Changing Climate, Landcover and Water Resources in the ... · Agricultural and industrial...
Transcript of Changing Climate, Landcover and Water Resources in the ... · Agricultural and industrial...
Changing Climate, Landcover and Water Resources in the Mountains of Central Asia over the last 60 years
Vladimir Aizen and Elena AizenUniversity of Idaho
• Population grows• Agricultural and industrial expansion/demand
1900 – 15M
2000 – 150M
Estimation of actual water resources of Central Asia and their changes for the last 60 years
- Climate- Seasonal snow cover- Glaciers
Water resources of Central Asia probability forecasting
The Goal
Data:Long-term surface observational data (air temperature, precipitation, annual datesof snow appearance and disappearance), large scale topographic maps, aerialphotographs, and assimilated remote sensing information. (Corona, Hexagon KH-9,Landsat, Aster, SRTM and MODIS)
Methods: - Differences in averages (d= AVE1973-03 - AVE1942-72) for two thirty-year periods 1973-2003 (AVE1973-03) and 1942-1972 (AVE1942-72) and T-test at 20% for precipitation and 10% for air temperature level- Linear trends (α) for two periods (α1942-72 and α1973-03); coefficients of determination, F tests, at 80% for precipitation and 90% for air temperature level of significance- Acceleration in changes for the last thirty years: a = α1973-03 - α1942-72; same significance as for the linear trends- Differences in standard deviations for two periods (dstd = std1973-03- std1942-72) and T-test at 20% for precipitation and 10% for air temperature level. - Geographically Weighted Regression (GWR) spatial interpolation method to interpolate spatial gaps in the meteorological data - Georeferenced and orthorectified image processing and spectral analysis
Some latest publications on climate changes in Central Asian
RegionNumber stations
PeriodResoluti
onResults
Authors
Tien Shan200 > 4000 m
110 1940-1991
MonthlyAir temperature +0.01C/yrPrecip.+1.2 mm /yr<2000m
Aizen, et al., 1997
Central Asia35-50N 75-120E
321951-1990
SummerMongolia and northern China
negative trendYatagai &
Yasunari, 1995
Tajikistan800-4000 m
41930-1991
AnnualAir temperature +2.2C and +0.4C/yr
Осадки+0.05 - 0.25 mm/yr<1000 m+1.82 - +5/37 >2000 m
Finaev, 1995
Central Asiaplains and foothills
26 +501891-1991
Annual and
summer
Steady positive trend in air temperature. Decrease of total river runoff and increase in its variability for 1962-91 comparing for 1931-60
Konovalov,2003; Konovalov&Williams, 2005
Central Asia68 - 3614 m
39- 45N 62-78E21
1879-2001
AnnualGrowth of air temperature+ 0.027 C /yr
Giese et al., 2007
Xinjiang 91979–1999
SummerSuccessive droughts for 3 summers(1997-99) at northern China
Xu, 2001
CLIMATE
8450
3828
1812117
0 100
<500500-1000
1000-15001500-20002000-25002500-30003000-3500
>3500m
Number of stations
23
111
117
Beginning of observations 1855-1900
1901-1941
1942-1952
Major period of observations: 1942-1972 and1973-2003
251 meteor-stations used in our analysis Differentiation by climatic regions
Number of stations by elevations
Difference in 30-year averages of annual (A) and summer (B) air temperature(ΔT = AVE1973-2003 – AVE1942-1972)
0.00.10.20.30.40.50.60.70.8
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
irE
aste
rn P
amir
Alta
i-Say
anN
orth
ern
Tien
Sha
nW
este
rn T
ien
Sha
nIs
sik
Kul
Cen
tral T
ien
Sha
nIn
ner T
ien
Sha
nE
aste
rn T
ien
Sha
nK
azak
hsta
n st
eppe
Mon
golia
Tarim
dTan , C
0.00 0.50 1.00
-135 -1000
1000 -2000
2000 -3000
> 3000
dTan , C
H, m
Difference by altitudinal zones, оС
Difference by regions, оС
A
B
,oC
,oC
0.65oC thirty year difference
0.64oC thirty year difference
-0.03
-0.02
-0.01
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
ir
East
ern
Pam
ir
Alta
i-Say
an
Nor
ther
n Ti
en S
han
Wes
tern
Tie
n Sh
an
Issi
k Ku
l
Cent
ral T
ien
Shan
Inne
r Tie
n Sh
an
Kaza
khst
an s
tepp
