optimal development of groundwater in a well …...optimal development of groundwater in a well...

IRRIGATION AND DRAINAGE

Irrig and Drain 62 363ndash376 (2013)

Published online 26 April 2013 in Wiley Online Library (wileyonlinelibrarycom) DOI 101002ird1736

OPTIMAL DEVELOPMENT OF GROUNDWATER IN A WELL COMMAND OF EASTERNINDIA USING INTEGRATED SIMULATION AND OPTIMIZATION MODELLINGdagger

S MOHANTY1 MADAN K JHA2 ASHWANI KUMAR1 AND P S BRAHMANAND1

1Directorate of Water Management Bhubaneswar Odisha India2AgFE Department IIT Kharagpur Kharagpur West Bengal India

ABSTRACT

Quantitative techniques such as simulation modelling and optimization play a vital role in the management of complex ground-water systems This study demonstrates the combined use of groundwater-flow and resource optimization models to scientif-ically address the water scarcity problem in well-based command areas through a case study in eastern India A transientsimulation-optimization model was developed for the study area using Visual MODFLOW (groundwater-flow simulation tool)and the response-matrix technique to maximize pumping from the existing tubewells The optimized maximum pumping ratesobtained from the integrated simulation-optimization model were further used in linear programming-based optimizationmodels to determine optimal cropping patterns for the wet normal and dry scenarios The net annual income from the optimalcropping patterns for the wet normal and dry scenarios were estimated at Rs 818 million Rs 764 million and Rs 716 mil-lion respectively The results of simulation-optimization modelling indicated that if the suggested optimal cropping patternsare adopted in the study area the net annual irrigation water requirements will be reduced by 28 35 and 40 and net annualincome will be increased by 28 23 and 17 during wet normal and dry scenarios respectively Copyright copy 2013 John Wileyamp Sons Ltd

key words simulation-cum-optimization modelling response matrix optimal cropping pattern groundwater management well command area

Received 22 September 2012 Revised 16 January 2013 Accepted 16 January 2013

REacuteSUMEacute

Les techniques quantitatives telles que la modeacutelisation de simulation et drsquooptimisation jouent un rocircle vital dans la gestion dessystegravemes complexes drsquoeaux souterraines Cette eacutetude de cas en Inde orientale deacutemontre que lrsquoutilisation combineacutee de modegravelesde preacutelegravevement et de gestion de lrsquoeau souterraine permet de srsquoattaquer au problegraveme scientifiquement poseacute de la rareteacute de lrsquoeaudans lrsquoaire contributive drsquoun puits Un modegravele transitoire de simulation-optimisation a eacuteteacute deacuteveloppeacute pour la zone drsquoeacutetude enutilisant drsquoune part Visual MODFLOW (outil de simulation de flux des eaux souterraines) et drsquoautre part la technique de lareacuteponse agrave matrice pour maximiser le pompage dans les forages existants Les taux de pompage maximiseacutes fournis par lesmodegraveles de gestion inteacutegreacutee ont ensuite eacuteteacute utiliseacutes dans les modegraveles de programmation lineacuteaire afin de deacuteterminer lesassolements optimaux pour les sceacutenarios humides normaux et secs Le revenu annuel optimal pour ces sceacutenarios humidesnormaux et secs a eacuteteacute estimeacute agrave 818 millions 764 millions et 716 millions de roupies respectivement Les reacutesultats de lamodeacutelisation indiquent que si les caracteacuteristiques optimales de reacutecolte proposeacutees sont adopteacutees dans la zone drsquoeacutetude lesbesoins annuels nets en eau drsquoirrigation seront reacuteduits de 28 35 et 40 alors que le reacutesultat net annuel sera augmenteacute de28 23 et 17 au cours des sceacutenarios humides normaux et secs respectivement Copyright copy 2013 John Wiley amp Sons Ltd

mots cleacutes simulationoptimisation de la modeacutelisation matrice de reacuteponse assolement optimal gestion des eaux souterraines aire contributive drsquoun puits

Correspondence to S Mohanty Directorate of Water ManagementChandrasekharpur PO- Railvihar Bhubaneswar Odisha 751 023 IndiaE-mail smohanty_wtceryahoocoindagger Le deacuteveloppement optimal des eaux souterraines dans lrsquoaire contributivedrsquoun puits de lrsquoInde orientale en utilisant la simulation inteacutegreacutee et desmodegraveles drsquooptimisation

