Ic

Sa

b

a

ARR1A

KCWUO

1

imultrssdip(daeni

M

0h

Resources, Conservation and Recycling 83 (2014) 176– 189

Contents lists available at ScienceDirect

Resources, Conservation and Recycling

jo ur n al hom epa ge : www.elsev ier .com/ locate / resconrec

ntegrated urban water management modelling underlimate change scenarios

antosh M. Pingalea,∗,1, Mahesh K. Jatb, Deepak Kharea

Department of Water Resource Development and Management, Indian Institute of Technology Roorkee, Roorkee 247 667 (UK), IndiaDepartment of Civil Engineering, Malaviya National Institute of Technology, Jaipur, Rajasthan, India

r t i c l e i n f o

rticle history:eceived 30 June 2013eceived in revised form8 September 2013ccepted 14 October 2013

eywords:limate change

a b s t r a c t

The concept of integrated water management is uncommon in urban areas, unless there is a shortageof supply and severe conflicts among the users competing for limited water resources. Further, prob-lem of water management in urban areas will aggravate due to uncertain climatic events. Therefore, anIntegrated Urban Water Management Model considering Climate Change (IUWMCC) has been presentedwhich is suitable for optimum allocation of water from multiple sources to satisfy the demands of dif-ferent users under different climate change scenarios. Effect of climate change has been incorporated innon-linear mathematical model of resource allocation in term of climate change factors. These factors

eather generatorrban water managementptimization

have been developed using runoff responses corresponding to base and future scenario of climate. Futurescenarios have been simulated using stochastic weather generator (LARS-WG) for different IPCC climatechange scenarios i.e. A1B, A2 and B1. Further, application of model has been demonstrated for a realisticwater supply system of Ajmer urban fringe (India). Developed model is capable in developing adaptationstrategies for optimum water resources planning and utilization in urban areas under different climate

change scenarios.

. Introduction

Various forms of single or multipurpose water management aren practice. However a comprehensive approach to water manage-

ent, referred as integrated water management, is still relativelyncommon, particularly in urban areas of developing countries

ike India (Jat, 2007). It can be defined as a process that promoteshe coordinated development and management of water, land andelated resources, in order to maximize the resulting economic andocial welfare in an equitable manner without compromising theustainability of vital ecosystems (Wolf and Hötzl, 2006). The wateremand and climate change always have the risk of uncertainty

nvolved in the future projections cautioning water managers torepare supply-demand strategies to face future crisis of waterWaterSmart, 2006). The emphasis should be given to identify andevelop the water management strategies, which lead to sustain-ble water resources and foolproof measures to thwart adverse

ffects of climate change. Such management practices are furtherecessary in a scenario of climate change due to uncertain precip-

tation and water availability. Therefore, there is a need for water

∗ Corresponding author. Tel.: +91 7409627187.E-mail address: pingalesm@gmail.com (S.M. Pingale).

1 Currently at Department of Water Resources and Irrigation Engineering, Arbainch University, Ethiopia.

921-3449/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.resconrec.2013.10.006

© 2013 Elsevier B.V. All rights reserved.

sector policy makers and professionals to understand the dynamicurban environment for solving water resources related problemsin urban areas, which will further aggravate in scenarios of climatechange.

The concept of integrated water management considering cli-mate change has not been well discussed and reported in literature(Asano, 1994; Kulga and Cakmak, 1997; Yang et al., 1999; Bouwer,2002; Haddad, 2002; Liu et al., 2007; Lin et al., 2010). Further,due attention has not been given to such practices in developingcountries like India (Zuidema, 1982; Tjallingii, 1988; Witter andBogardi, 1993; Lund and Cabrera, 2002; White and Fane, 2005;Jat, 2007; Qin and Xu, 2011). Few attempts have been made tomodel the climate change for urbanized areas (Wigmosta andBurges, 1990; Solecki and Oliveri, 2004; Wurbs et al., 2005; Stracket al., 2008) without considering climate change impacts on watermanagement practices. A few attempts have also been made toaddress the water resources management issues considering oneor another issue of climate change (Taleb and Maher, 2000; Ragaband Prudhomme, 2002; Mitchell et al., 2007; Qin et al., 2008; Shaoet al., 2011). However, integrated water management consideringintegration of various possible water sources to satisfy the demandsof different users, environment protection, land and urban planning

have not been considered.

The integrated approach to the water resources planning andmanagement requires a comprehensive consideration of waterrequirements and characteristics (hydrological and hydraulic) of

rvatio

dirBT2saebtrpdlhfwtud

go1aVBoHNasClmwiwis(iamSbtt

smsAmeWsw(tta((p

S.M. Pingale et al. / Resources, Conse

ifferent available sources (such as availability, cost and reliabil-ty). There are a number of management models which have beeneported in literature for rural areas and canal commands (e.g.,elaineh et al., 1999; Barlow et al., 2003; Karamouz et al., 2004;ripathi et al., 2004; Pulido-Velazquez et al., 2006; Khare et al.,006; Jha et al., 2008; Nikam, 2012), but a very few models areuitable for the urban areas (Dracup, 1966; Onta et al., 1993; Syauktnd Fox, 2004; Jat, 2007; Qin and Xu, 2011). The management mod-ls and strategies developed for rural and canal commands cannote applied directly to the urban areas. The water resources sys-em of urban areas is quite different and complex as compared toural areas because of the dynamic nature of various hydrologicalrocesses, different type of conflicting water demands and theirynamic nature, different environmental aspects, like water pol-

ution and wastewater generation and its consequences, and moreuman interference. The dynamic complexity in space and/or time

ound in urban water systems presents great difficulties for urbanater management (Senge, 1994). It becomes a complex system

han on the management of a few isolated issues. We are generallynable to relate causes with effects that are removed by time oristance (Zarghami and Akbariyeh, 2012).

The optimum allocation of water from surface sources androundwater has been attempted using different types of theptimization techniques, like dynamic programming (e.g., Buras,963; Aron and Scott, 1971), linear programming (e.g., Nieswandnd Grandstrom, 1971; Lakshminarayana and Rajagopalan, 1977;edula, 1985), simulation based models (e.g., Young andredehoeft, 1972; Bredehoeft, 1983; O’mara, 1984), multilevelptimization technique (e.g., Maddock and Haimes, 1975; Yu andaimes, 1974; Morel-Seytoux and Daly, 1975; Sharma, 1987) andon-linear programming (e.g., Kashyap and Chandra, 1982; Lefkoffnd Gorelick, 1990; Mishra et al., 2005). Nishikawa (1998) pre-ented an optimization model for the optimal management ofity of Santa Barbra’s water resources during the drought. Prob-

em is formulated as a linear programming problem. Optimizationodel is linked to MODFLOW programme to couple the ground-ater system with the management model. The objective function

s formulated to minimize the cost of water supply, subjected toater demand and hydraulic head constraint to control seawater

ntrusion and depletion of ground water. Jenkins et al. (2004) pre-ented the results of an economic-engineering optimization modelCALVIN) of California’s water supply system. This model explic-tly integrates the operation of various water facilities, sourcesnd demands of different users. Model allocates water to maxi-ize the economic value of statewide agricultural and urban uses.

hao et al. (2011) have developed a conditional value-at-risk (CVaR)ased inexact two-stage stochastic programming (CITSP) modelo a water resources allocation problem involving a reservoir andhree competing water users under uncertainty.

Wang and Huang (2011) formulated an interactive two stagetochastic fuzzy programming approach for the water resourcesanagement. The developed approach was applied for the case

tudy to demonstrate the water resources allocation problem. set of solutions under different feasibility degrees were esti-ated to plan the water resources allocation based on economic

fficiency, degree of satisfaction and risk of constraint violation.ang and Huang (2012) also developed an interactive multi-stage

tochastic fuzzy programming approach for identifying optimalater resources allocation strategies. Zarghami and Akbariyeh

2012) developed a model for the Tabriz’s urban water sys-em using a system dynamic approach. This model consideredhe water supply resources (groundwater, imported fresh water

nd treated wastewater), sources of demand for water resourcesdomestic, irrigation and industry uses) and management toolswastewater reuse and recycling, inter-basin water transfer, waterrice and conservation tools). Zarghami and Hajykazemian (2013)

n and Recycling 83 (2014) 176– 189 177

developed a new optimization algorithm for urban water resourcesplanning called as particle swarm optimization with mutationsimilarity (PSOMS). The application of PSOMS was successfullydemonstrated to an urban water problem for Tabriz city of Iran. Theproblem was formulated with an objective function to minimizethe cost, maximize water supply and minimize the environmen-tal hazards. The pipelines capacity, ground water, the demand andthe impact of conservation tools were considered as constraints.Wang and Huang (2013) applied an interval-parameter two stagestochastic fuzzy programming with type-2 membership functions(ITSFP–T2MF) approach for the water resources allocation prob-lem under uncertainty. However, in these studies, climate changeand resulting impact on water resources allocation have not beenconsidered. In these studies individual aspects of integrated watermanagement, within an optimization framework have been consid-ered. Therefore, an attempt has been made to develop an integratedurban water management model suitable in developing adaptationmeasures in optimum integration of various sources of water forurban water supply systems in climate change scenarios.

2. Study area and data used

Ajmer urban fringe is located between 26◦20′ to 26◦35′ N lati-tude and 74◦33′ to 74◦45′ E longitudes (Fig. 1). It spreads over anarea of about 85 km2 and has population of 542,580 (Census 2011).In the North-East side, water flow towards Samber lake. Ajmervalley drains eastward and Pushkar valley drains westward by trib-utaries of Luni river. There is a large lake i.e., Anasagar in the northof city.