e
East
ern
Tien
Sha
n
Mon
golia
Tari
m
aTan
aTs
°C
0.00 0.01 0.02 0.03 0.04
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
aTs
aTan
aT°C
Acceleration of changing annual (aTan) and summer (aTs) air temperatures in Central Asia by regions and altitudes for the last 30 years
Acceleration by regions, оС Acceleration by altitudinal zones, оС
-4-202468
10121416
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
irEa
ster
n Pa
mir
Alta
i-Say
anN
orth
ern
Tien
Sha
nW
este
rn T
ien
Shan
Issi
k Ku
lCe
ntra
l Tie
n Sh
anIn
ner T
ien
Shan
Kaza
khst
an s
tepp
eEa
ster
n Ti
en S
han
Mon
golia
Tari
m
dPan,km3
-30.0
-25.0
-20.0
-15.0
-10.0
-5.0
0.0
5.0
10.0
15.0
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
irEa
ster
n Pa
mir
Alta
i-Say
anN
orth
ern
Tien
Sha
nW
este
rn T
ien
Shan
Issi
k Ku
lCe
ntra
l Tie
n Sh
anIn
ner T
ien
Shan
Kaza
khst
an s
tepp
eEa
ster
n Ti
en S
han
Mon
golia
Tari
m
dPan,mm
-20 0 20
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
dPan, km3
-20 -10 0 10
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
dPan, mm
Differences in 30-year averages of annual precipitation (ΔPan= avePan1973-2003 – avePan1942-1972)over Central Asia
Difference in annual precipitation by regions, mm
Difference in annual precipitationby area of regions, km3Average weighted altitudinal
difference in annual precipitation by area, km3
Average weighted altitudinal difference in annual precipitation, mm
, mm
Surplus in annual precipitation 355 km3
0
10
20
30
40
50
60
70
80
90
1930194019501960197019801990
Annual and decadal total particle concentration
0.0E+00
1.0E+05
2.0E+05
3.0E+05
4.0E+05
1910192019301940195019601970198019902000Year
Conc
(num
/mL)
Overall decadal trends show the high dust loading for the 1960’s and 70’s, with maximum dust loadingapparent for the 30’s and that is in accordance with results from 154 Chinese stations on maximumfrequency of dust weather for the mid-1960’s (Qian et al; Sun et al., 2002) and the lowest in the 90’s to beone-fifth that of the 60’s.
-25
-20
-15
-10
-5
0
5
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
ir
East
ern
Pam
ir
Alta
i-Say
an
Nor
ther
n Ti
en S
han
Wes
tern
Tie
n Sh
an
Issi
k Ku
l
Cent
ral T
ien
Shan
Inne
r Tie
n Sh
an
Kaza
khst
an s
tepp
e
East
ern
Tien
Sha
n
Mon
golia
Tari
m
dPs, mm
-12
-10
-8
-6
-4
-2
0
2
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
ir
East
ern
Pam
ir
Alta
i-Say
an
Nor
ther
n Ti
en S
han
Wes
tern
Tie
n Sh
an
Issi
k Ku
l
Cent
ral T
ien
Shan
Inne
r Tie
n Sh
an
Kaza
khst
an s
tepp
e
East
ern
Tien
Sha
n
Mon
golia
Tari
m
dPs, km3
-15 -10 -5 0
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
dPs,mm
H, m
-15 -10 -5 0
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
dPs,km3
H, m
Differences in 30-year averages of summer precipitation (ΔPs= avePs1973-2003 – avePs1942-1972) over Central Asia
Difference in average weighted summer precipitation by regions, mm
Difference in average weighted summerprecipitation by area of regions, km3
Altitudinal differences in average weighted summer precipitation, mm
Altitudinal differences in summer
precipitation by area, km3
, mm
Difference in 30 -year averages of winter precipitation(ΔPw = AVE1973-2003 – AVE1942-1972)
-10
-5
0
5
10
15
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
irEa
ster
n Pa
mir
Alta
i-Say
anN
orth
ern
Tien
Sha
nW
este
rn T
ien
Shan
Issi
k Ku
lCe
ntra
l Tie
n Sh
anIn
ner T
ien
Shan
Kaza
khst
an s
tepp
eEa
ster
n Ti
en S
han
Mon
golia
Tari
m
dPw, mm
-202468
1012
Ara
l-Cas
pean
…W
este
rn P
amir
East
ern
Pam
irA
ltai-S
ayan
Nor
ther
n Ti
en S
han
Wes
tern
Tie
n Sh
anIs
sik
Kul
Cent
ral T
ien
Shan
Inne
r Tie
n Sh
anKa
zakh
stan
ste
ppe
East
ern
Tien
Sha
nM
ongo
liaTa
rim
dPw, km3
-5 0 5 10 15 20
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
dPw, km3
-5 0 5 10
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
dPw, mm
