O

Fa

NMa Ab T

Ecohydrology & Hydrobiology 14 (2014) 182–191

A

Art

Re

Re

Ac

Av

Ke

Ind

Cli

An

SW

§

Sy

Ha

Te

sp

Hy*

am

(R.

(M

htt

16

res

riginal Research Article

low regime alteration due to anthropogenicnd climatic changes in the Kangsabati River, India§

eha Mittal a,*, Ashok Mishra a, Rajendra Singh a, Ajay Gajanan Bhave a,ichael van der Valk b

gricultural and Food Engineering Department, Indian Institute of Technology, Kharagpur, India

he Netherlands National Committee IHP-HWRP, The Netherlands

1. Introduction

Amidst a growing understanding that sufficient water isessential for the sustainability of riverine ecosystems,significant stress has been placed on assessing importantcharacteristics of the river flow regime needed to sustainecosystem functions. Poff et al. (1997) have discussed howthe ecological integrity of riverine ecosystems is a functionof flow characteristics and their interactions with waterquality, energy sources, physical habitat and biota. The

R T I C L E I N F O

icle history:

ceived 15 September 2013

ceived in revised form 27 May 2014

cepted 2 June 2014

ailable online 17 June 2014

ywords:

icators of Hydrologic Alteration

mate change

thropogenic intervention

AT

A B S T R A C T

According to the ‘natural flow paradigm’, any departure from the natural flow condition

will alter the river ecosystem. Flow regimes have been modified by anthropogenic

interventions and climate change is expected to cause additional impacts by altering

precipitation extremes. This study aims to evaluate the observed hydrologic alteration

caused by dam construction and simulate alteration due to expected climatic changes in a

monsoon dominated mesoscale river basin in India. To analyze the natural flow regime, 15

years of observed streamflow (1950–1965) prior to dam construction is used. Future flow

regime is simulated by a validated hydrological model Soil and Water Assessment Tool

(SWAT), using four high resolution (�25 km) Regional Climate Model (RCM) outputs for

the near future (2021–2050) based on the SRES A1B scenario. Finally, to quantify the

hydrological alterations of different flow characteristics, the Indicators of Hydrologic

Alteration (IHA) method which is based on the Range of Variability approach is used. This

approach enables the assessment of ecologically sensitive streamflow parameters for the

pre- and post-impact periods. Results of our analysis indicate that flow variability in the

river has been significantly reduced due to dam construction with high flows being

reduced and low flows during non-monsoon months considerably enhanced. Streamflow

simulated based on projected climatic changes reveals reduced monsoonal flows with

marginal changes in non-monsoon streamflow. The combined effect will reduce flow

variability, potentially affecting the ecosystem. We conclude that in such modified basins,

adaptive river basin management will be necessary to maintain such an extreme river flow

regime for the long term viability of riverine ecosystems.

� 2014 European Regional Centre for Ecohydrology of Polish Academy of

Sciences. Published by Elsevier Urban & Partner Sp. z o.o. All rights reserved.

This paper in its early version was presented at the International

mposium Ecohydrology, Biotechnology and Engineering: Towards

rmony between the Biogoeosphere and Society on the basis of Long-

rm Ecosystem Research held in Łodz, 17–19th September 2013. The

ecial issue of the symposium papers was published in Ecohydrology &

drobiology, Vol. 14, No. 1. 2014. Corresponding author. Tel.: +91 9775040454.

E-mail addresses: nehamitts@gmail.com (N. Mittal),

ishra@agfe.iitkgp.ernet.in (A. Mishra), rsingh@agfe.iitkgp.ernet.in

Singh), ajaybhave84@gmail.com (A.G. Bhave), info@hydrology.nl

. van der Valk).

Contents lists available at ScienceDirect

Ecohydrology & Hydrobiology

jo ur n al ho mep ag e: www .e lsev ier . c om / loc ate /ec o hyd

p://dx.doi.org/10.1016/j.ecohyd.2014.06.002

42-3593/� 2014 European Regional Centre for Ecohydrology of Polish Academy of Sciences. Published by Elsevier Urban & Partner Sp. z o.o. All rights

erved.

nvtifrase(P2

cineapesddflevdc(P2dtiHbuimdoac

sodTdfldcthga(MameS(Rthhfokmac

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191 183

atural dynamism of a river is demonstrated by the flowariability across hourly, daily, seasonal, annual and longerme scales. Primarily, five flow components; magnitude,equency, duration, timing and rate of change, which have

significant influence on the ecological dynamics of riverystems, have to be ‘mimicked’ in order to maintaincosystem diversity and the diversity of organisms thereinoff et al., 1997; Gibson et al., 2005; Arthington et al.,