For demonstration of model, data related to water supply systemof the Ajmer fringe like land use/cover, population, water demandsand existing water supply from various sources and other relevantdata have been collected from various departments and used in thepresent study (RUIDP, 1998; CPHEEO, 1999; Ajmera, 2000; Censusof India, 2001; PHED, 2004; Jat, 2007). The gridded rainfall datasetswere used in many hydrological and climatological studies world-wide, including Australia for hydroclimatic forecasting, climateattribution studies and climate model performance assessments(Tozer et al., 2012). The 0.5◦ × 0.5◦ gridded data set of daily rainfallfor Ajmer fringe from year 1971 to 2005 has been used which isprocured from India Meteorological Department, Pune. The Cana-dian Global Climate Model (CGCM3.1/T47) daily output data wereobtained from Canadian Centre for Climate Modelling and Anal-ysis (CCCma) for the grids covering Ajmer fringe of base period(1971–1990) and 20s (2011–2030) for A1B, A2 and B1 emission sce-narios. It uses a 360 days per year and has a spatial grid resolution2.8◦ × 2.8◦ latitude and longitude (McFarlane et al., 1992).

3. Methodology

The Integrated Urban Water Management Model consideringClimate Change (IUWMCC) is a mathematical programming model,which is formulated for the optimum allocation of water fromthe various water supply sources to satisfy the water require-ments of different users along with considering various system andgeometric constraints. The integrated water management modeldeveloped by Jat (2007), has been modified and improved in thepresent study to include climate change effects. The model appli-cation has been demonstrated further for the actual water supplysystem of Ajmer urban fringe.

In the present study, climate change impacts on surface and

groundwater sources have been incorporated in the model in termof variation in water availability from these sources as a result ofchange in rainfall and temperature due to climate change (climatechange factors). The stochastic weather generator (i.e. LARS-WG)

178 S.M. Pingale et al. / Resources, Conservation and Recycling 83 (2014) 176– 189

urban fringe in the Rajasthan State.

hi(a

3

omocplcn

3

q

Sou rces o f impo rted water

SSi wit h annual capacit y Qi,i = 1……..I

Ground water source GSjwith annual capacit y Qrj = 1 J

Loca l surface sources LSkwith annual capacit y Qk, k= 1…… ..K

Re-use o f wastewater from

sou rce WSl with ann ual

capacit y Ql,l = 1 L

Ass ociated cost

CGj t

Ass ociated cost

CLSk t

Ass ociated cost

CWl t

Ass ociated cost

CSi t

CCFj

CCFk

CCFiWater demand of

different u ser wi th

mon thly d emand,

dn,t,n = 1………N

Mathemati cal

Programming

Formulation

Dec ision Variables

Quantity o f water to be

used fr om d ifferent

sou rces

Fig. 1. The location of Ajmer

as been used to generate future climate change scenario and cal-brated rainfall runoff model (SWMM) and groundwater modelMODFLOW) has been used to generate future scenarios of surfacend groundwater.

.1. Model formulation

The formulation of the IUWM model includes definition ofbjective function and constraints of different components. Theodel was formulated to incorporate the climate change effects

n water availability from different sources in the form of climatehange factors. Formulated model is a non-linear mathematicalrogramming problem, which has been solved using successive

inearization technique. Non-linearity of the groundwater pumpingost has been linearized through successive linearization tech-ique. The framework of the model has been presented in Fig. 2.

.2. Objective function

The objective function has been formulated to determine theuantity of water supplied from various sources during different

i, j, k, l, are type of sources, t = 1…T, number of planning periods, n = 1…N, are number of users CCF…climate change factor

Fig. 2. General conceptual framework of IUWMCC model.

rvatio

pi

M

Sk,t

⎞⎟⎟⎟⎟⎠

wigqpptcd(ios(tttm

3

tc

3

tst

∑

s

∑

wj

(

∑

w

S.M. Pingale et al. / Resources, Conse

lanning periods such that overall cost of the system is minimizedn different climate change scenarios.

inimise, Z =

⎛⎜⎜⎜⎜⎝

I∑i=1

T∑t=1

qi,tCSi,t +J∑

j=1

T∑t=1

qj,tCGj,t +K∑

k=1

T∑t=1

qk,tCL

+L∑

l=1

T∑t=1

ql,tCWl,t

here qi,t is quantity of water transferred from imported source in time period t (m3/day); qi,t is quantity of water pumped fromroundwater potential zone j in time period t (m3/day); qk,t isuantity of water supplied from local surface source k in timeeriod t (m3/day); and ql,t is quantity of treated wastewater sup-lied from wastewater source/treatment facility l in time period

(m3/day). The CSi,t,CGj,t, CLSk,t and CWl,t represents the unit costoefficients of imported water from source (i), groundwater with-rawal from zone (j), water supply from local surface water sourcesk) and reuse of treated wastewater from source/treatment facil-ty for time period (t) in ( /m3), respectively. The total numberf imported sources, groundwater potential zones, local surfaceources and reuse of treated wastewater sources in time periodT) are represented by I, J, K and L, respectively. The solution ofhe water management/optimum allocation problem requires thathe decision variables qi,t, qj,t, qk,t and ql,t be determined such thathe following constraints are satisfied and the objective function is

inimized.

.3. Constraints

Constraints are physical, geometrical and operational limita-ions imposed on the model to represent the actual operationalharacteristics of a water resources system.

.4. Water availability constraints

Eqs. (2)–(5) are the water availability constraints for each poten-ial water source in each time period under different climate changecenario. These constraints represent the upper bound on the quan-ity of water, which may be tapped from each source.

Imported source

T

t=1

qi,t ≤ CCFi ∗ Qi ∀i (2)

where CCFi = climate change factor for imported water supplyource, i

Qi = annual capacity of imported source, i (Mm3)Groundwater

J

j=1

12∑t=1

qj,t ≤ CCFj ∗ (1 − �) ∗ Qr (3)

here CCFj = climate change factor for ground water supply source,

Qr = annual quantity of groundwater available from all zones, rMm3)

� = mining allowance for groundwater pumping.Local surface source

T

t=1

qk,t ≤ CCFk ∗ Qk ∀k (4)

here CCFk = climate change factor for local surface sources, kQk = annual capacity of local surface source, k (Mm3).

n and Recycling 83 (2014) 176– 189 179

(1)

Treated wastewater

T∑t=1

ql,t ≤ Ql ∀l (5)

where Ql = annual capacity of reusable treated wastewater availablesource/facility, l (Mm3)

3.5. System capacity constraints

System capacity constraint (Eqs. (6)–(9)) for each source isthe limiting capacity of any system component, like capacity oftransmission system, treatment facility or storage facility. This con-straint is used to ensure that quantity of water allocated from anysource during any time period should not be more than the mini-mum of capacity of any system component.

Imported source

qi,t ≤ Qi,t ∀i, t (6)

where Qi,t = maximum quantity of water available from importedsource i, in time period t (Mm3/year).

Groundwater pumping

qj,t ≤ Qj,t ∀j, t (7)

where Qj,t = monthly pumping capacity of zone j, in time period t(Mm3/year).

Local surface source

qk,t ≤ Qk,t ∀k, t (8)

where Qk,t = maximum quantity of water available from the localsurface source k, in time period t (Mm3/year).

3.5.1. Treated wastewaterMinimum of monthly capacity of any component of sewer-

age system, like capacity of wastewater treatment plant can beconsidered as the system capacity constraint for the re-usablewastewater.

ql,t ≤ Ql,t ∀l, t (9)

where Ql,t = maximum quantity of treated wastewater availablefrom waste water source/facility l, in time period t (Mm3/year).

3.6. Water balance constraints

Water balance should be maintained within the water supplyand sewerage system. Quantity of treated wastewater availableduring a particular time period t should be equal to sum of quantityof water supplied from various sources minus consumptive use and

1 rvatio

lo⎛⎜⎜⎜⎜⎝

wc �

witekwsisftwr

3

weth⎛⎜⎜⎜⎜⎝

n,t∀

wtt

q

3

sbad

3

iR(iqesair

80 S.M. Pingale et al. / Resources, Conse

osses at different stages. This constraint represents the continuityf flow within the system.

I∑i=1

qi,t�ci�ti�d +J∑

j=1

qj,t�cj�tj�d +K∑

k=1

qk,t�ck�tk�d

+L∑

l=1

ql,t�cl�tl�d

⎞⎟⎟⎟⎟⎠

∗ (1 − ˛)�

here �ci is efficiency of the water transmission system formported source i, �cj is efficiency of the water transmission sys-em for groundwater pumped from zone j (if applicable), �ck isfficiency of the water transmission system for local surface source, �cl is efficiency of the water transmission system for treatedastewater sources (if applicable). �ti, �tj, �tk, �tl are repre-

ents the efficiency of water treatment plant for imported source, groundwater pumped from zone j (if applicable), local surfaceource k and additional treatment to the recyclable wastewaterrom source/facility l, respectively. �d, �wc and �tw are efficiency ofhe water distribution system, sewerage system for treated waste-ater and wastewater treatment plants, respectively. ̨ is a fraction

epresents the consumptive use of water by the users.

.7. Water requirement constraints

Water requirement constraint ensures that the quantity ofater supplied from all the sources in any time period must be

qual to the total quantity of water required by all the users for sameime period. In case of single water supply system, total demandsave been assumed as potable.

