrdquoEarth System Dynamics vol 5 no 1 pp 67ndash87 2014
[4] H J Fowler and D R Archer ldquoHydro-climatological vari-ability in the Upper Indus Basin and implications for waterresourcesrdquo in Proceedings of the 7th IAHS Scientific Assemblyvol 295 pp 131ndash138 Foz do Iguaccedilu Brazil April 2005
[5] R Sharma ldquoImpacts on human health of climate and land usechange in the hindu kush-himalayan regionrdquo MountainResearch and Development vol 32 no 4 pp 480ndash486 2012
[6] P D Jones and A Moberg ldquoHemispheric and large-scalesurface air temperature variations an extensive revision andan update to 2001rdquo Journal of Climate vol 16 no 2pp 206ndash223 2003
[7] S E Fick and R J Hijmans ldquoWorldClim 2 new 1 km spatialresolution climate surfaces for global land areasrdquo Interna-tional Journal of Climatology vol 37 no 12 pp 4302ndash43152017
[8] D E Parker and E B Horton ldquoGlobal and regional climate in1998rdquo Weather vol 54 no 6 pp 173ndash184 1999
[9] R W Portmann S Solomon and G C Hegerl ldquoSpatial andseasonal patterns in climate change temperatures and pre-cipitation across the United Statesrdquo Proceedings of the Na-tional Academy of Sciences vol 106 no 18 pp 7324ndash73292009
[10] S K Muhammad Qasim ldquoHydro-meteorological character-istics of Indus river basin at extreme north of PakistanrdquoJournal of Earth Science amp Climatic Change vol 05 no 1pp 1ndash6 2013
[11] A F Lutz H W ter Maat H Biemans A B ShresthaP Wester and W W Immerzeel ldquoSelecting representativeclimate models for climate change impact studies an ad-vanced envelope-based selection approachrdquo InternationalJournal of Climatology vol 36 no 12 pp 3988ndash4005 2016
[12] A F Lutz W W Immerzeel P D A KraaijenbrinkA B Shrestha and M F P Bierkens ldquoClimate change im-pacts on the upper Indus hydrology sources shifts and ex-tremesrdquo PLoS One vol 11 no 11 Article ID e0165630 2016
[13] Z H Dahri F Ludwig E Moors B Ahmad A Khan andP Kabat ldquoAn appraisal of precipitation distribution in thehigh-altitude catchments of the Indus basinrdquo Science of eTotal Environment vol 548-549 pp 289ndash306 2016
[14] MM Sheikh N Manzoor M Adnan J Ashraf and A KhanClimate Profile and Past Climate Changes in Pakistan GCISC-RR-01 Islamabad Pakistan 2009
[15] H J Fowler and D R Archer ldquoConflicting signals of climaticchange in the upper Indus basinrdquo Journal of Climate vol 19no 17 pp 4276ndash4293 2006
[16] D Bocchiola and G Diolaiuti ldquoRecent (1980ndash2009) evidenceof climate change in the upper Karakoram Pakistanrdquo e-oretical and Applied Climatology vol 113 no 3-4 pp 611ndash641 2013
[17] M Khattak M Babel and M Sharif ldquoHydro-meteorologicaltrends in the upper Indus River basin in Pakistanrdquo ClimateResearch vol 46 no 2 pp 103ndash119 2011
[18] B Mukhopadhyay and A Khan ldquoA reevaluation of thesnowmelt and glacial melt in river flows within upper Indusbasin and its significance in a changing climaterdquo Journal ofHydrology vol 527 pp 119ndash132 2015
[19] Y Latif M Yaoming and M Yaseen ldquoSpatial analysis ofprecipitation time series over the upper Indus basinrdquo eo-retical and Applied Climatology vol 131 no 1-2 pp 761ndash7752018
[20] D Archer ldquoContrasting hydrological regimes in the upperIndus basinrdquo Journal of Hydrology vol 274 no 1ndash4pp 198ndash210 2003
Advances in Meteorology 17
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology
[21] A F Lutz W W Immerzeel A B Shrestha andM F P Bierkens ldquoConsistent increase in high Asiarsquos runoffdue to increasing glacier melt and precipitationrdquo NatureClimate Change vol 4 no 7 pp 587ndash592 2014
[22] R R Wijngaard A F Lutz S Nepal et al ldquoFuture changes inhydro-climatic extremes in the upper Indus Ganges andBrahmaputra river basinsrdquo PLoS One vol 12 no 12 ArticleID e0190224 2017
[23] D R Archer and H J Fowler ldquoSpatial and temporal varia-tions in precipitation in the upper Indus basin global tele-connections and hydrological implicationsrdquo Hydrology andEarth System Sciences vol 8 no 1 pp 47ndash61 2004
[24] W W Immerzeel N Wanders A F Lutz J M Shea andM F P Bierkens ldquoReconciling high-altitude precipitation inthe upper Indus basin with glacier mass balances and runoffrdquoHydrology and Earth System Sciences vol 19 no 11pp 4673ndash4687 2015
[25] K Hewitt ldquo(e karakoram anomaly Glacier expansion andthe elevation effectrsquo Karakoram Himalayardquo Mountain Re-search and Development vol 25 no 4 pp 332ndash340 2005