006; Laize et al., 2013).Several studies over the past decade have discussed

hanges in river flow regimes brought about by humantervention (Gibson et al., 2005; Sun and Feng, 2012; Yang

t al., 2012; Poff and Matthews, 2013). Dams are known tolter natural flow regimes of rivers when compared to there-impoundment period and have been found to homog-nize distinct river flow regimes across broad geographiccales (Poff et al., 2007). Simplistic and static rules forelineating environmental flow requirements may beetrimental to the critical ecological functions of riverow variability and sustenance of ecosystems (Arthingtont al., 2006). Dampened seasonality and interannual flowariability of rivers due to dams alters the habitatynamics which makes assessment of these impactsritical from the point of view of biodiversity conservationoff et al., 2007). In their global analysis, (Doll and Zhang,

010) find that along with ecological impacts of wateriversion by dams, climate change induced flow altera-ons can have a strong impact on freshwater ecosystems.owever, a strong regional variation in dam impacts maye observed in such studies, and also might have beennderestimated. For an improved ability to gauge potential

plications for riverine ecosystems, which experienceifferent types of flow regimes across different regions,bserved changes at a basin scale need to be analyzed,long with flow alterations caused by future climatehange.

In this study, we analyze long-term observed andimulated streamflow at a location which lies downstreamf a large earthen dam on the Kangsabati river to evaluateam and future climate change induced flow alterations.his river, located in the lower Ganges basin, has beenammed to create the Kangsabati reservoir which providesood/drought relief and diverts water for irrigation andomestic purposes in the command area. Flow regimehanges due to dam construction could be exacerbated ine coming decades as future climate projections forecast a

eneral decrease in precipitation, increase in temperaturend an intensification of climatic extremes in the basin

ittal et al., 2013). Use of hydrologic indicators forssessing environmental flow requirements. From theyriad hydrologic indicators available for assessing

nvironmental flow requirements (Olden and Poff, 2003;uen, 2010), we employ the Range of Variability ApproachVA) (Richter et al., 1997) derived from aquatic ecologyeory which involves a detailed assessment of the

ydrological variability of river. This method is usefulr detecting changes in the hydrological regime due to a

nown perturbation and for establishing flow-based riveranagement targets (Chen, 2012) making it suitable for

nalyzing the impact of human intervention and climate

2. Methodology

2.1. Study area

The Kangsabati river basin is located mostly within theIndian state of West Bengal, bounded by the 868 E and878300 E longitudes and the 228200 N and 238300 Nlatitudes and having an approximate area of 5796 km2.The Kangsabati river flow in a south-easterly directionbefore becoming the last contributing river to the Gangesriver in India. The river basin is traditionally considereddrought prone (Saxena, 2012), although it receives anaverage annual rainfall of 1200 mm. The high concentra-tion of rainfall during monsoon (JJAS) months results in askewed streamflow distribution with large differences indaily and monthly river flows within the natural flowregime. The Kangsabati river which originates in theuplands of the Chhota Nagpur plateau has caused highlyeroded and gullied topography in the upper reacheswhere ephemeral 1st and 2nd order streams are domi-nant. This eroded landscape turns into undulatinguplands towards the middle reaches and finally alluvialplains in the lower reaches. The Kangsabati reservoirproject is located at the confluence of the Kangsabati riverand its major tributary; the Kumari. This reservoir, built intwo phases (1965 and 1973), receives inflow from bothsub-basins. Along with the provision for irrigation anddomestic water consumption needs the project alsoprovides 246.7 million m2 storage for flood relief. Alongwith these benefits, drought mitigation, pisciculture andrecreational facilities are also a part of the project (Saxena,2012). Some of the storage capacity has diminished overthe past 40 years due to siltation. As a result, the currentreservoir storage capacity is about 1/3rd of the annualreservoir inflow (Bhave et al., 2013b). Previous studies inthis basin (Bhave et al., 2013a,b; Mittal et al., 2013) havediscussed stakeholder experiences of past climaticchanges. Moreover, projected climatic changes in mid-21st century demonstrate an amplification of the impactsdue to changing precipitation extremes, particularly,consecutive dry days (CDD), highest five day precipitation(RX5D) and number of days with precipitation greaterthan 50 mm (PD50).

2.2. Analytical approach

Availability of long duration river discharge data is anessential requirement for assessing ecologically importantcharacteristics of the natural flow regime. Kennard et al.(2010) suggest that hydrologic metrics should be based onmore than 15 years of discharge data to reduce bias andincrease accuracy. Mohanpur, a discharge gauging stationof the Irrigation & Waterways Department, Government ofWest Bengal, approximately 80 km downstream of theKangsabati reservoir has recorded and made availabledaily discharge data from 1950 onwards. Natural river flowconditions existed for a 16-year period between 1950 and1965. The period between 1965 and 1973 constitutes anintermediate period between the completion of the Phase Iand II of the Kangsabati reservoir project. The 37-year

eriod from 1974–2010 marks the post-dam period where

hange on the Kangsabati river basin. p

dath

rudusigTahaflomcoclietpecliva

Ta

SW

R

S

T

R

K

M

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191184

m effects on flow regime are studied using Mohanpur ase reference location.A calibrated SWAT model is used for simulating future