I∑i=1

qi,t�ct�ti�d +J∑

j=1

qj,t�cj�tj�d +K∑

k=1

qk,t�ck�tk�d

+L∑

l=1

ql,t�cl�tl�d

⎞⎟⎟⎟⎟⎠

≥N∑

n=1

dn,t�

here �n,t is rationing factor for potable water demand of user n inime period t, and dn,t is demand of potable water by the user n inime period t (Mm3/year).

Non-negativity constraints

i,t≥0 ∀i&t; qj,t≥0 ∀j&t; qk,t≥0 ∀k&t; ql,t≥0 ∀l&t (12)

.8. Application of model

Application of developed mathematical model has been demon-trated for the water supply system of Ajmer urban fringe. Theasic characteristics of the water supply system of the study areand models used for parametrisation of different process have beeniscussed below.

.9. Sources of water

Presently, water is supplied to Ajmer urban fringe through anntegrated distribution system which is imported from Bisalpureservoir (Tonk, Rajasthan). Local groundwater and surface sourcesi.e. Anasagar lake, Foysagar lake and Khanpura pond) have beendentified as the other potential sources of water supply. Annualuantity of water available from these sources is required to be

stimated for future water resources planning. In the presenttudy, quantity of water available from local surface sourcesnd groundwater have been estimated using calibrated empir-cal rainfall-runoff and groundwater models of the Ajmer area,espectively (Jat, 2007). Ajmer urban fringe has water rights for

n and Recycling 83 (2014) 176– 189

tw≥L∑

l=1

ql,t∀t (10)

t (11)

31.025 Mm3/year of water from 1st phase of the Bisalpur project(RUIDP, 1998; PHED, 2004).

Quantity of water available from local surface sources in futurewithout considering the climate change scenario has been esti-mated for Ajmer urban fringe corresponding to normal seasonalrainfall (monsoon) of 451.9 mm. The normal climate needs tobe defined because climate correction involves in estimating theamount of water use, if weather conditions are normal. Nor-mal climate is usually defined as the 30 year average of climaticparameters (Maheepala and Roberts, 2006). However, the conceptof normal climate will be uncertain in climate change scenariobecause the average of climate over a period differs from decade todecade (Perera et al., 2009). Therefore, the future quantity of wateravailable from these sources in climate change scenario has beenestimated based on climatic parameters simulated using CGCMoutputs. For each local surface source, average quantity of surfacerunoff available at the end of September and December monthshas been considered as the quantity of water available for use andconsidered as the source capacity constraint in the optimizationmodel.

Quantity of water available from groundwater has been esti-mated using calibrated groundwater model, where recharge isestimated corresponding to the normal rainfall in monsoon(June–November) and non-monsoon (November–May) season.Annual capacity of groundwater has been adopted equal to theannual recharge, which is estimated as 3.956 Mm3 for year 2005and 3.957 Mm3 for other years (corresponding to normal rain-fall). Total groundwater recharge comprises of rainfall recharge andrecharge from the lakes. Groundwater recharges from lakes andponds have been considered equal to recharge in year 2005, havingrainfall equal to the normal rainfall. Quantity of available treatedwastewater has been estimated from the data obtained from RUIDP,Ajmer (RUIDP, 1998; PHED, 2004). Quantity of treated wastewaterhas been estimated assuming 20% consumptive use of water outof total water supplied. Annual capacities of different sources havebeen used as the water availability constraints in the optimizationmodel.

3.10. Development of climate change factors

Climate change factors (CCF) have been incorporated in thesource constraint of all potential sources except treated waste-

water. The results of resources allocation under climate changescenario have been compared with the results of resources alloca-tion without considering climate change. The CCF for the importedand groundwater source have been adopted similar to local surface

S.M. Pingale et al. / Resources, Conservation and Recycling 83 (2014) 176– 189 181

Table 1Runoff scenarios generated for Anasagar using CGCM output.

Year Population ofAjmer (million)

Imperviousarea (%)

Runoff generated (mm)

Baseline A1B A2 B1

2006 0.531 11.723 97.65 63.31 56.24 56.922007 0.539 12.269 99.22 64.88 57.81 58.492008 0.548 12.826 100.82 66.47 59.41 60.092009 0.556 13.393 102.44 68.10 61.03 61.722010 0.564 13.97 104.10 69.76 62.69 63.372011 0.572 14.558 105.79 71.45 64.38 65.062012 0.583 15.300 107.92 73.58 66.51 67.192013 0.593 16.058 110.09 75.75 68.68 69.362014 0.603 16.831 112.31 77.97 70.90 71.582015 0.613 17.621 114.58 80.24 73.17 73.852016 0.623 18.426 116.89 82.55 75.48 76.162017 0.634 19.247 119.24 84.90 77.84 78.522018 0.644 20.085 121.65 87.31 80.24 80.92

scbbifdsTiAote(aw

R

R

%gmu

I

wrtel

A

A

w

bChK

each groundwater potential zones (GPZ) has been adopted as thesystem capacity constraint (Eq. (7)), which remains same for allthe months in a year. In the present study, four GPZ have beenconsidered suitable for use i.e. Anasagar, Khanpura, Durai and

Table 2The climate change factor for local surface sources, imported water supply andground source in A1B scenario (Case 2 and 3).

Year CCFi CCFj CCFk (A) CCFk (KH) CCFk (F)

2006 0.62 0.62 0.65 0.70 0.522007 0.63 0.63 0.65 0.70 0.522008 0.63 0.63 0.66 0.71 0.522009 0.63 0.63 0.66 0.71 0.522010 0.63 0.63 0.67 0.71 0.522011 0.64 0.64 0.68 0.71 0.522012 0.64 0.64 0.68 0.72 0.522013 0.64 0.64 0.69 0.72 0.522014 0.65 0.65 0.69 0.72 0.522015 0.65 0.65 0.70 0.73 0.522016 0.65 0.65 0.71 0.73 0.522017 0.66 0.66 0.71 0.73 0.522018 0.66 0.66 0.72 0.74 0.522019 0.66 0.66 0.72 0.74 0.52

2019 0.654 20.938

2020 0.664 21.808

2021 0.674 22.693

ource i.e. Anasagar assuming that they will be affected by thelimate change similarly. The CCF has been estimated based onaseline and future runoff scenarios generated corresponding toase line and future climate scenarios. The future climate scenar-

os have been developed using LARS-WG based on CGCM outputsor the Ajmer urban fringe. In LARS-WG, the calibration and vali-ation for the observed climate has been performed in threeteps: (1) calibration, (2) QTest (Validation) and (3) Generation.he model is calibrated (1971–1990 for the base line) and val-dated for the observed climate to obtain future projections forjmer urban fringe. The model performance is validated basedn statistical characteristics (mean and standard deviation) andests (Kolmogorov–Smirnov test, F-test and T-test). The empiricalquations were used, which was verified by hydrological modelSWMM) and observed data to estimate the surface water avail-bility. Runoff from the urbanized and non-urbanized catchmentsere estimated from the equations as given below (Jat, 2007):

O = 0.367S + 2.87Ai + 10269I − 31.078 (13)

O = 0.197S + 3.065I − 28.53 (14)

where Ai and S are the impervious area and topographic slope in, respectively, I is the rainfall intensity factor and RO is the runoffenerated in mm. The rainfall intensity factor is derived using theonsoon rainfall depth (R) in mm and number of rainy days (NR)

sing the following equation (Jat, 2007):

=(

R

NR˛

)(15)

here constant ̨ depends upon the geographical region for whichelationship is sought. For Ajmer area, this factor has been foundo be 0.5 (Jat, 2007). The percentage impervious area (Ai) has beenstimated for Anasagar and Khanpura sub-watershed using the fol-owing developed equations (Jat, 2007), respectively:

i = 63.94P2.2 − 4.18 (16)

i = 48.08P1.1 − 6.26 (17)

here P is population in Million.The rainfall and number of rainy days during the monsoon have

een simulated for A1B, A2 and B1 scenarios of 20th century fromGCM output. Using the above Eqs. (13)–(17), the runoff scenariosave been generated for the base and future scenarios at Anasagar,hanpura and Foysagar lakes/pond (e.g. Table 1). The CCF for these

124.10 89.76 82.69 83.37126.59 92.25 85.19 85.87129.13 94.79 87.73 88.41

lakes or ponds has been generated using the baseline and futurerunoff scenarios upto year 2020–2021 (e.g. Table 2).

4. System capacities

For water supply system of Ajmer, system capacities includegeometric limitation of the different components. These capacitieshave been taken from previous studies (Jat, 2007) (Table 3).

Imported source: For imported supplies from Bisalpur, monthlycapacity of transmission system has been adopted as system capac-ity constraint (Eq. (6)) and considered as constant for all monthsof a year. Capacity of transmission system of Ajmer has beenadopted equal to the average of water supplied (69,460 m3/day) inlast five years, which is found to be between 65,430 m3/day and74,780 m3/day. This is less than the designed capacity of trans-mission system (85,000 m3/day), which has been reduced due towear and tear. Thus, the average quantity of water supplied in year2005–2006 has been assumed as the system capacity constraint forthe imported source.

Groundwater: For groundwater, monthly pumping capacity of

2020 0.66 0.66 0.73 0.74 0.522021 0.67 0.67 0.73 0.75 0.52

Note: A, F and KH represents Anasagar, Foysagar and Khanpura lakes/pond, respec-tively

182 S.M. Pingale et al. / Resources, Conservatio

Tab

le

3M

onth

ly

cap

acit

y

con

stra

ints

of

dif

fere

nt

sou

rces

for

dif

fere

nt

year

s

(in

1000

m3).