[26] M Sharif D R Archer H J Fowler and N Forsythe ldquoTrendsin timing and magnitude of flow in the upper Indus basinrdquoHydrology and Earth System Sciences vol 17 no 4pp 1503ndash1516 2013
[27] A T Wolf J A Natharius J J Danielson B S Ward andJ K Pender ldquoInternational river basins of the worldrdquo In-ternational Journal of Water Resources Development vol 15no 4 pp 387ndash427 1999
[28] I Ahmad F Zhang M Tayyab et al ldquoSpatiotemporal analysisof precipitation variability in seasonal annual and extremevalues over upper Indus River basinrdquo Atmospheric Researchvol 213 pp 346ndash360 2018
[29] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindukush-Karakoram-Himalaya upper Indus basinrdquo Earth System Dynamics Dis-cussions vol 6 no 1 pp 579ndash653 2015
[30] S R Bajracharya and B Shresthae Status of Glaciers in theHindu Kush-Himalayan Region Springer Kathmandu Nepal2011
[31] R Bhambri T Bolch P Kawishwar D P DobhalD Srivastava and B Pratap ldquoHeterogeneity in glacier re-sponse in the upper Shyok valley northeast Karakoramrdquo eCryosphere vol 7 no 5 pp 1385ndash1398 2013
[32] S Hasson J Bohner and V Lucarini ldquoPrevailing climatictrends and runoff response from Hindu-kushndashKarakoramndashHimalaya upper Indus basinrdquo Earth SystemDynamics vol 8 no 2 pp 337ndash355 2017
[33] S U Hasson S Pascale V Lucarini and J Bohner ldquoSeasonalcycle of precipitation over major river basins in south andsoutheast Asia a review of the CMIP5 climate models data forpresent climate and future climate projectionsrdquo AtmosphericResearch vol 180 pp 42ndash63 2016
[34] F Wilcoxon ldquoIndividual comparisons by ranking methodsrdquoBiometrics Bulletin vol 1 no 6 p 80 1945
[35] S Yue and C Wang ldquo(e Mann-Kendall test modified byeffective sample size to detect trend in serially correlatedhydrological seriesrdquo Water Resources Management vol 18no 3 pp 201ndash218 2004
[36] F Fathian Z Dehghan M H Bazrkar and S EslamianldquoTrends in hydrologic and climatic variables affected by fourvariations of Mann-Kendall approach in Urmia lake basinIranrdquo Hydrological Sciences Journal vol 6667 pp 1ndash13 2016
[37] N M Kehrwald L G (ompson Y Tandong et al ldquoMassloss on Himalayan glacier endangers water resourcesrdquo
Geophysical Research Letters vol 35 no 22 Article IDL22503 2008
[38] W K M Lau M-K Kim K-M Kim and W-S Lee ldquoEn-hanced surface warming and accelerated snow melt in theHimalayas and Tibetan Plateau induced by absorbing aero-solsrdquo Environmental Research Letters vol 5 no 2 Article ID025204 2010
[39] T Rientjes A T Haile and A A Fenta ldquoDiurnal rainfallvariability over the upper blue Nile basin a remote sensingbased approachrdquo International Journal of Applied EarthObservation and Geoinformation vol 21 pp 311ndash325 2013
[40] A A Fenta H Yasuda K Shimizu et al ldquoSpatial distributionand temporal trends of rainfall and erosivity in the easternAfrica regionrdquo Hydrological Processes vol 31 no 25pp 4555ndash4567 2017
[41] X Zhang L A Vincent W D Hogg and A NiitsooldquoTemperature and precipitation trends in Canada during the20th centuryrdquo Atmosphere-Ocean vol 38 no 3 pp 395ndash4292000
[42] L Pizzolato S E L Howell C Derksen J Dawson andL Copland ldquoChanging sea ice conditions and marinetransportation activity in Canadian Arctic waters between1990 and 2012rdquo Climatic Change vol 123 no 2 pp 161ndash1732014
[43] P K Sen ldquoEstimates of the regression coefficient based onKendallrsquos Taurdquo Journal of the American Statistical Associationvol 63 no 324 pp 1379ndash1389 1968
[44] T G Huntington and M Billmire ldquoTrends in precipitationrunoff and evapotranspiration for rivers draining to the Gulfof Maine in the United Statesrdquo Journal of Hydrometeorologyvol 15 no 2 pp 726ndash743 2014
[45] B Mukhopadhyay and A Khan ldquoRising river flows and glacialmass balance in central Karakoramrdquo Journal of Hydrologyvol 513 pp 192ndash203 2014
[46] E Palazzi J Von Hardenberg and A Provenzale ldquoPrecipi-tation in the hindu-kush karakoram himalaya observationsand future scenariosrdquo Journal of Geophysical Research At-mospheres vol 118 no 1 pp 85ndash100 2013
[47] U Minora D Bocchiola C D A Agata et al ldquo2001ndash2010glacier changes in the central Karakoram national park acontribution to evaluate the magnitude and rate of theldquoKarakoram anomalyrdquo e Cryosphere Discussions vol 7no 3 pp 2891ndash2941 2013
18 Advances in Meteorology