noff (Section 2.3). Kangsabati reservoir is not consideredring the model setup, so as to isolate the climate changenal and analyze only the impact of climate change.king due notice of the oft-discussed uncertainty issues, its been proposed that climate-induced changes in riverw regime can be best analyzed using multiple climateodels (Doll and Schmied, 2012). Regionally relevant,mprehensive and state-of-the-art high resolution RCMmate simulations are therefore used here (Mathison

al., 2013; Mittal et al., 2013). SWAT simulations for theriod 1970–1999 are based on historical simulations ofmate models which represent the natural climaticriability in the basin, while SWAT simulations for the

period 2021–2050 represent future hydrologic conditionsunder a climate change scenario.

Based on this understanding, observed and simulatedflows are then analyzed with the help of the Indicators ofHydrologic Alteration (IHA) software of The NatureConservancy (2009). Pre- and post-impact periods arecompared in this approach to evaluate the degree ofhydrologic alteration caused by the intervention. More-over, recent addition of the Environmental Flow Compo-nents (EFCs) in the IHA software provide greater supportfor translating key flow characteristics into environmentalflow management (Mathews and Richter, 2007). The entirerange of hydrologic metrics available are compared todetermine key alterations in the natural flow regime andtheir potential implications for the riverine ecosystem arediscussed.

ble 1

AT model performance of five calibrated subbasins in the Kangsabati basin.

iver discharge station Calibration Validation

P factor R factor NSE R2 P factor R factor NSE R2

imulia 0.74 1.08 0.63 0.69 0.71 0.94 0.53 0.69

usuma 0.78 0.75 0.72 0.86 0.76 0.65 0.76 0.79

angagora 0.68 1.1 0.74 0.66 0.62 1.02 0.67 0.66

haridwar 0.72 0.84 0.64 0.75 0.79 0.9 0.64 0.75

ohanpur 0.4 0.59 0.68 0.87 0.63 0.72 0.49 0.85

Fig. 1. Map of the Kangsabati basin showing land cover land use classes, location of hydro-meteorological stations and Kangsabati reservoir.

2

rtysmrecthaeFcsomogthvqsth9rdth

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191 185

.3. SWAT model and data

SWAT 2009 (Neitsch et al., 2009) is used to simulateiver discharges for observed and future period. SWAT

pically operates on a daily time step and accounts forpatial heterogeneities by subdividing the basin intoultiple hydrological response units (HRUs). Each HRU

epresents a unique combination of soil, land cover andlevation. The rainfall-runoff model simulates the dis-harge from each sub basin and routes the streamflow to

e watershed outlet (Neitsch et al., 2009). Preprocessingnd model setup are performed using the Arc-SWATxtension for ArcGIS 9.3. The Sequential Uncertaintyitting algorithm (SUFI-2) (Abbaspour, 2007) is used toalibrate SWAT and to perform uncertainty analysis at fivetreamflow gauge stations in the Kangsabati Basin. Twoptimization criteria are used to assess model perfor-ance: Nash-Sutcliffe efficiency (NSE) and the coefficient

f determination (R2). Both NSE and R2 describe theoodness-of-fit between simulated and observed flow; ande model simulation would be considered best when their

alues approach one (Moriasi et al., 2007). SUFI-2uantifies the uncertainty using P factor and R factortatistics. The P factor, which varies from 0 to 1, represents

e fraction of observed discharge which falls within the5PPU band, while the R factor is derived by taking theatio of the average width of the 95PPU and the standardeviation of the observed discharge. While a value of lessan 1 is considered desirable for R factor, the ideal value

for P factor is 1 (100% values within the band) (Vaghefiet al., 2013). Input data used for SWAT simulation includesa 90 m � 90 m resolution Shuttle Radar TopographyMission (SRTM) DEM and soil data from survey toposheetsof the National Bureau of Soil Survey and Land UsePlanning (NBSS&LUP), Govt. of India. Unsupervised landuse classification of LandSat ETM and ETM+ images for thetime periods 1990 and 2001 are used for the calibration(1991–2000) and validation (2001–2010) periods respec-tively, while 2011 land use information is used as referencefor the future (2021–2050) period.