Mon

th

Jun

e

July

Au

gust

Sep

tem

ber

Oct

ober

Nov

embe

r

Dec

embe

r

Jan

uar

y

Febr

uar

y

Mar

ch

Ap

ril

May

Imp

ort

for

2005

–21

(Bis

alp

ur)

2083

.8

2153

.3

2153

.3

2083

.8

2153

.3

2083

.8

2153

.3

2153

.3

1944

.9

2153

.3

2083

.8

2153

.3A

nas

agar

for

cap

acit

y

in

2005

a22

4.4

231.

9

231.

9

224.

4

231.

9

224.

4

231.

9

231.

9

209.

4

231.

9

224.

4

231.

9A

nas

agar

for

2005

–21

(wit

h

urb

aniz

atio

n

effe

ct)b

448.

8

463.

8

463.

8

448.

8

463.

8

448.

8

463.

8

463.

8

418.

9

463.

8 44

8.8

463.

8Fo

ysag

ar

for

2005

–21

108.

0

111.

6

111.

6

108.

0

111.

6

108.

0

111.

6

111.

6

100.

8

111.

6 10

8.0

111.

6K

han

pu

ra

for

2005

–21

(wit

hou

t

urb

aniz

atio

n

effe

ct)a

448.

8

463.

8

463.

8

448.

8

463.

8

448.

8

463.

8

463.

8

418.

9

463.

8

448.

8

463.

8K

han

pu

ra

up

to

2011

(wit

h

urb

aniz

atio

n

effe

ct)c

448.

8

463.

8

463.

8

448.

8

463.

8

448.

8

463.

8

463.

8

418.

9

463.

8

448.

8

463.

8K

han

pu

ra

from

2012

–21

(wit

h

urb

aniz

atio

n

effe

ct)c

516.

053

3.2

533.

251

6.0

533.

2

516.

0

533.

2

533.

2

481.

6

533.

2

516.

0

533.

2Tr

eate

d

WW

(up

to

2011

)

1500

.0

1550

.0

1550

.0

1500

.0

1550

.0

1500

.0

1550

.0

1550

.0

1400

.0

1550

.0

1500

.0

1550

.0Tr

eate

d

WW

for

2012

–201

5

2100

.0

2170

.0

2170

.0

2100

.0

2170

.0

2100

.0

2170

.0

2170

.0

1960

.0

2170

.0

2100

.0

2170

.0Tr

eate

d

WW

for

2016

–202

1

2400

.0

2480

.0

2480

.0

2400

.0

2480

.0

2400

.0

2480

.0

2480

.0

2240

.0

2480

.0

2400

.0

2480

.0

aC

apac

ity

of

trea

tmen

t

pla

nt,

if

wat

er

avai

labi

lity

assu

med

at

the

leve

l of y

ear

2005

.b

Cap

acit

y

of

trea

tmen

t

pla

nt

pro

pos

ed, i

f eff

ect

of

urb

aniz

atio

n

is

con

sid

ered

wh

ile

esti

mat

ing

the

quan

tity

of

wat

er

avai

labl

e

from

An

asag

ar

lake

.c

Cap

acit

y

of

trea

tmen

t

pla

nt

pro

pos

ed

corr

esp

ond

ing

to

the

wat

er

avai

labl

e,

con

sid

erin

g

the

effe

ct

of

urb

aniz

atio

n, f

rom

Kh

anp

ura

pon

d. W

ater

avai

labi

lity

from

imp

ort

is

the

aver

age

cap

acit

y

of

tran

smis

sion

syst

em. W

W

isw

aste

wat

er

and

TWW

is

trea

ted

was

tew

ater

.

n and Recycling 83 (2014) 176– 189

Parvatpura, as other two zones are found unsuitable because oftheir low potential.

Anasagar: Monthly capacity of water treatment plant has beenadopted as the system capacity constraint for Anasagar (Eq. (8)),and considered as constant for all the planning periods of a year.The capacity of transmission line for the Anasagar have been cal-culated based on 15 h of pumping and safe capacity of lake in year2005. Two pipe lines of 7.48 MLD (million litres per day) capacities(each) have been proposed with all necessary pumping capabilities.So, total capacity of transmission system is 5.46 Mm3/year. Further,its capacity can be enhanced by increasing the pumping hours. Toaccommodate the effect of urbanization, capacity of water treat-ment plant for Anasagar has been assumed equal to the 14.96 MLD,as compared to 7.48 MLD considered suitable for the year 2005(without considering the effect of urbanization). Unit cost of watersupply with this capacity would have been 2.81 /m3 (at pricesof year 2005) as compared to 3.13 /m3 for 7.48 MLD capacity oftreatment plant.

Foysagar: For Foysagar, monthly capacity of water treatmentplant (3.6 MLD) (already available) has been adopted as the sys-tem capacity constraint, which remains same for all the planningperiods (month) of a year.

Khanpura: Monthly capacity of water treatment plant has beenadopted as the system capacity constraint for Khanpura, whichremains constant for all the months of a year. Presently, watersupply infrastructure is not available at Khanpura pond. Suitablewater supply infrastructure has been proposed similar to Anasagar.Water supply infrastructure has been proposed for two capacities;(i) 5.46 Mm3 annually up to service level of year 2011, and (ii)6.278 Mm3 up to service level of year 2021. It has been assumedthat water will be directly fed to the distribution system from thetreatment plant.

Treated wastewater: RUIDP, Ajmer has designed and executedthe sewerage system for Ajmer urban fringe. A wastewatertreatment plant (both primary and secondary) of 50,000 m3/daycapacity was designed for the service level up to year 2011. RUIDP,in principal, has decided to further augment the sewerage sys-tem of Ajmer by 30 MLD in two phases for a service level upto year 2021. Capacity of sewerage system increased by 20 MLDby the year 2011, and further it would be increased by 10 MLDby the year 2015. Hence, quantity of sewage available after year2011 was 70 MLD and by the end of 2015, would be 80 MLD.Monthly capacity has been assumed to be constant. These capac-ities have been adopted as the system capacity constraints in themodel.

4.1. Efficiency of different system components

Water losses in water supply system were considered andadopted as the hydraulic efficiency of different system compo-nents in the model. The information of losses was obtained fromliterature (RUIDP, 1998; Ajmera, 2000; PHED, 2004; Jat, 2007).Transmission losses for different sources were reported between0.5% and 2% (Ajmera, 2000). For model demonstration, these losseswere considered as 0.5% (local surface sources), 1% (groundwater)and 2% (imported water). Losses of water in treatment plants werereported between 0.5% and 1% (PHED, 2004). These were adoptedas 1% in this study. Losses of wastewater in sewerage system wereadopted as 10% (RUIDP, 1998). Presently, losses of water in dis-tribution system have been found about 30% to 40% of the water

supply (Ajmera, 2000; PHED, 2004). In this study, efficiency of thedistribution system was assumed as 70%. Total demand of waterfrom all sectors has been considered at the entrance of distributionsystem.

rvatio

4

htosbtycaTf

4

mUAdc

4

wetbp

4

a

(

(

(

(

4

dmws

ptNt1aT

S.M. Pingale et al. / Resources, Conse

.2. Cost of water

In the present study, unit cost of water supply from any sourceas been expressed in term of /1000 m3. These costs representhe unit cost of delivery of water from each source to starting pointf distribution system. Unit cost of water supply for the importedource, Anasagar, Foysagar, Khanpura and treated wastewater haseen adopted as Rs. 1150, 3130, 2860, 2810 and 15,920 m–3, respec-ively for the base year (2005) as obtained from RUIDP. For otherears, cost has been estimated considering 8% rate of inflation. Unitost of water supply includes cost incurred on source, transmissionnd treatment from different sources has been adopted (Jat, 2007).hese have been adopted as the cost coefficients in the objectiveunction (Eq. (1)).

.3. Water demands

For the present study, annual water demand of various users i.e.unicipal, industrial, fire, railways and floating water demands ofrsh has been obtained from RUIDP and PHED, Ajmer (RUIDP 1998;jmera, 2000; PHED, 2004; Jat, 2007). Average monthly wateremands have been considered, as the monthly water demandonstraints (Eq. (13)).

.4. Re-use of treated wastewater

Re-use of treated wastewater has been considered as a potentialater demand management option. It has been divided in two cat-

gories: (i) potable and (ii) non-potable. In the present study onlyhe first case have been explored that all the water demands haveeen considered as potable, and treated wastewater is used as aotable water source after proper treatment.

.5. Model assumptions

For demonstrating the application of model the followingssumptions have been made:

a) The time horizon has been adopted as 15 years (2005–2006 to2020–2021). Each planning year has been assumed to start from1st June,

b) The inflation rate has been adopted as 8% to estimate the futurecost of water supply from different sources,

(c) Whole Ajmer urban fringe has been considered as the singlewater supply zone and water available from different sourcesis integrated with the same distribution system,

d) Quality of groundwater has been assumed as potable, and about100% of groundwater recharge is available for the use,

e) Model has been formulated to account for the monthly andyearly variation of water available in different year.