Copyright copy 2013 John Wiley amp Sons Ltd

INTRODUCTION

Groundwater is a very important and invaluable naturalresource on the earth Its unique qualities of being easilyaccessible generally free from pathogens and suspended

364 S MOHANTY ET AL

particles and requires no or little treatment have made it themost important and preferred source of water for domesticagricultural and industrial uses It also serves as the onlyreliable source of water supply during emergency periodsHowever aquifer depletion worldwide due to over-exploitation and the growing pollution of groundwater arethreatening the sustainability of water supply and ecosys-tems on the earth (Shah et al 2000 Zektser 2000Sophocleous 2005 Biswas et al 2009) Hence a seriousconcern is how to maintain a long-term sustainable yieldfrom aquifers (eg Hiscock et al 2002 Alley and Leake2004) in the face of looming climate change and socio-economic changes

In India the demand for water has already increased man-ifold over the years due to growing urbanization increasingpopulation agriculture expansion rapid industrializationand economic development and it has an increasing trendin all sectors (Kumar et al 2005 Mall et al 2006)Roughly 52 of irrigation consumption across the countryis extracted from groundwater and there are several partsof the country that face water scarcity due to intensivegroundwater exploitation (Central Ground Water Board(CGWB) 2006) and mismanagement Excessive groundwa-ter exploitation has led to an alarming decrease in ground-water levels in several parts of the country such as TamilNadu Gujarat Rajasthan Odisha West Bengal Punjaband Haryana (Sharma and Gupta 1999 Mall et al 2006CGWB 2006) In recent studies analysis of GRACE satel-lite data has revealed that the groundwater reserves in thestates of Rajasthan Punjab and Haryana are being depletedat a rate of 177 45 km3 yr1 (Rodell et al 2009) Itwas also found that between August 2002 and December2008 the above-mentioned north-western states of India lost109 km3 of groundwater which is double the capacity ofIndiarsquos largest reservoir Wainganga and almost triple thecapacity of rsquoLake Meadrsquo the largest man-made reservoir inthe United States (Rodell et al 2009) Thus the depletionof groundwater resources has increased the cost of pumpingcaused seawater intrusion in coastal areas and has raisedquestions about sustainable groundwater supply as well asenvironmental sustainability Therefore efficient and judiciousutilization of surface and groundwater resources is essential toprotect vital groundwater resources as part of sustainable landand water management strategies

The state of Odisha in eastern India is no exception andhas its own share of water problems with diverse situationsin different parts such as the recurrence of drought inwestern parts pockets of saline water in the coastal tract andacute water scarcity in many other parts Because of theuneven nature of rainfall and its capricious distribution thereis an increasing dependence on groundwater resources tomeetthe growing water demand from the agriculture industrial anddomestic sectors The overexploitation of groundwater has


resulted in declining groundwater levels in several areas andseawater intrusion in coastal areas (CGWB 2006) Eventhough sufficient water is available in coastal areas in themonsoon (wet) season there is a water shortage for irriga-tion in the post-monsoon (dry) season Therefore thereis a need to develop optimal groundwater managementstrategies to increase the area under post-monsoon seasoncrops and thereby sustain agricultural productivity andlivelihoods

Quantitative techniques are required to best satisfy thecompeting water demands in a basin Simulation models inconjunction with optimization models constitute a powerfulset of tools for maximizing utilization of available land andwater resources minimizing adverse impacts on the envi-ronment and for providing cost-effective managementgoals In the last four decades groundwater simulationmodels (Ting et al 1998 Asghar et al 2002 Lin andMedina 2003 Sarwar and Eggers 2006 Zume andTarhule 2008 Al-Salamah et al 2011) and groundwatersimulation-optimization models (Peralta and Datta 1990Hallaji and Yazicigil 1996 Jonoski et al 1997 Belainehet al 1999 Barlow et al 2003 Uddameri and Kuchanur2007 Ahlfeld and Gemma 2008 Safavi et al 2010) havebeen widely used for developing optimal groundwater man-agement strategies in different parts of the world Howeverin developing countries like India basin-wide groundwatermodelling studies are still in the initial stage due to severalsocio-scientific factors such as the lack of adequate andgood-quality field data financial resources infrastructureand proper technical expertise As a result as yet very fewstudies on basin-wide groundwater-flow modelling (Ahmedand Umar 2009 Raul et al 2011) and groundwatersimulation-cum-optimization modelling (Garg and Ali2000 Rejani et al 2009) have been carried out in India ingeneral and eastern India in particular