SWAT simulations are driven by four RCM simulationsand their ensemble mean (refered as MME hereafter). RCMsimulations are obtained by forcing two regional climatemodels REMO and HadRM3 with two CMIP3 GCMsECHAM5 (Roeckner, 2003) and HadCM3 (Gordon et al.,2000), under the SRES A1B scenario. The resulting foursimulations are REMO-ECHAM5, HadRM3-ECHAM5,HadRM3-HadCM3 and REMO-HadCM3. This is the onlyand the most comprehensive set of high resolution(�25 km) future climatic projections available for theKangsabati river basin developed under the EU projectHighNoon (http://www.eu-highnoon.org/). The perfor-mance of these RCMs for the Kangsabati basin has beenvalidated by comparing 20 year model simulations for theperiod 1989–2008, driven by lateral boundary forcingsfrom ERAInterim reanalysis data (Simmons et al., 2007),with the observational datasets; Climate Research Unit(CRU) for temperature and Asian Precipitation Highly

0

1

2

3

4

5

6

7

8

9

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Str

eam

flo

w (

m3/s

)

7-day min imum 30-day min imum

Ist ph ase IInd phase

a)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010

Str

eam

flo

w (

m3/s

)

1-day maxi mum 3-day maxi mum 7-day maxi mum 30 -day maxi mum

Ist ph ase IInd phase b)

Fig. 2. Changes in annual (a) minimum and (b) maximum streamflow from 1950 to 2010.

Retioabnu(Mapwbe(BsimsimsimSR

2.4

thdeda19efimthunthchdu

Fig

(19

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191186

solved Observational Data (APHRODITE) for precipita-n. Both the RCMs have demonstrated an adequateility to capture the seasonal characteristics and interan-al variability (IAV) of temperature and precipitationittal et al., 2013). The study also evaluates theplicability of these simulations for the Kangsabati basin,

hile the high spatial resolution of these simulations haveen found to be useful for hydrological assessmentshave et al., 2013a,b; Bhave et al., 2014). Two sets of RCM

ulations are used for the analysis; (i) historicalulations (20c3 m) for the period 1970–1999 (ii) futureulations for the period 2021–2050 based on the A1B

ES emission scenario.

. Indicators of Hydrologic Alteration

The IHA methodology, which is conceptually based one Range of Variability Approach (RVA), is used to assessgree of departure from the natural flow regime due tom and projected climate change (Richter et al., 1996,97). IHA, originally designed to analyze hydrologic

fects of dams by comparing streamflow in pre- and post-pact periods, assumes that natural flow alteration alterse ecosystem. It is useful to compare flow parametersder different scenarios (observed and future period inis study) and particularly for assessing key flow regimearacteristics, such as magnitude, timing, frequency andration (Pradhanang et al., 2013). To analyze the degree

of hydrologic alteration in ecologically relevant statistics,the IHA estimates 67 hydrologic parameters derived fromdaily flow statistics. These are subdivided into twocategories: 33 IHA parameters and 34 Environmental FlowComponents (EFC). Most widely used IHA parametersconsist of five major categories: (i) magnitude of monthlystreamflows; (ii) magnitude of and duration of annualextreme flows; (iii) timing of annual extreme flows; (iv)frequency and duration of high and low pulses and (v) rateand frequency of flow changes. The five flow componentswhich describe the ways in which an organism experiencesriver flow variability include: (i) extreme low flows; (ii)low flows; (iii) high-flow pulses; (iv) small floods and (v)large floods (Mathews and Richter, 2007). The IHA coversmost flow components and is useful for analysing highinformation and non-redundant indices. However, thechoice of relevant indices has to be assessed on a case-to-case basis (Olden and Poff, 2003). In this study, only thoseresults which depict significant changes in flow regime arepresented and discussed to increase their relevance.

3. Results and discussion

3.1. SWAT model calibration and validation

Results of the model performance and degree to whichit accounts for uncertainty are evaluated in terms of P

factor, R factor, NSE and R2 for both calibration and

908070605040302010

10

1

100

1,000

10

1

100

1,000

100

November

Excee dence prob abilit y

908070605040302010

908070605040302010

1

Excee dence prob abilit y

10

100

1

10

February

Pre-dam Post-dam

May

August

Flow

rate

(m3 /s

)

908070605040302010

. 3. Flow duration curves of seasonal streamflow for reference months February, May, August and November for pre-dam (1950–1965) and post-dam

74–2010) periods.

vfothmvinfotov0gpvaV

3

c

F

a

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191 187

alidation periods (Table 1). The P factor indicates that datar majority of the stations are well bracketed in predictinge uncertainty of the model, whereas the R factors areostly below or around 1. The small values for P and larger

alues for R factor in case of Mohanpur gauge stationdicate higher prediction uncertainties. NSE and R2 valuesr all stations reveal that discharge simulation is satisfac-ry based on the criteria given by Moriasi et al. (2007). NSE

alues lie between 0.6 and 0.8 whereas R2 values vary from.6 to 0.9 for majority of the stations except for Mohanpur. Ineneral, in the downstream of Kangsabati dam, the modelrediction has larger uncertainties. The poor calibration andalidation results in case of managed downstream flow haslso been observed in other studies (Faramarzi et al., 2010;aghefi et al., 2013) (Fig. 1).