.6. Solution of the model

In the present study, LINGO (version 10), a comprehensive toolesigned for building and solving linear, nonlinear and integer opti-ization models, has been used for the solving the formulatedater management models under climate change scenario. This

oftware is developed by the LINDO Systems Inc. (Chicago, USA).Formulated model is a non-linear mathematical programming

roblem, which has been solved using successive linearizationechnique using Successive Linear Programming (SLP) algorithm.on-linearity of the groundwater pumping cost has been linearized

hrough successive linearization technique. The LINGO (version0) also has Global and Multi-start solvers, which are capable ofchieving global solution of non-convex and non-linear problem.he Global solver converts the original non-convex and nonlinear

n and Recycling 83 (2014) 176– 189 183

problem into several convex and linear sub-problems. It then usesthe branch-and-bound technique to exhaustively search over thesesub-problems for the global solution.

5. Results and discussion

The application of IUWMCC has been demonstrated for findingoptimum allocation of water from multiple water supply sources tosatisfy the water requirements of various users in uncertain futureclimate scenarios. Impact of climate change has been incorporatedin terms of change in water availability from various sources due tochange in precipitation and runoff under different climate changescenarios. Results of different management strategies in differ-ent scenarios have been discussed below. Optimal managementstrategies in changing climate are suggested through comparativeanalysis of all the strategies for Ajmer urban fringe.

5.1. Management scenarios investigated

The following management strategies have been investigated:

(1) Present water supply condition (Case 1), Integration ofgroundwater, local surface sources with present water supplycondition (Case 2), and

(2) Integration of groundwater, local surface sources with urban-ization effect and treated wastewater for potable uses (Case3).

5.2. Model solution

The required data and various model assumptions have beenused while solving the model for water resources allocations indifferent management strategies under climate change scenarios.For each scenario, model solution has been obtained using LINGOcomputer package (version 10). Summary of the various scenariosinvestigated and size of problem have been presented in Table 4.

The present water supply system and its sustainability with dif-ferent management strategies have been evaluated for the existingand future climate change scenarios i.e. A1B, A2 and B1. Ground-water is available in small quantity but its use is unaccounted(Jat, 2007) and not considered in Case 1 i.e. present condition.Model constraints have been modified to represent the presentwater supply system of Ajmer. For different years, water availabil-ity constraint (annual capacity) has been adopted as 31.025 Mm3.Monthly capacity constraint has been adopted as per the presentcapacity of water transmission system (25.352 Mm3 annually) fromBisalpur to Ajmer. For other sources i.e. groundwater, local surfacesources and treated wastewater capacity constraints have beenmodified accordingly. Losses have been represented in the formof efficiency of different system components, and adopted in themodel. Monthly water demands of various users in different yearshave been considered as the demand constraint. For different years,cost coefficients have been adopted.

In Case 2 and Case 3, the dynamic interaction of groundwatersystem and management model has been considered as incorpo-rated by Jat, 2007 through external coupling of groundwater model(MODFLOW) and optimization model. For the first run of the model(year 2005–2006), water table elevations of pre-monsoon period(i.e. at the end of May) of year 2005 have been considered as theinitial water table elevations. Groundwater mining coefficient hasbeen kept as zero, as no mining is allowed. The model has been run

for three strategies (Case 1–Case 3) of water supply for the entireanalysis period in A1B, A2 and B1 scenario, separately. Initially,rationing factor has been kept as zero in all three cases. However,model solution has not been found feasible at 0% rationing in Case

184 S.M. Pingale et al. / Resources, Conservation and Recycling 83 (2014) 176– 189

Table 4Size of problem for different management strategies investigated.

S. no. Scenario Number of sources No. of decisionvariables

No. ofconstraints

Type of problem

1. Present water supply condition One (Import) 12 775 Linear2. Integration of groundwater and

local surface sources with presentcondition

Six (Import, GW, Threelocal sources and TWW)

96 775 Linear, non-linearity of GW

3. Integration of import source, Six (Import, GW, Three)

108 895 Linear, non-linearity of GW

1lffahilcctfc

anwcct

lpasfswwTmvaAy

mbabapcitdtBbwfac(

change in climate as found from the simulated climate using CGCMoutput.

12000

17000

22000

27000

32000

37000

42000

47000

52000

57000

Quan

tity

(1

00

0 m

3)

Yea rCase 1 wit hout CC Case 1 wit h CC

groundwater and local surfacesources with urbanization effectand treated wastewater

local sources and TWW

and Case 2, as water demands are more than supplies, which vio-ates the demand constraint. Further, different values of rationingactor are tried to make the solution feasible. Value of rationingactor, at which solution becomes feasible, represents the percent-ge of overall shortage or demands unmet. In a similar way, modelas been run for different years and results are obtained, which

nclude monthly allocation of water, minimized cost, percentageosses and percentage of demand satisfied under different climatehange scenarios. The results of the model for various years underhanging climate scenarios have been determined. Results revealedhat throughout the analysis period, water demand of the urbanringe has not been met fully in all the three scenarios of climatehange in Case 1 and 2.

In Case 2, water supply system of Ajmer fringe is assumed to beugmented with local surface sources. Effect of urbanization hasot been considered in Case 2. The climate change effects on urbanater demand have not been considered, only effects have been

onsidered on water availability from different sources in changinglimate. Monthly variation of demands has not been considered inhe present study due to non availability of data.

In Case 3, integration of treated wastewater, groundwater andocal surface sources has been considered into present water sup-ly system. In this case, effect of urbanization (i.e. increase in watervailability from local surface sources with increase in imperviousurface) and treated wastewater as supply augmentation measuresor potable uses with proper additional treatment has been con-idered. It is assumed that the availability of quantity of treatedastewater will not be affected due to change in climate as treatedastewater is a function of water supplied from other sources.

reated wastewater has been considered as a demand manage-ent measures, however it has been considered as a separate

ariable in the model, being a major source of water with properdditional treatment. Unit cost of water supply for Khanpura andnasagar has been found to be same for both the capacities up toear 2011–2012.

In this Case 3, rationing factor has been found to be zero if cli-ate change effects have not been considered. Model solution has

een found to be feasible in this strategy as full water demandsre satisfied. However, model solution for A1B scenario has noteen found to be feasible in the beginning of years 2005–2011,s full water demands are not satisfied, which is more than sup-lies in changing scenarios of climate change, violating the demandonstraint. Value of rationing factor has been found to be zeron this case after the year 2011. Model solution has been foundo be feasible in A1B scenario after year 2010–2011 as full wateremands are satisfied because of proposed increased capacity ofreated wastewater. Similarly, model has been run for the A2 and1 scenarios, separately. In both these scenario, solution has noteen found feasible at 0% rationing, as it does not satisfies the fullater demand in changing climate. For sample, the model results

or A1B scenario of climate change are presented in Table 5. Waterllocations from different sources for different years in changinglimate scenario have been shown for A1B scenario as a sampleFig. 3).

5.3. Comparison of management strategies

5.3.1. Total quantity of water suppliedTotal quantity of water supplied from different sources

depends upon number of sources considered in that particu-lar management strategy. Total quantity of water supplied indifferent years in different management strategies has beeninvestigated with and without considering climate changeeffect.

Total quantity of water supplied in the Case 1 is limited up tomaximum quantity of water available from the imported sources,which is equal to 20.78 Mm3, 18.61 Mm3 and 18.92 Mm3, respec-tively for A1B, A2 and B1 scenarios in the year 2020–2021. Thisquantity is less than the water rights (31.025 Mm3) of Ajmer due tolimited capacity of transmission system. In Case 2, total quantity ofwater supplied from various sources has increased to 30.63 Mm3,27.34 Mm3 and 27.83 Mm3, respectively due to integration ofgroundwater and local surface sources into the present water sup-ply system in the year 2020–2021. Total quantity of water suppliedin Case 3 varies from 39.76 Mm3, 36.76 Mm3 and 37.18 Mm3 foryear 2005–2006 to 59.34 Mm3, 54.03 Mm3 and 54.98 Mm3 for year2020–2021 in A1B, A2 and B1 scenarios, respectively (e.g. Fig. 3for A1B scenario). Groundwater and local surface sources have alsobeen utilized fully in Case 2 and Case 3 in all three scenarios. Thesesources are utilized first as compared to imported supplies andtreated wastewater, being the cheapest sources. In Case 3, quantityof total water supplied in different years has increased as com-pared to Case 2 on account of effect of urbanization and integrationof treated wastewater for potable use. With an increase in popula-tion, quantity of water available from local surface sources wouldincrease due to increase in impervious areas (Jat, 2007) as resultof increase in built-up activities. However, total quantity of wateravailable from different sources would significantly reduce due to

Case 2 wit hout CC Case 2 wit h CCCase 3 wit hout CC Case 3 wit h CC

Fig. 3. Total quantity of water supplied from different sources in various casesconsidering without and with climate change in A1B scenario.

S.M.

Pingale et

al. /

Resources,

Conservation and

Recycling

83 (2014) 176– 189185

Table 5Integration of imported, groundwater and local surface sources with urbanization effect and reuse of treated wastewater in A1B scenario (Case 3).