Difference in average weighted winter precipitation by regions, mm
Difference in winterprecipitation by area of regions , км3
Difference in average weightedwinter precipitation by elevation, mm
Difference in winter precipitation by elevations, km3
, mm
-3
-2
-1
0
1
2
3
4
Ara
l-Cas
pean
des
erts
Wes
tern
Pam
ir
East
ern
Pam
ir
Alta
i-Say
an
Nor
ther
n Ti
en S
han
Wes
tern
Tie
n Sh
an
Issi
k Ku
l
Cent
ral T
ien
Shan
Inne
r Tie
n Sh
an
East
ern
Tien
Sha
n
Kaza
khst
an s
tepp
e
Mon
golia
Tari
ma, mm yr-1
-2 0 2 4 6
-135 - 1000
1000 - 2000
2000 - 3000
> 3000
a, mm yr-1
H, m
Acceleration of changing annual precipitation for the last 60 years (Δ = SLOPE1973-2003 – SLOPE1942-1972)
Annual acceleration by regions
Annual acceleration by altitudes
, mm
Snow covered areas by 1,000m isohyps over the Tien Shan for the last twenty years reconstructed by surface observational, AVHRR and MODIS data
0
20
40
60
80
snow cover, %
1987 2000 2001 2002 2003 2004 2005 2006 2007199919981997199619951994199319921991199019891988
ten days AVHRR data calibrated with surface observational data eight days MODIS data
Duration of snow melt from the date of maximum snow cover to date of it’s disappearance reduced on 30 days during the last twenty years, equal 138 days in 2007. Snow melt 30 days faster then 20 years ago. The decrease of snow cover is not linear process.
SEASONAL SNOW COVER
Tien Shan, number of days with snow 2000-2001
The seasonal snow covered area in Tien Shan decreased by 15% approximately 120 000 km2
Tien Shan, number of days with snow 2006-2007
GLACIERSSome recent publication on Central Asia glacier changes
Authors Regions Period Data and methods Area of glacier recession, km2
Khromova , etal., 2003
Akshiirak, Inner Tien Shan
1977-2001 Map 1977, ASTER image 406.8-93.6 (-23%)
Vilesov and Uvarov, 2001
Zailiiskiy Alatau’N. Tien Shan
1955-1990 Aerial photography 287.3-81.8 (-29%)
Bolch, 2006 Zailiiskiy AlatauN. Tien Shan
1979-1999 1:100000 maps, Landsat ETM 198.37-34.2 (-17.3%)
Niederer et al.2008
Sokuluk River basin, N. TienShan
1963-1986 1:25,000 maps, KFA1000 satellite photo 31.7-4.2 (-13.3%)
1986-2000 Landsat ETM+ 27.5-4.7 (-17.1%)
Ye et al. 2005 Glacier № 1Urumqi R. basinE. Tien Shan
1962-2003 Maps 1962, 1964, 1986, 1992, 1994, 2000 and 2001
1.94-0.24 (-12.4%)
Narama et al.,2006
Terskei AlatooN. Tien Shan
1971-2002 Corona , Landsat ETM+ 245-18 (-8%)
Liu et al., 2006 Aksu R. basinC. Tien Shan
1963-1999 Maps 1:100000, Landsat TM и ETM 176-58.6 (-3.3%)
Kaidu R. basin,C. Tien Shan
1963-2000 33-38.5 (-11.6%)
Aizen, et al. 2007 Akshiirak,Inner Tien Shan
1977-2003 Aerial photography, ASTER 406.835.15 (-8.6%)
Glacier covered area recession during the last 30 years:
• Тянь-Шань -709 км2 (-7.1%)• Алтай -86 км2 (-6.2%)
1973
2003
8.3% 12%4,6% 6.5%3%
Atbashi glacierized area, Inner Tien Shan, -5.6% area reduction for the last 30 yearsSeptember 2003 September 1973
Borohoro glacierized area, Eastern Tien Shan, -5.7% area reduction for the last 30 yearsSeptember 2003 September 1973
Djungarskiy Alatau glacierized area, Northern Tien Shan -8.0% area reduction for the last 30 years
September 2003 September 1973
Aksiirak glacierized massif
182 glaciers
427 km2 glacierized area
(aerial photogrammetry 1943)
19431977182 glaciers
406.8 km2 glacierized area
4.2% area reduction
(aerial photogrammetry 1943/1977)
2003178 glaciers
371.6 km2 glacierized area
8.7% area reduction
(aerial photogrammetry 1977/ASTER2003)
Inner Tien Shan
2003 2002199519771956194318691800
Petrov Glacier
To maintain Central Asian glaciers at the current state, the increasing summer airtemperature at the ELA (TsELA) (equilibrium line altitude) must be offset by acorresponding increase in annual precipitation. For example, the glaciers of Tien Shanwill not retreat if an increase in mean summer air temperature on 1.0 ºC atELAcoincides with an increase of annual precipitation of 100 mm atELA.