.2. Flow regime alterations due to dam

Natural flow variability is essential to maintainhannel and floodplain habitats, which support aquatic

and riparian species (Poff et al., 1997). Alterations due todam (Figs. 2, 3 and 4a) reveal a marked reduction in peakflows, increase in low flows and because of water diversionfor irrigation and a reduction in total annual discharge. InFig. 4(a), middle, low and high alterations correspond tochanges in the median, 25th percentile and 75th percentileflows. Dam construction modifies the natural flow regimedirectly by storing high flows and maintaining certainminimum flows during the non-monsoon months (Fig. 4a).Absorbing high flows, which cause downstream floods, isan important function of the dam, while providing waterfor domestic consumption and irrigation in downstreamsections necessitates regular dam releases during non-monsoon months. Such alterations have been known tocause geomorphic simplification, floodplain disconnectionand disruption of lateral and longitudinal connectivity,thereby affecting habitat dynamics and making it difficultfor native biota to adapt (Bunn and Arthington, 2002; Poffet al., 2007). From pre-dam to post-dam period, the 1-daythrough 7-day minimum flows remain almost the same,

J F M A M

Middle alteration Low alteration High alteration Pre-dam Post-dam

Middle alteration Low alteration High alteration Pre-impact Post-impact

(a)

(b)

J J A S O N D

J F M A M

Dis

char

ge m

3 /s

Dis

char

ge m

3 /s

Hyd

rolo

gic

Alte

ratio

nH

ydro

logi

c A

ltera

tion

J J A S O N D

1.5

1

0.8

0.6

0.4

0.2

0

-0.2

-0.4

-0.6

-0.8

-1

-1.2

9

160

140

120

100

80

60

40

20

0

8

7

6

5

4

3

2

1

0

1

0.5

0

-0.5

-1

-1.5

ig. 4. Monthly median river flow rates and hydrologic alterations for (a) pre-dam (1950–1964) and post-dam (1974–2010) and (b) control (1970–1999)

nd future climate change (2021–2050) periods at the Mohanpur gauging station.

wabdiduan

Fig

ch

Ta

Ch

lev

S

(

1

3

7

3

9

1

3

7

3

9

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191188

hereas there is a marked increase in 30-day minimum ofout 127% (Table 2). There is a statistically significantfference (Mann–Whitney U-test) in the maximum flowsring the post-dam period. As shown in Fig. 2, the 30d 90-day minimum flows increase dramatically after

completion of dam construction. Such an effect is part ofthe regularly observed dampening effect of dams (Poffet al., 1997; Bunn and Arthington, 2002; Poff et al., 2007).Conversely 1-day through 90-day maximum flows havereduced significantly because the dam absorbs and stores

0

1

2

3

4

5

6

1970

1974

1978

1982

1986

1990

1994

1998

2022

2026

2030

2034

2038

2042

2046

2050

160

180

200

220

240

260

280

300

80

100

120

140

160

180

200

0

5

10

15

20

25

30

35

40

1970

1974

1978

1982

1986

1990

1994

1998

2022

2026

2030

2034

2038

2042

2046

2050

0

2

4

6

8

10

12

14

16

10

15

20

25

30

35

Frequency

Duration

Timing

Duration

Timing

Frequency

Extreme low flows Hi gh flow p ulses

Day

sFr

eque

ncy

Juli

an d

ays

(1-

366)

Control Future 25th perc entile 75th perc entile median

ble 2

anges in annual mean, minimum and maximum flows between pre- and post-dam periods. p-values from Mann–Whitney U-test at 10% significance

el.

treamflow

m3/s)

Pre-dam

(1950–1965)

Post-dam

(1974–2010)

Relative change

(%)

p-value

-day minimum 0.0 0.0 0.0 0.117

-day minimum 0.0 0.0 0.0 0.094

-day minimum 0.0 0.6 0.0 0.777

0-day minimum 1.1 2.5 127.3 0.287

0-day minimum 4.3 4.5 4.7 0.642

-day maximum 2240.0 1105.0 �50.7 0.007

-day maximum 1647.0 774.7 �53.0 0.014

-day maximum 1024.0 474.3 �53.7 0.040

0-day maximum 476.1 253.2 �46.8 0.051

0-day maximum 254.1 190.5 �25.0 0.077

. 5. Timing, frequency and duration of ecologically important extreme low flows and high flow pulses for the control (1970–1999) and future climate

ange (2021–2050) periods.

hw

cdaAMthwScwisAroinchfiM

hfitiec

F

c

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191 189

igh flows generated by isolated precipitation events alongith the monsoon high flows (Fig. 3).