Year 2010–2011 2011–2012 2012–2013 2013–2014 2014–2015 2015–2016 2016–2017 2017–2018 2018–2019 2020–2021

Water demands of different users (1000 m3)Domestic 30,890 31,340 31,897.3 32,460 33,013.6 33,570 34,130 34,688 35,246 36,360Industrial 406.9 419.7 433.1 446.5 459.9 468.6 480 486.18 494.9 534.7Railways 2299.5 2299.5 2299.5 2299.5 2299.5 2299.5 2299.5 2299.5 2299.5 2299.5Fire 881.8 890.6 900 910.06 919.8 923.4 930 930.7 934.4 978.91Ursh 446 456.2 466 475.9 485.8 489.1 490 495.6 498.9 538.3

Total 34924.2 35406.0 35995.9 36592.0 37178.6 37750.6 38329.5 38900.0 39473.7 40711.4Utilization of water from different sources (1000 m3)Import 19856 19856 19856 20166 20166 20166 20476 20477 20476 20787GW 2532.5 2532.5 2532.5 2572.1 2572.1 2572.1 2611.6 2611.6 2611.6 2651.2Anasagar 2308 2355 2450 2511 2611 2713 2780 2889 2959 3148Foysagar 799 811 811 811 822 822 822 833 833 845Khanpura 2765 2801 2845 2890 2934 2979 3023 3068 3113 3203TWW 18,245 19,037 24,009 24,416 25,112 25,547 26,180 26,834 27,549 28,707

Total 46505.5 47392.5 52503.5 53366.1 54217.1 54799.1 55892.6 56712.6 57541.6 59341.2Losses (%) 31.51 31.49 31.44 31.43 31.43 31.46 31.42 31.41 31.4 30.89Percentage of demand satisfiedDemand 91.2 91.7 100.0 100.0 100.0 99.5 100.0 100.0 100.0 100.0Unit cost of water supply ( /1000 m3)Import 16897.3 18249.1 19,709 21285.7 22988.6 24827.6 26813.8 28,959 31275.7 36479.9Anasagar 4128.8 4459.1 4815.8 5201.1 5617.2 6066.6 6551.9 7076.1 7642.1 8913.8Foysagar 4202.3 4538.5 4901.5 5293.7 5717.2 6174.5 6668.5 7202 7778.1 9072.4Khanpura 4128.8 4459.1 4815.8 5201.1 5617.2 6066.6 6551.9 7076.1 7642.1 8913.8Weighted annual average depth of water table below ground surface in different zones (m)Anasagar 38.98 38.91 38.85 38.77 38.69 38.60 38.53 38.44 38.37 38.14Khanpura 27.29 27.57 27.72 27.92 28.10 28.26 28.41 28.55 28.68 28.90Parvatpura 37.17 37.30 37.41 37.51 37.66 37.77 37.91 38.01 38.13 38.34Durai 25.57 25.76 25.86 25.97 26.08 26.17 26.26 26.33 26.39 26.50Groundwater recharge, GWR (1000 m3)GWR 699.53 760.62 924.43 1012.29 1113.58 1214.85 1339.06 1467.13 1608.77 1933.36Minimum cost of water supply for Case 2 (M. )Cost 699.53 760.62 924.43 1012.29 1113.58 1214.85 1339.06 1467.13 1608.77 1933.36

Results of some years have been presented in Table due to space restriction, however in figures results of all the years have been presented.

186 S.M. Pingale et al. / Resources, Conservation and Recycling 83 (2014) 176– 189

0

3000

6000

9000

12000

15000

18000

21000

24000

27000

30000

33000

2005-06 2009 -10 2013 -14 2017 -18

Quan

tity

(1

00

0 m

3)

Yea rCase 1 Case 2 Case 3

Fc

5u

mcalss

mcttgIhy

5

tdSwdycm

Fe

25

35

45

55

65

75

85

95

105

2005-06 2009-10 2013-14 2017-18

Dem

and

met

(%

)

Yea r

Case 1 wit hout CC Case 1 wit h CCCase 2 wit hout CC Case 2 wit h CCCase 3 wit hout CC Case 3 wit h CC

tainable even with the integration of local sources with imported

ig. 4. Utilization of treated wastewater in various scenarios for different yearsonsidering without considering climate change.

.3.2. Total quantity of imported water and treated wastewatertilized

Total quantity of imported water utilized in different manage-ent strategies (i.e. Case 1–3) has been compared for different

limate change scenarios to highlight the dependence of Ajmer onn imported and uncertain source. Quantity of imported water uti-ized for different year remains same for Case 1–Case 3 in all threecenarios (e.g. Fig. 5), as water demand are more than availableupplies in Case 1 and Case 2.

Quantities of treated wastewater utilized in different manage-ent strategies have also been compared. It is assumed that in

hanging scenario of climate, the quantities of treated wastewa-er must be available as proposed. In first two cases, quantity ofreated wastewater utilized has been zero, as it has not been inte-rated with the water supply system of Ajmer urban fringe (Fig. 4).n third case, total quantity of treated wastewater utilized wouldave been 13.88 Mm3 for the year 2005–2006 to 26.68 Mm3 for theear 2020–2021.

.3.3. Sustainability of water resources systemIn present study, sustainability of water resources system of

he study area has been evaluated in terms of percentage of wateremand satisfied for different years in changing climate scenarios.imultaneously, it has been compared in terms of percentage ofater demand satisfied without considering the climate change forifferent years. Percentage of water demands satisfied in different

ears under different management strategies in different climatehange scenarios have been compared (e.g. Fig. 6). When the cli-ate change scenario is not considered then results revealed that

12500

15000

17500

20000

22500

25000

27500

2005-06 2009 -10 2013 -14 2017 -18

Quan

tity

(1

00

0 m

3)

Yea r

Case 1,2,3 wit hout CC Case 1,2,3 wit h CC

ig. 5. Utilization of imported water in various scenarios in different years consid-ring without and with climate change in A1B scenario.

Fig. 6. Water demands satisfied in various scenarios for different years consideringwithout and with climate change in A1B scenario.

water supply system satisfies the full demand in Case 3 and donot satisfies the full demand in Case 1 and 2. Results revealed thatunder A2 and B1 scenario of climate change, water supply system isdeficient in all the three strategies. However, water supply systemis deficient in first two cases for A1B scenario of climate change.Initially in A1B scenario upto year 2010–2011, water supply sys-tem is found deficient and do not satisfies full demand in Case 3,because lower capacity of treated waste water available in theseyears (Fig. 6). In deficient water supply system, water demands cor-responding to recommended rate of supply will not been satisfiedin changing scenarios of climate change (see Fig. 7).

In Case 1, percentage of demand satisfied has been found tovary from 42.06%, 37.31% and 37.99% for year 2005–2006 to 34.68%,31.06% and 31.57% for year 2020–2021 in A1B, A2 and B1 scenar-ios, respectively. In Case 2, supply scenario has been improvedslightly on account of integration of supplies from groundwaterand local surface sources. However, water supply system is stilldeficient in changing climate scenario. Percentage of demands sat-isfied in Case 2 has been found to vary from 56.82%, 50.24% and51.15% for year 2005–2006 to 51.36%, 45.85% and 46.66% for year2020–2021 in A1B, A2 and B1 scenario, respectively. Average per-centage unmet demand has increased to 24.34% due to climatechange. The demand satisfaction has decreased in changing sce-nario of climate as compared to previous (Case 1). The Case 2revealed that water supply system of Ajmer urban fringe is not sus-

source of water. Therefore, need of integration of other possiblesources or demand management is felt so that demand of watercan be met in changing climate. The model results revealed that

5500

10500

15500

20500

2005-06 2008-09 2011 -12 2014 -15 2017 -18 2020 -21

Quan

tity

(1

00

0 m

3)

Yea rCase 1 wit hout c c Case 1 wit h ccCase 2 wit hout c c Case 2 wit h ccCase 3 wit hout c c Case 3 wit h cc

Fig. 7. Total losses of water in various scenarios for different years consideringwithout and with climate change in A1B scenario.

S.M. Pingale et al. / Resources, Conservatio

200

400

600

800

1000

1200

1400

1600

1800

2005-06 2008-09 2011 -12 2014 -15 2017 -18 2020 -21

Co

st (

M.

)

Yea rCase 1 wit hout CC Case 1 wit h CCCase 2 wit hout CC Case 2 wit h CCCase 3 wit hout CC Case 3 wit h CC

Fc

a1ncchunaor2Ary

pw1twasisfiao

5

mowwi

ma(soHt

to

ig. 8. Minimum cost of water supply system in various scenarios for different yearsonsidering without and with climate change in A1B scenario.

verage shortage of water supply increases to 10.14%, 14.32% and3.74% due to change in climate in different climate change sce-arios i.e. A1B, A2 and B1, respectively as compared to presentondition (without considering climate change). After consideringlimate change scenario, the average percentage unmet demandas been found to be increased 12.73% as compared to percentagenmet demand in present condition if climate change effects wereot considered. Present system is supplying water at the rate ofround 83 l per capita per day (lpcd), as compared to 150 lpcd rec-mmended by the CPHEEO (CPHEEO, 1999). In Case 2, The modelesults revealed that average shortage of water supply increases to0.51%, 26.72% and 25.80% in different climate change scenarios of1B, A2 and B1, respectively. Therefore, Case 1 and Case 2 resultevealed that water demand have not been met (100%) fully for allears due to possible changing climate.

In Case 3, treated wastewater has been integrated as a source ofotable water into present water supply system in addition to localater and groundwater sources. However, only in A1B scenario

00% water demands have been met after year 2010–2011 due tohe proposed increased capacity of treated wastewater (Fig. 3). Fullater demand has not been satisfied in other two scenarios (i.e. A2

nd B1). Therefore, water supply system of Ajmer urban fringe isustainable only with the integration of treated wastewater includ-ng local water and groundwater sources in a changing climatecenario. Hence, integration of treated wastewater is inevitableor the water deficient system of Ajmer urban fringe in a chang-ng climate. Similarly, demand management strategies have to bedopted to manage the increased demand for changing scenariosf climate change (like A2 and B1 scenario).