The present study was conceived in order to demonstratethe efficacy of combined groundwater-flow simulation andresource optimization models in developing efficientstrategies for the sustainable management of groundwaterin command areas dominated by well irrigation To achievethis objective a study area the KathajodindashSurua inter-basinlocated in Orissa state in eastern India was selected Thepresent study is first of its kind in the study area

STUDY AREA

Location and climate

The study area the KathajodindashSurua inter-basin issurrounded on both sides by the Kathajodi River and itsbranch the Surua The study area is located around the con-fluence of the Mahanadi River with the Bay of Bengal alongthe eastern coast of India (Figures 1 and 2) It is located


Figure 1 Map of the study area showing geographical location and other details

365OPTIMAL GROUNDWATER DEVELOPMENT IN WELL COMMAND

between 85o54rsquo21 to 86o00rsquo41 E and 20o21rsquo48 to 20o

26rsquo00 N The total geographical area of the study area is35 km2 The area is characterized as a tropical humid climatewith an average annual rainfall of 1650mm of whichapproximately 80 occurs during June to October

Cropping pattern and irrigation scenario

Agriculture is the major occupation of the inhabitants Thetotal cultivated area in the study area is 2445 ha of which1365 ha (558) is irrigated land The area of low land is408 ha (167) medium land 1081 ha (442) and highland 956 ha (391) Paddy is the major crop in themonsoon season whereas vegetables potato groundnutgreengram blackgram and horsegram are grown in thepost-monsoon season Groundwater is the only source ofirrigation in the study area There are 69 functioninggovernment tubewells in the area (Figure 2) which constitute


major sources of groundwater withdrawal for irrigation Thesetubewells were constructed and managed by the Odisha LiftIrrigation Corporation Government of Odisha India Nowthey have been handed over to the local water usersrsquo associa-tions Thus the command area of this study area can be calleda rsquowell command arearsquo as opposed to the widely used termlsquocanal command arearsquo

As such there is nowater shortage during themonsoon (wet)season in the study area but in the latter part of post-monsoon(dry) season farm ponds and dug wells dry up and ground-water levels start to go down As a result water scarcityoccurs in the study area during the dry season and it threatensthe sustainability of agricultural production and livelihoodsThis situation warrants that water management strategiesshould be evolved using modern tools and techniques for thesustainable management of water resources particularlygroundwater in the study area


Figure 2 Location of observation and pumping wells and geological cross-sections in the study area

366 S MOHANTY ET AL

Hydrogeology and groundwater scenario

The detailed hydrological and hydrogeological investiga-tions in the study area are presented in Mohanty et al(2012) The river basin is underlain by a semi-confined aquiferwhich mostly comprises coarse sand The thickness of theaquifer varies from 20 to 55m and the depth from 15 to 50mover the basin The aquiferrsquos hydraulic conductivity deter-mined by pumping tests varies from 113 to 968mday1whereas the storage coefficient ranges between 143 10-4

and 99 10-4 The annual recharge in the study area isestimated to vary from 288 to 385mm (Mohanty 2012)Analysis of river stage and groundwater level dataindicated the existence of streamndashaquifer interaction in thebasin Comparison of groundwater contour maps of dryand wet seasons indicated that there is about 3ndash4m spatialvariation and about 5ndash6m seasonal variation of groundwa-ter levels over the river basin

MATERIALS AND METHODS

Data collection and analysis

Daily rainfall data of 20 years (1990ndash2009) and daily panevaporation data of 4 years (2004ndash2007) were collectedfrom a nearby meteorological observatory at the Central


Rice Research Institute (CRRI) Cuttack Odisha locatedabout 2 km from the study area The 20 years of rainfall datawere analysed for investigation of annual and monthlyvariations of rainfall in the study area Frequency analysis ofthe monthly rainfall data was also performed consideringprobability distribution functions of normal 2-parameterlog-normal 3-parameter log-normal Pearson type III logPearson type III Gumbel type 1 extremal and generalizedextreme value using SMADA 60 software The best-fit prob-ability distribution function was determined based on chi-square error Using the best-fit probability distribution func-tions for different months the probabilities of monthly rainfallat 20 50 and 80 exceedance of rainfall were found Basedon regional experience they are represented as monthlyrainfall under wet normal and dry scenarios respectively