Fig. 3 and Table 2 provide an estimate of flow durationurves, which depict the percentage of time a specifiedischarge has been equaled or exceeded during the pre-nd post-dam period, for four months (February, May,ugust and November) representative of four seasons (DJF,AM, JJAS and ON) respectively. In the post-dam period,ere is an overall increase in the flow rate in February,hile the zero-flow duration has significantly reduced.

ummer rice (Boro rice) water demand in the reservoirommand area during summer months may be associatedith significant reduction of high flows during May. There

also a reduction in flow rate especially for low flows inugust and November while high flows in Novemberemain consistent. Due to dam construction, the frequencyf occurrence of small floods (defined as having a returnterval between a 2 and 10 years) has reduced signifi-

antly. The dam has eliminated the large floods (defined asaving a return interval greater than 10 years). Siluroidshes, especially species such as Bagarius bagarius andystus gulio, which are found in the Kangsabati River, are

ighly sensitive to reduction in high flows. Benthic siluroidshes are very good indicators of environmental degrada-on (Wootton et al., 1996) and changes in the timing ofxtreme low flows and high flow pulses affects their lifeycle by disrupting various stages such as spawning, egg

hatching, rearing, movement onto the floodplains forfeeding and reproduction or migration upstream anddownstream (Poff et al., 1997). Channel bed sedimentationbrought about by reduction in floods also causes degrada-tion of their habitat (Lisle, 1989; Mishra et al., 2009). Theobserved decline in these two species since 1960 and theconsequent absence by 2002 near Mohanpur station(Mishra et al., 2009) may be associated with the hydrologicchanges brought about by Kangsabati dam. Overall, it isobserved that while low flows are enhanced, the high flowsare reduced by the dam leading to a dampening of theextreme seasonality of river flows.

3.3. Flow regime alterations due to climate change

The isolated impact of climate change on runoff, basedon MME climate projections (2021–2050), is simulatedwithout considering the effect of the Kangsabati dam usingSWAT. The results show an increase in monthly medianstreamflow for winter months, with the maximumincrease during December (Fig. 4b) as compared to controlperiod (1970–1999). Analysis of IHA and EFC parametersshows that there is an overall reduction in median flowsduring monsoon season with maximum reduction ob-served during August. As shown in Fig. 5, the mediantiming of the extreme low flows and high flow pulsescomes later in the year. However, whereas there is an

908070605040302010

10

100

1,000

908070605040302010

1

10

100

1,000

9080706050403020100.1

1

10

100

1,000

9080706050403020100.01

0.1

1

10

100

February May

August November

Flow

rate

(m3 /

s)

Flow

rate

(m3 /

s)

Excee dence prob abilit yExcee dence prob abilit y

Control Future

ig. 6. Flow duration curves of seasonal streamflow for reference months February, May, August and November for the control (1970–1999) and future

limate change (2021–2050) periods.

inhithputointhwmcoNoinRxFofo1 m80MmapduspinOnmhachaqth

4.

ofdocanacocemexthanDaanprdadeotnudausReinwreeclimtamch

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191190

crease in variability of timing of extreme low flows, forgh flows there is a decrease. There is a slight reduction ine frequency of both, extreme low flows and high flowlses, which affects the overall flow variability. Compared

the control period, the duration of low flows hascreased by approximately 11% while for high flow pulsesere is a change in the 25th–75th percentile rangeithout any change in the median duration. There is aarked difference between the flow duration curves forntrol and future period for months May, August andvember (Fig. 6). For May and August high flows are

tensified which may be associated with the increase in5D and PD50 precipitation extremes (Mittal et al., 2013).r February, results show a slight increase in streamflowr the extreme low flows, which always remain above the

3/s flow rate. The medium flows from approximately m3/s to about 1 m3/s show a significant decrease duringay. On the other hand, high flows during the post-onsoon November month shows are reduced fromproximately 1050 m3/s to 850 m3/s. An increase in theration of extreme low flows has been known to decreaseecies richness of macroinvertebrate assemblages andcrease the macrophyte biomass (Suren and Riis, 2010).

the other hand, floods of different sizes are essential foraintaining diversity of riparian plant species and aquaticbitat (Caruso et al., 2013). Therefore the projectedanges due to climate change may have an impact on theuatic habitat of the Kangsabati river thereby affectinge riparian plant species and aquatic biota.

Conclusions

This study provides a detailed basin scale assessment ecologically relevant flow alterations in a monsoon-minated river basin. 67 flow alteration parameters arelculated using IHA and are compared to pre-impacttural flow conditions for analysing the impact of damnstruction and future climate change by mid-21stntury. Parameters describing changes in long-termonthly average, timing, duration and frequency oftreme flow conditions show a significant change frome natural flow regime. Such changes impact aquaticd riparian habitat and will adversely affect key species.mpening effect of dams on hydrological variabilityd the extreme seasonality of river flows is highlyonounced. Significant overall flow reduction by them for provision of irrigation and domestic watermands may be exacerbated by climate change. On theher hand, the effects of existing dam operations mayllify some impacts of climate change. Moreover, futurem operations may not be similar to current and may beeful for managing the impacts of climate change.duction in precipitation and higher temperatureduced increase in evapotranspiration when combinedith expected increase in human water demand willduce water availability for the sustenance of riverineosystems. Lack of sufficient long term ecological data, a

itation in the assessment of habitat changes, necessi-te long-term integrated and interdisciplinary assess-ents, for addressing this gap in understanding. Other

nutrient influx and water quality parameters also need tobe monitored for better understanding of ecosystemimpacts which will inform the method of river restora-tion. Integrated assessments in the future may aid muchneeded adaptive river basin management, in such amodified river basin, to maintain the naturally extremeriver flow regime.