.3.4. Overall cost of water supplyGenerally, overall cost of any project is considered as one of the

ost important decision criterions for selection of alternative withptimal performance in view of changing climate. Overall cost ofater supply for different years in different management strategiesith and without considering climate change has been presented

n Fig. 8.Overall unit cost of water supply (cost of water supply per cubic

etres of water) in Case 1, is more than Case 2 and Case 3. Over-ll cost of water supply has been found to 221.16 Million RupeesM. ) for year 2005–2006 to 758.19 M. for year 2020–21 in A1Bcenario (Fig. 8) however demands are not satisfied fully. Similarly,verall cost of water supply in A2 and B1 has been found as lowest.owever, full demands have not been met. In Case 2, results found

hat average minimum cost of water supply reduced to 141.96 M., 194.90 M. and 187.22 M. in A1B, A2 and B1 scenarios, respec-

ively. In these cases, the overall cost is low due to unmet demandf water in all three scenarios of climate change. If penalty cost

n and Recycling 83 (2014) 176– 189 187

is imposed due to unmet demands cost of water supply would bemore in Case 1 and Case 2, as demands have not been met in thesescenarios.

In Case 3, overall cost of water supply in different years havebeen found to vary from 459.4 M. , 432.23 M. and 436.06 M. foryear 2005–2006 to 1933.36 M. , 1767.74 M. and 1798.45 M. foryear 2020–2021 in A1B, A2 and B1 scenario, respectively. Cost ofwater supply has been increased in this case as compared to Case2 i.e. from 840.4 M. to 1933.36 M. for year 2020–2021 in A1Bscenario, because of increase in supplies from treated wastewater.Similarly, cost of water supply has increased in case 3 as comparedto Case 2 for A2 and B1 scenario from 751.88 M. to 1767.74 M. and764.62 M. to 1798.45 M. for year 2020–2021. The comparativeresults of with and without considering climate change indicatesthat average minimum cost of water supply increases to 75.15 M., 10.44 M. and 25.17 M. in A1B, A2 and B1 scenarios, respectivelydue to adverse impact of climate change. The increase in cost isattributed to the increased use of treated wastewater due to declinein water availability from surface sources due to change in climate(e.g. Fig. 8). Overall cost of water supply has increased in this caseas compared to previous cases on account of supplemental sup-ply from local surface sources after considering the urbanizationeffect and treated wastewater for potable use with proper addi-tional treatment. However, if penalty cost for unmet demands inCase 1 and Case 2 is considered, overall cost of water supply forAjmer fringe will be lowest. Also water supply system is found tobe sustainable in Case 3 event after year 2020–2021. Therefore, case3 is the optimum water management strategy for the Ajmer fringein a climate change scenario.

5.3.5. Social acceptabilityResults revealed that present water supply system is sustain-

able only with the re-use of treated wastewater along with use ofgroundwater and local surface sources after year 2011–2012 in A1Bscenario of climate change. The average percentage unmet demandhas been increased to 7.94% due to the climate change in thesemanagement strategies. Therefore, re-use of treated water have asignificant importance in managing the growing demands of waterfrom different users in changing climate and can be a major sourceof augmentation of water supply for potable and non-potable useswith proper additional treatment. In present scenario, re-use oftreated wastewater has been proposed for the potable uses withproper additional treatment, which is technically and economicallyfeasible. However, social acceptability of such a management pol-icy is difficult as people’s perception about the re-use of treatedwastewater for potable uses in India is still not favourable. For waterdeficient urban centres, where conventional sources are available inlimited quantity, re-use of treated wastewater is inevitable to sat-isfy the growing demand of water and for the sustainability of waterresources system. There is a need in current scenario of climatechange for creating awareness to increase the social acceptabilityof re-use of treated wastewater, and similarly other possible alter-native ways can be explored like re-use of treated wastewater fornon-potable domestic uses, agriculture, landscape gardening andgroundwater recharge etc.

5.3.6. Environmental benefitsVarious management strategies in different scenarios of climate

change have been compared in terms of environmental bene-fits, like demand satisfied, pollution of water sources and watertransfers from other regions. Water transfer from other regionswould bring imbalance in regional water systems which may lead

to unwanted environmental and other socio-economic conflicts.Therefore, a water management strategy which uses less quantityof water from imported source is better. Thus, it can be concludedthat Case 3 (A1B scenario), in which treated wastewater has been

1 rvatio

ptihscmws

6

CdAgsssihcifimwwrTwsmida

A

EcTDd

R

A

A

A

B

B

B

B

B

CC

88 S.M. Pingale et al. / Resources, Conse

roposed to be integrated is most suitable in changing climate. Inhis case, water supply system is sustainable and social acceptabil-ty is also better as compared to other two cases. Therefore, Case 3as been proposed as the best management solution for the waterupply of Ajmer urban fringe in a scenario of changing climate. Itan be concluded that with the suitable integration of supply aug-entation and demand management measures, present deficientater supply system of Ajmer urban fringe can be converted into a

ustainable system in a changing climate.

. Conclusions

An Integrated Urban Water Management Model consideringlimate Change has been developed and its application has beenemonstrated successfully for realistic water supply system ofjmer fringe. Model has been run for different management strate-ies under different climate change scenarios, and finally, optimumtrategy has been found for different years. Present water supplyystem (imported source) has been examined and integration ofome of other potential sources has been investigated in scenar-os of climate change. Effect of urban dynamic land use/land coveras been incorporated into water resources planning of the area inhanging climate scenario. Urbanization effect has been integratedn terms of increase in quantity of water available from local sur-ace sources. Rationing and re-use of treated wastewater have beenncorporated within the optimization model as demand manage-

ent options. From the comparative study, Case 3 (A1B scenario), inhich treated wastewater has been integrated for the potable usesith prior additional treatment, has been found to be the best and

ecommended for the implementation in the Ajmer urban fringe.he application of model has successfully demonstrated for theater supply system of Ajmer urban fringe under climate change

cenarios. The model has been found capable in finding out the opti-um integration of various sources into the water supply system

n changing climate scenario. Presented model would be useful forecision makers/authorities of urban areas in optimum planningnd utilization of water resources in changing climate.

cknowledgements

The authors are thankfully acknowledged to M.A. Semenov and.M. Barrow for proving the LARS-WG 5.11 and India Meteorologi-al department (IMD), Pune for providing meteorological data sets.he financial assistance provided by Ministry of Human Resourceevelopment, Government of India in the form of a scholarship isuly acknowledged.

eferences

jmera AK. Institutional analysis in water and sanitation sector: a case study ofAjmer water supply system. MREC, Jaipur (India): Department of Civil Engi-neering; 2000 (Unpublished M.E. Dissertation).

ron G, Scott VH. Dynamic programming for conjunctive use. Journal of HydraulicDivision 1971;97(5):705–21.

sano T. Reusing urban waste water—an alternative and a reliable water resource.Water International 1994;19(1):1–6.

arlow David PM, Ahlfeld P, Dickerman DC. Conjunctive-management model forsustained yield of stream-aquifer systems. Journal of Water Resource Planningand Management 2003;129(1):35–48.

elaineh G, Peralta RC, Hughes TC. Simulation/optimisation modelling for waterresources management. Journal of Water Resource Planning and Management1999;125(3):154–60.

ouwer H. Integrated water management for the 21st century: problems and solu-tions. Journal of Irrigation and Drainage Engineering 2002;128(4):193–202.

redehoeft JD. Conjunctive use of groundwater and surface water for agriculturerisk aversion. Water Resource Research 1983;19(5):1111–21.

uras N. Conjunctive operation of dams and aquifers, Journal of the Hydraulic Divi-sion. In: Proceedings of the ASCE, vol. 89, No. HY6; 1963. p. 111–31.

ensus of India, 2001 and 2011.PHEEO. Manual on water supply and treatment. In: Central Public Health and

Environmental Engineering Organization, Govt. of India; 1999.

n and Recycling 83 (2014) 176– 189

Dracup JA. The optimal use of ground and surface water system: a parametric linearprogramming approach. Univ. of California: Water Resources Center, Contribu-tion No. 107; 1966.

Haddad BM. Monterey county water recycling project: institutional study. Journalof Water Resource Planning and Management 2002;128(4):280–7.

Jat MK. An integrated conjunctive use modelling of water in an urban area usingremote sensing and GIS. Roorkee 247 667, UK, India: IIT Roorkee; 2007 (Unpub-lished Ph.D. thesis).

Jenkins MW, Lund JR, Howitt RE, Draper AJ, Msangi SM, Tanaka SK, et al. Optimisa-tion of California’s water supply system: results and insights,. Journal of WaterResource Planning and Management, ASCE 2004;130(4):271–80.

Jha R, Sharma KD, Singh VP. Critical appraisal of methods for the assessment ofenvironmental flows and their application in two river systems of India. KSCEJournal of Civil Engineering 2008;12(3):213–9.

Karamouz M, Kerachian R, Zahraie B. Monthly water resources and irrigation plan-ning: case study of conjunctive use of surface and groundwater resources.Journal of Irrigation and Drainage Engineering 2004;130(5):93–8.

Kashyap D, Chandra S. A distributed conjunctive use model for optimal croppingpattern. IAHS Publication 1982;135:377–84.

Khare D, Jat MK, Ediwahyunan. . Assessment of conjunctive use planning options: acase study of Sapon irrigation command area of Indonesia. Journal of Hydrology2006;328:764–77.