The lithologic data at 70 sites over the study area werecollected from the Odisha Lift Irrigation Corporation(OLIC) Office Cuttack Odisha The lithologic data wereanalysed in detail which along with other field data wereused to develop a numerical groundwater-flow model ofthe study area Since no groundwater data were availablein the study area a groundwater monitoring programmewas initiated by the authors Monitoring of groundwaterlevels in the study area was carried out by selecting 19tubewells in such a way that they represent approximatelyfour west-east and four northndashsouth cross-sections of the



study area The locations of the 19 monitoring-cum-pumpingwells are shown as filled circles (A to S) and the locations ofother 50 pumping wells are shown as open circles (1 to 50) inFigure 2 Weekly groundwater-level data at the 19 sites weremonitored from February 2004 to October 2007 and wereused for studying the groundwater characteristics in the studyarea and calibration of the groundwater-flow simulationmodel The river stage data at 10 sections were monitoredon a monthly basis and were used for assigning boundaryconditions to the groundwater-flow simulation model

Calculation of crop water requirement

The irrigation requirement of the crops in the study area ismostly in the post-monsoon (rabi) season Paddy sugar-cane potato onion groundnut vegetables greengramblackgram horsegram and mustard are the major crops inthe study area during this season The Department ofAgriculture Government of Odisha has fixed target areas ofcoverage under these rabi crops which are given in Table IThe water requirements of these crops (ie ETc) were deter-mined by the pan evaporation method (Doorenbos and Pruitt1977 Allen et al 1998) as follows

Tablecrops

Sl N

1234567891011

Copy

ETo frac14 Kp Epan (1)

ETcrop frac14 kc ETo (2)

where ETo = reference evapotranspiration (mm day1)Kp = pan coefficient Epan = pan evaporation (mm day1)ETcrop = crop evapotranspiration (mm day1) and kc = cropcoefficient

Given the water requirements of the crops the net irriga-tion requirements of the crops were calculated by deductingthe effective rainfall from the water requirements of thecrops The effective rainfall was estimated by the USDA-SCS method (Doorenbos and Pruitt 1977)

I Targeted area of coverage under different post-monsoon

o Crop Area (ha)

Paddy 710Sugarcane 130Potato 200Onion 20Groundnut 82Winter vegetables 400Summer vegetables 165Greengram 210Blackgram 320Horsegram 150Mustard 36

right copy 2013 John Wiley amp Sons Ltd

Development of groundwater-flow simulation model

A groundwater-flow simulation model of the study area wasdeveloped using Visual MODFLOW software to analysegroundwater conditions in the study area The model wasdeveloped following the groundwater modelling protocol(Anderson and Woessner 1992) which consisted of threemain phases conceptual model development model designand model calibration and validation

Conceptual model The conceptual model of the aqui-fer system prevalent in the study area was developed basedon hydrogeological investigations The thickness of theaquifer varies from 20 to 55m and its depth from the groundsurface varies from 15 to 50m over the basin The aquifermaterial comprises medium sand to coarse sand whereasthe upper confining layer mostly consists of clay Thereare patches of medium and coarse sand within the clay bedwhich are most likely to be leaky in nature Also there aresome clay lenses present in the study area To simplify theactual field situation for groundwater-flow modelling theseclay lenses were ignored when developing the conceptualmodel of the study area The eastern boundary of the studyarea is the Kathajodi River and the western boundary theSurua River (Figure 2) Therefore these boundaries weresimulated as Cauchy boundary conditions The conceptualmodel of the study area along Section JndashJrsquo (Figure 2) isillustrated in Figure 3 which provided a basis for the designand development of a numerical groundwater-flow model ofthe study area

Numerical model design The study area wasdiscretized into 40 rows and 60 columns using the Gridmodule of the Visual MODFLOW software This resultedin 2400 rectangular cells with a dimension of approxi-mately 222 215m The river boundaries were simulatedas Cauchy boundary conditions and the base of the aquifermodelled as a no-flow boundary The location of pumpingwells observation wells and extent of the well screens ofrespective pumping wells were assigned to the model Themodel parameters such as hydraulic conductivity specificstorage groundwater abstraction in all the pumping wellsand groundwater recharge were provided as inputs to themodel

Model calibration and validation The developednumerical groundwater-flow model was calibrated usingweekly groundwater-level data of 19 sites from February2004 to May 2006 Hydraulic conductivity specific storageand recharge were used as calibration parameters Thereaf-ter the calibrated model was validated using weeklygroundwater-level data from June 2006 to May 2007 Theresults of model calibration and validation were evaluated