Conflict of interest

None declared.

Financial disclosure

None declared.

Acknowledgement

This work has been supported by the HighNoon project(www.eu-highnoon.org), funded by the European Com-mission Framework Programme 7 under Grant No. 227087.

References

Abbaspour, K.C., 2007. User manual for SWAT-CUP, SWAT calibration anduncertainty analysis programs. Swiss Federal Institute of AquaticScience and Technology, Eawag, Duebendorf, Switzerland.

Arthington, A.H., Bunn, S.E., Poff, N.L., Naiman, R.J., 2006. The challenge ofproviding environmental flow rules to sustain river ecosystems. Ecol.Appl. 16, 1311–1318.

Bhave, A.G., Mishra, A., Groot, A., 2013a. Sub-basin scale characterizationof climate change vulnerability, impacts and adaptation in an IndianRiver basin. Reg. Environ. Change. 1–12, http://dx.doi.org/10.1007/s10113-013-0416-8.

Bhave, A.G., Mishra, A., Raghuwanshi, N.S., 2013b. A combined bottom-upand top-down approach for assessment of climate change adaptationoptions. J. Hydrol., http://dx.doi.org/10.1016/j.jhydrol.2013.08.039.

Bhave, A.G., Mishra, A., Raghuwanshi, N.S., 2014. Evaluation of hydrologi-cal effect of stakeholder prioritized climate change adaptationoptions based on multi-model regional climate projections. Clim.Chang. 1–15, http://dx.doi.org/10.1007/s10584-014-1061-z.

Bunn, S.E., Arthington, A.H., 2002. Basic principles and ecological con-sequences of altered flow regimes for aquatic biodiversity. Environ.Manage. 30, 492–507.

Caruso, B.S., Edmondson, L., Pithie, C., 2013. Braided river flow andinvasive vegetation dynamics in the Southern Alps, New Zealand.Environ. Manage. 1–18.

Chen, H., 2012. Assessment of hydrological alterations from 1961 to 2000 inthe Yarlung Zangbo River. Tibet. Ecohydrol. Hydrobiol. 12 (2) 93–103.

Doll, P., Zhang, J., 2010. Impact of climate change on freshwater ecosys-tems: A global-scale analysis of ecologically relevant river flowalterations. Hydrol. Earth Syst. Sci. Discuss. 7 (1) 1305–1342.

Doll, P., Schmied, H.M., 2012. How is the impact of climate change on riverflow regimes related to the impact on mean annual runoff? A global-scale analysis. Environ. Res. Lett. 7 (1) .

Faramarzi, M., Yang, H., Mousavi, J., Schulin, R., Binder, C.R., Abbaspour,K.C., 2010. Analysis of intra-country virtual water trade strategy toalleviate water scarcity in Iran. Hydrol. Earth Syst. Sci. Discuss. 7 (2) .

Gibson, C.A., Meyer, J.L., Poff, N.L., Hay, L.E., Georgakakos, A., 2005. Flowregime alterations under changing climate in two river basins: impli-cations for freshwater ecosystems. River Res. Appl. 21 (8) 849–864.

Gordon, C., Cooper, C., Senior, C.A., Banks, H., Gregory, J.M., Johns, T.C.,Mitchell, J.F.B., Wood, R.A., 2000. The simulation of SST, sea ice extentsand ocean heat transports in a version of the Hadley Centre coupledmodel without flux adjustments. Clim. Dyn. 16 (2–3) 147–168.

Kennard, M.J., Mackay, S.J., Pusey, B.J., Olden, J.D., Marsh, N., 2010.Quantifying uncertainty in estimation of hydrologic metrics for eco-hydrological studies. River Res. Appl. 26 (2) 137–156.

Laize, C.L.R., Acreman, M.C., Schneider, C., Dunbar, M.J., Houghton-Carr,H.A., Florke, M., Hannah, D.M., 2013. Projected flow alteration and

ecological risk for Pan-European Rivers. River Res. Appl., http://dx.doi.org/10.1002/rra.2645. anges such as sediment yield, water temperature,

L

M

M

M

M

M

N

O

P

P

P

P

N. Mittal et al. / Ecohydrology & Hydrobiology 14 (2014) 182–191 191

isle, T.E., 1989. Sediment transport and resulting deposition in spawninggravels, north coastal California. Water Resour. Res. 25 (6) 1303–1319.

athews, R., Richter, B.D., 2007. Application of the indicators of hydro-logic alteration software in environmental flow setting. J. Am. WaterResour. Assoc. 43 (6) 1400–1413.