Kulga D, Cakmak C. Towards sustainable water management in the southeast Ana-tolia Project (GAP). Water Resource Development 1997;13(4):541–6.

Lakshminarayana V, Rajagopalan SM. Optimal cropping pattern for basin in India.Journal of Irrigation and Drainage Engineering 1977:53–70.

Lefkoff LJ, Gorelick SM. Simulating physical processes and economic behavior insaline, irrigated agriculture: model development. Water Resource Research1990;26(7):1359–69.

Lin QG, Huang GH, Brad B, Nie XH, Zhang XD, Qin XS. EMDSS: an optimization-baseddecision support system for energy systems management under changing cli-mate conditions—an application to the Toronto-Niagara Region, Canada. ExpertSystems with Applications 2010;37(7):5040–51.

Liu Y, Qin XS, Guo HC, Zhou F. ICCLP—an inexact chance-constrained linear pro-gramming model for land-use management of lake areas in urban fringes.Environmental Management 2007;40(6):966–80.

Lund JR, Cabrera E. Regional water system management: water conservation, watersupply and system irrigation. In: Regional Water System Management, WaterConservation, Water Supply, System Integration, Cabrera E, Cobacho R, Lund JR,editors. Swets and Zeitlinger; 2002. p. 3–22.

Maddock T III, Haimes YY. A tax system for groundwater management. WaterResource Research 1975;11(1):7–14.

Maheepala S, Roberts P. Towards development of an improved method for esti-mation of climate-corrected urban water use. In: 30th Hydrology and WaterResources Symposium, Conference CD 4–7; December 2006.

McFarlane NA, Boer GJ, Blanchet JP, Lazare M. The Canadian Climate Centre second-generation general circulation model and its equilibrium climate. Journal ofClimate 1992;5:1013–44.

Mitchell VG, Cleugh HA, Grimmond CSB, Xu J. Linking urban water balance andenergy models to analyze urban design options. Hydrological Processes 2007.,http://dx.doi.org/10.1002/hyp.

Mishra A, Singh R, Raghuwanshi NS. Development and application of an integratedoptimization simulation model for major irrigation project. Journal of Irrigationand Drainage Engineering, ASCE 2005;131(6):504–13.

Morel-Seytoux HJ, Daly CJ. A discrete kernel generator for stream aquifer studies.Water Resource Research 1975;11(2):253–60.

Nieswand GH, Grandstrom ML. A chance constrained approach to the conjunctiveuse of surface and groundwater. Water Resources Research 1971;7(6):1425–36.

Nikam B. R. Distributed conjunctive use modelling using remote sensing and GIS.Roorkee 247 667, UA, India: IIT Roorkee; 2012 (Unpublished Ph.D. thesis).

Nishikawa T. Water resources optimization model for Santa Barbara, California. Jour-nal of Water Resource Planning and Management, ASCE 1998;124(5):252–63.

O’mara, Gerald T, Duloy JH. Modeling efficient water allocation in a conjunctiveuse regime: the Indus Basin Pakistan. Water Resource Research, The AmericanGeophysical Union 1984;20(11):1489–98.

Onta PR, Shakya RM, Gupta AD. Integrated use of surface and groundwater for urbanwater supply in greater Kathmandu. In: Proc. of International Conference onEnvironmentally Sound water resources Utilization, held at Bangkok, Thailand,from 8 to 11 November; 1993. p. 10–5.

Perera BJC, Muttil N, Hasofer M. Climate corrected urban water use and estima-tion of water savings. In: 18th World IMACS/MODSIM Congress, Cairns; 2009.p. 3633–9.

PHED. Ajmer water supply system: present status and future. A Technical Report.Ajmer: Govt. of Rajasthan; 2004.

Pulido-Velazquez M, Andreu J, Sahuquillo A. Economic optimisation of conjunc-tive use of surface water and groundwater at the basin scale. Journal of WaterResources Planning and Management 2006;132(6):454–67.

Qin XS, Huang GH, Chakma A, Nie XH, Lin QG. A MCDM-based expert sys-tem for climate-change impact assessment and adaptation planning—a casestudy for the Georgia Basin, Canada. Expert Systems with Applications2008;34(3):2164–79.

Qin XS, Xu Y. Analyzing urban water supply through an acceptability-index-basedinterval approach. Advances in Water Resources 2011;34(7):873–86.

Ragab R, Prudhomme C. Climate change and water resources management in aridand semi-arid regions: prospective and challenges for the 21st century. Biosys-tems Engineering 2002;81(1):3–34.

rvatio

R

S

S

S

S

S

S

T

T

T

T

V

W

W

S.M. Pingale et al. / Resources, Conse

UIDP. Ajmer City investment plan. Final Report to Asian Development Bank (ADB).Prepared by Louis Berger International Inc., for Govt. of Rajasthan; 1998.

enge PM. The fifth discipline: the art & practice of the learning organization. NewYork: Dell Publishing Group Inc; 1994.

hao LG, Qin XS, Xu Y. A conditional value-at-risk based inexact water allocationmodel. Water Resources Management 2011;25(9):2125–45.

harma SP. Multistage modelling and conjunctive management of water resourcesin a canal command. India: Department of Civil Engineering, Indian Institute ofTechnology Kanpur; 1987 (Unpublished Ph.D. thesis).

olecki DW, Oliveri C. Downscaling climate change scenarios in an urban land usechange model. Journal of Environmental Management 2004;72:105–15.

track JE Sr, Steyaert RAP, Knox LTRG. Sensitivity of June near-surface temperaturesand precipitation in the eastern United States to historical land cover changessince European settlement. Water Resources Research 2008;44(W11401).,http://dx.doi.org/10.1029/2007WR006546.

yaukt S, Fox GC. Conjunctive surface and ground water management in theJakarta region of Indonesia. Journal of the American Water Resource Association2004;40(1):241–50.

aleb A, Maher F. Impacts of global climate change scenarios on water supply anddemand in Jordan. Water International 2000;25(3):457–63.

jallingii SP. Strategies in urban water design. In: Preprints of the international sym-posium on hydrological processes and water management in urban areas; April24–29, 1988. p. 791–802.

ozer CR, Kiem AS, Verdon-Kidd DC. On the uncertainties associated with using grid-ded rainfall data as a proxy for observed. Hydrology and Earth System Sciences2012;16:1481–99.

ripathi MP, Panda RK, Raghuwanshi NS, Singh R. Hydrological modelling of a smallwatershed using generated rainfall in the soil and water assessment tool model.Hydrological Processes 2004;18:1811–21.

edula S. Optimal irrigation planning in river basin development: the case studyof upper Cauvery Basin. In: Chaturvedi MC, Rogers P, editors. Water resourcessystem planning: some case studies for India. Bangalore, India: Indian Academyof Sciences; 1985. p. 223–52.

ang S, Huang GH. An interval-parameter two-stage stochastic fuzzy program with

type-2 membership functions: an application to water resources management.Stochastic Environmental Research and Risk Assessment 2013;27:1493–506.

ang S, Huang GH. Identifying optimal water resources allocation strategiesthrough an interactive multi-stage stochastic fuzzy programming approach.Water Resource Management 2012;26:2015–38.

n and Recycling 83 (2014) 176– 189 189

Wang S, Huang GH. Interactive two-stage stochastic fuzzy programmingfor water resources management. Journal of Environmental Management2011;92:1986–95.

WaterSmart 2006. Draft Supply Demand Strategy for Melbourne 2006–2055. Pub-lished by WaterSmart; April 2006.

White SB, Fane SA. Designing cost effective water demand management programsin Australia. Water Science and Technology 2005;76(6–7):225–32.

Wigmosta BMS, Burges SJ. Proposed model for evaluating urban hydrologicchange. Journal of Water Resources Planning and Management 1990;116(6):742–63.

Witter VJ, Bogardi JJ. Optimal integrated water management: a quantitativeapproach. In: International conference on environmentally sound waterresources utilization, held at Bangkok, Thailand, from 8 to 11 November; 1993.p. 1–9.

Wolf L, Hötzl H. Different levels and examples of integrated concepts in waterresources management. In: Paper presented at international conference onhydrology and watershed management (ICHWAM-2006). Centre for WaterResources, Institute of Science & Technology; 2006.

Wurbs RA, Muttiah RS, Felden F. Incorporation of climate change in water availabilitymodeling. Journal of Hydrologic Engineering 2005:375–85.

Yang Y, Lin X, Zhou L, Kalin RM. Sustainable ground water management underimpacts of urban growth in a developing country—a case study in thecity of Shijadran, China. Environmental Management and Health 1999;10(1):18–27.

Young RA, Bredehoeft JD. Digital computer simulation for solving management prob-lem of conjunctive groundwater and surface water systems. Water ResourceResearch, The American Geophysical Union 1972;8(3):533–56.

Yu W, Haimes YY. Multilevel optimisation for conjunctive use of groundwater andsurface water. Water Resources Research 1974;10(4):625–36.

Zarghami M, Akbariyeh S. System dynamics modeling for complex urban watersystems: application to the city of Tabriz, Iran. Resources, Conservation andRecycling 2012;60:99–106.

Zarghami M, Hajykazemian H. Urban water resources planning by using a modifiedparticle swarm optimization algorithm. Resources, Conservation and Recycling

2013;70:1–8.

Zuidema FC. Aspects of water resources planning and management in the urban-ization process. In: Laconte P, Haimes YY, editors. Water resources land-useplanning: a systems approach. The Hague: Martinus Nijhoff Publishers; 1982. p.297–340.