Figure 3 Conceptual model of the study area

368 S MOHANTY ET AL

using goodness-of-fit criteria such as residual mean abso-lute residual mean standard error of estimate (SEE) rootmean squared error (RMSE) normalized RMSE and corre-lation coefficient (r) Also scatter plots of observed andsimulated groundwater levels together with 1 1 line95 interval lines and 95 confidence interval lines wereprepared in order to examine the efficacy of the flow modelsin simulating groundwater levels

Development of simulation-optimization model

Generation of response matrices The calibrated andvalidated flow-simulation model was used to develop a tran-sient response matrix Transient response functions describethe influence of a unit pulse of pumpage on drawdown overspace and time (Hallaji and Yazicigil 1996 Theodossiou2004) A transient response function in the discrete form isgiven as (Maddock 1972)

Copy

skn frac14XNPW

ifrac141

Xn

jfrac141

bki njthorn1eth THORNqij (3)

where Skn= drawdown at site k at the end of total pumpingperiod n (m) bki(n-j+1)= average drawdown at site k at theend of pumping period n due to a unit pulse of pumpageat the ith well during the jth pumping period (sm2) qij =average discharge at the ith well during the jth pumpingperiod (m3 s1) i= index for the number of wells j= indexfor the time period (months) k = index for the site numbern = total number of time periods and NPW= total numberof pumping wells

A planning horizon of 7months (November to May cor-responding to the post-monsoon irrigation season in theregion with a monthly time step was considered for the devel-opment of the simulation-optimization model Consequently


transient response functions bki(n-j+1) were generated fromrepeated runs of the flow-simulation model for a period of7months by successively subjecting each of the pumpingwells to a discharge of 1 l s1 for the first 1month periodand zero discharge for the rest of the 6months Drawdownsat 69 pumping wells were simulated by pumping of thewells over the 7-month pumping period The responsefunctions thus obtained (a total of 33 327) were thenassembled to form a transient response matrix Theresponse-function approach is usually applicable to alinear system ie a confined aquifer system (Hallaji andYazicigil 1996) Since the KathajodindashSurua inter-basincomprises a confined aquifer use of this approach in thisstudy is justified

Formulation of the transient hydraulic managementmodel A transient hydraulic management model was de-veloped to maximize pumping from the basin by integratingthe transient response matrix generated from groundwaterflow simulation model with the non-linear optimizationmodel The maximum drawdowns in the aquifer usuallyoccur at pumping well sites The prediction of hydraulic heador drawdown at each well location as a response to thepumping is of great interest for proper groundwater planningand management This response was represented in terms ofa response matrix as described in the previous section In thehydraulic management model drawdowns at all the 69 wellswere constrained so that the cumulative effect of pumpingfrom all the wells does not exceed the maximum allowabledrawdown at individual wells Constraints were alsointroduced to ensure that the annual water demand of thestudy area is met to the maximum extent possible A finalconsideration was the physical capability of the systemimposed by well capacity limitations The objective functionof the model was to maximize the total pumpage from thebasin That is



Copy

MaxQ frac14X69

ifrac141

X7

jfrac141

qijdj (4)

where Q= total pumpage from the existing wells (m3) qij=pumpage of the ith well in the jth month (m3 s1) dj=numberof days of pumping in the jth month i= index for well numberand j= index for time period (j=1 to 7 1 for November 2 forDecember and so on) When applying Equation (4) it was as-sumed that the pumps are operated for maximum 12h a day

The above objective function was subjected to the follow-ing three constraints

bull drawdown constraint At the pumping wells the draw-down must not exceed the maximum permissible draw-down for all time steps Using the transient responsefunctions this constraint can be expressed as follows(Maddock 1972)

X69

ifrac141

X7

jfrac141

bki njthorn1eth THORNqijleSmaxk k frac14 1 2 3 69 (5)

Table II Maximum permissible drawdowns (Smax) at 69 sites over

where bki(n-j+1)= average drawdown at the kth site due to aunit pulse of pumpage at the ith well during the jth monthn = total number of time periods k = index for site numberSkmax = maximum allowable drawdown at the kth site (m)

bull water demand constraint The total groundwater de-mand of the crops in each month must be satisfied Thatis

the study area

Site Smax (m) Site Smax (m) Site Smax (m)

A 678 5 572 28 408

X69

ifrac141

qijdj geDj j frac14 1 2 7 (6)

B 740 6 680 29 408C 767 7 580 30 474D 680 8 676 31 474E 703 9 680 32 463F 676 10 582 33 457G 572 11 682 34 463H 703 12 703 35 457I 680 13 703 36 463

where Dj =water demand of the crops in the jth month (m3)dj= number of days of pumping in the jth month

bull pumping capacity constraints The pumping in individ-ual wells cannot exceed the pumping capacity of thewells and it has to be a positive value That is