athison, C., Wiltshire, A., Dimri, A.P., Falloon, P., Jacob, D., Kumar, P.,Moors, E., Siderius, C., Stoffel, M., Yasunari, T., 2013. Regional projec-tions of North Indian climate for adaptation studies. Sci. Total Envi-ron., http://dx.doi.org/10.1175/2011jhm1388.1.

ishra, S.S., Acharjee, S.K., Chakraborty, S.K., 2009. Development of toolsfor assessing conservation categories of siluroid fishes of fresh waterand brackish water wetlands of South West Bengal, India. Environ.Biol. Fishes 84 (4) 395–407.

ittal, N., Mishra, A., Singh, R., 2013. Combining climatological andparticipatory approaches for assessing changes in extreme climaticindices at regional scale. Clim. Change 1–13.

oriasi, D.N., Arnold, J.G., Van Liew, M.W., Bingner, R.L., Harmel, R.D.,Veith, T.L., 2007. Model evaluation guidelines for systematic quanti-fication of accuracy in watershed simulations. Trans. ASABE 50 (3)885–900.

eitsch, S.L., Arnold, J.G., Kiniry, J.R., Williams, J.R., 2009. Soil and WaterAssessment Tool: Theoretical Documentation, Version 2009, http://twri.tamu.edu/reports/2011/tr406.pdf.

lden, J.D., Poff, N.L., 2003. Redundancy and the choice of hydrologicindices for characterizing streamflow regimes. River Res. Appl. 19 (2)101–121.

off, N.L., Matthews, J.H., 2013. Environmental flows in the Anthropo-cence: past progress and future prospects. Curr. Opin. Environ. Sus. 5,667–675.

off, N.L., Allan, J.D., Bain, M.B., Karr, J.R., Prestegaard, K.L., Richter, B.D.,Sparks, R.E., Stromberg, J.C., 1997. The natural flow regime. BioScience47 (11) 769–784.

off, N.L., Olden, J.D., Merritt, D.M., Pepin, D.M., 2007. Homogenization ofregional river dynamics by dams and global biodiversity implications.Proc. Natl. Acad. Sci. 104 (14) 5732–5737.

radhanang, S.M., Mukundan, R., Schneiderman, E.M., Zion, M.S., Anandhi,A., Pierson, D.C., Frie, A., Easton, Z.M., Fuka, D., Steenhuis, T.S., 2013.

Streamflow responses to climate change: analysis of hydrologicindicators in a New York City water supply watershed. J. Am. WaterResour. Assoc., http://dx.doi.org/10.1111/jawr.12086.

Richter, B.D., Baumgartner, J.V., Powell, J., Braun, D.P., 1996. A method forassessing hydrologic alteration within ecosystems. Conserv. Biol. 10(4) 1163–1174.

Richter, B.D., Baumgartner, J.V., Wigington, R., Braun, D.P., 1997. Howmuch water does a river need? Freshw. Biol. 37 (1) 231–249.

Roeckner, E., 2003. The atmospheric general circulation model ECHAM5:Part 1: model description. Max-Planck-Institut fur Meteorologie.

Saxena, R.P., 2012. Impacts of Kangsabati Project, India. In: Impacts ofLarge Dams: A Global Assessment. Springer, Berlin/Heidelberg, pp.277–298.

Simmons, A.S., Uppala, D.D., Kobayashi, S., 2007. ERA-interim: newECMWF reanalysis products from 1989 onwards. ECMWF Newsl.110, 29–35.

Suen, J.P., 2010. Potential impacts to freshwater ecosystems caused byflow regime alteration under changing climate conditions in Taiwan.Hydrobiologia 649 (1) 115–128.

Sun, T., Feng, M.L., 2012. Multistage analysis of hydrologic alterations inthe Yellow River, China. River Res. Appl., http://dx.doi.org/10.1002/rra.2586.

Suren, A.M., Riis, T., 2010. The effects of plant growth on stream inverte-brate communities during low flow: a conceptual model. J. Am. Benth.Soc. 29 (2) 711–724.

The Nature Conservancy, 2009. Indicators of hydrologic alteration, ver-sion 7.1. User’s manual, http://www.conservationgateway.org/Con-servationPractices/Freshwater/EnvironmentalFlows/Methodsand-Tools/IndicatorsofHydrologicAlteration/Pages/IHA-Software-Down-load.aspx.

Vaghefi, A.S., Mousavi, S.J., Abbaspour, K.C., Srinivasan, R., Yang, H., 2013.Analyses of the impact of climate change on water resources com-ponents, drought and wheat yield in semiarid regions: Karkheh RiverBasin in Iran. Hydrol. Process., http://dx.doi.org/10.1002/hyp.9747.

Wootton, J.T., Parker, M.S., Power, M.E., 1996. Effects of disturbance onriver food webs. Science, New York Then Washington, pp. 1558–1560.

Yang, Z., Yan, Y., Liu, Q., 2012. Assessment of the flow regime alterations inthe Lower Yellow River, China. Ecol. Inform. 10, 56–64.

View publication statsView publication stats