PELA = 0.0995TsELA + 0.7441
R2 = 0.980.6
0.7
0.8
0.9
1.0
1.1
-1 0 1 2 3 4 TsELA
PELA, m
Glaciers exist while ELA is below the upper boundary of GCA (glacier covered area) inthe basin. This chain can be diagrammed as follows: climate → ELA → Glacierdimensions/configuration, and, ultimately, glacier ice volume.
Forecasted decrease in number of glaciers (K), glacier covered areas (S) andvolume (V) relatively to current glacier covered area under theELA moving up
Both models forecast that significant glacier degradation begins when ELA is increased by 600 m . The Central Asia GCA may shrink to about half of the current state if ELA increases another 1000 m. The number of glaciers could decrease by 40% and glacier volume by 60%.
The annual runoff of the major Tien Shan rivers is on average 67 km3 yr-1, which includes glacial melt of about 14 km3 yr-1 (20%)
For the last thirty years (1973-2003), the long-term mean runoff on average increased by 2% compared with previous thirty years, while thirty year mean in annual maximum runoff decreased by 5% of average
RIVER RUNOFF
Relative changes of the last thirty year annual mean (dQan/Qan) and maximum (dQmax/Qmax) river runoff in comparison to sixty year averages, % , and changes in dates of maximum river runoff (ddQmax).
%
Differences in 30-year averages of annual (dQan) and glacier (Qgl) river runoff (a) and their relation (b).
a b
Significant temporal and spatial changes have occurred in Central Asia in the inter-annualsurface water distribution between upper, middle, and lower river reaches, while annualrunoff did not changed significantly.
Precipitation and potential evaporation have similar ranges.
1957 1961 1965 1969 1973 1977 1981 1985 1989 1993
Years
Laye
r w.e
., m
m
Precipitation Runoff
-1000
100
300
500
700
900
1100
1997 2001 2005
Potential Evaporation
Average computational river runoff , precipitation and potential evaporation for the Tien Shan region during last 50 years
0.3
0.8
1.3
1.8
2.3
T T+1 T+2 T+3 T+4 T+5
E/Ec,P/PcR/Rc, E*/E*c
Sir Dar'ya P/PcE/EcR/RcE'/Ec'
Ratios between predicted by 2100 and current river runoff (R/Rc), evapotranspiration(E/Ec), potential evaporation (E*/E*c) under predicted changes of air temperature (Tc+d;d=1,2,…5°C) and precipitation (mPc, m=0.9; 1, 1.1, …1.5) for the Sir Dar’ya R. basin.
The Magicc&ScenGen Global Climatic Model (IPCC, 2001) scenarios considered that annual average air temperature in Central Asia by 2100 increases between 1.8 and 4.4°С and precipitation by 6% of current rate.
Rapid current decline of water resources in central Asia related to factors such as :
- (i) the rise of global and regional air temperatures
- (ii) shrinkage of seasonal snow cover and degradation of glaciers
- (iii) decrease of precipitation in the Alpine areas
- (iv) partitioning among snow and rain, evaporation fluxes
- (v) poor management of regional water resources.
The diminishing natural water storages significantly affect river runoff, lake levels, andgroundwater in aquifers, and contribute to progressive droughts that causesalinization and desertification in central Asia.
Conclusion