J 582 14 703 37 392K 463 15 703 38 392L 408 16 740 39 416M 457 17 703 40 464

0le qij le qmaxij i frac14 1 2 3 69 j frac14 1 2 7 (7)

N 392 18 740 41 474O 474 19 680 42 464P 416 20 740 43 423Q 464 21 678 44 416R 423 22 678 45 464S 464 23 680 46 4161 767 24 703 47 4162 767 25 576 48 4643 767 26 408 49 4644 767 27 463 50 464

where qmaxij = maximum pumping capacity of the ith well in

the jth monthThe above optimization model was solved using the

LINGO 80 software package

Inputs of the hydraulic management model The in-puts of the hydraulic management model are the monthly wa-ter demands of the crops drawdown coefficients bki(n j+1)maximum permissible drawdowns Smax

k at 69 well sites


and maximum rates of pumping of the wells At presentthe farmers in the study area operate their pumps withoutany restrictions As a result the pumping from each wellis at its maximum capacity during the dry non-monsoonperiod Therefore the upper limits of pumpage from indi-vidual wells in the months of November to May were con-sidered to be the current rates of pumping during thesemonths The maximum possible drawdown at a site couldbe determined considering the pumping cost land subsi-dence and other socio-scientific factors In this study thedrawdown in individual wells during the dry period wasconsidered as the maximum permissible drawdown whichis summarized in Table II

Optimization model for land and water resourcesmanagement

The transient hydraulic management model estimates themaximum permissible pumpage from the groundwater res-ervoir by the existing pumping wells With the availableland and water resources a linear programming optimizationmodel was formulated to determine the optimal croppingpattern in the study area by maximizing the net annual returnsubject to various land and water-related constraints Themodel was developed to suitably allocate 11 crops vizpaddy sugarcane potato onion groundnut winter vegeta-bles summer vegetables greengram blackgram horsegramand mustard in different land types ie high land medium


370 S MOHANTY ET AL

land and low land so that net annual return can be maxi-mized The optimization model was developed for the wetnormal and dry scenarios respectively

Formulation of optimization model The objectivefunction of the optimization model was to maximize thenet annual return in wet normal and dry scenarios whichis expressed as follows

Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)

where Z= net total annual income corresponding to wetnormal and dry scenarios (Rs) Pj=market price of the jthcrop (Rs kg1) Yj= yield of the jth crop (kg ha

1) Cj= costof cultivation per unit area for the jth crop excluding the costof irrigation water (Rs ha1) Iij= irrigation cost for the jthcrop in the ith type of land (Rs ha1) aij = area under thejth crop in the ith type of land (ha) i= index for land type(1 for high land 2 for medium land and 3 for low land)and j= index for crop type (1 2 3 11)

The objective function was subjected to the following setsof constraints

bull land availability constraint Land allocated to variouscrops must not exceed the total available land undereach category (high land medium land and low land)That is

X11

jfrac141

aijleAi i frac14 1 2 3 (9)

where aij= area under the jth crop in the ith type of land andAi= total area of the ith type of land (i =1 for high land i= 2for medium land and i= 3 for low land)

bull water requirement constraint The net irrigation re-quirement of all the crops grown in the study area inall the months must satisfy the available maximumgroundwater withdrawals for all the pumping wells ina particular month

X3

ifrac141

X11

jfrac141

aijkwijkleWk k frac14 1 2 7 (10)

where aijk = area under the jth crop in the ith type of land inthe kth month wijk =water requirement of the jth crop in theith type of land in the kth month Wk= available irrigationwater (ie maximum groundwater withdrawals) in the kthmonth (k =1 for January k = 2 for February and k = 7for May) The available irrigation water Wk in differentmonths is obtained by the following equation


Wk frac14 X69

ifrac141

qikdk k frac14 1 2 7 (11)

where = irrigation efficiency qik =maximum allowablepumpage of the ith well in the kth month (m3 s1) dk =number of days of pumping in the kth month (k = 1 forJanuary k = 2 for February and k = 7 for May)

bull minimummaximum area constraint Managementconsiderations restrict some maximum and minimumareas under each crop to meet the basic food requirementof the area This constraint is expressed as

ALj le

X3

ifrac141

aijleAUj j frac14 1 2 3 11 (12)

where ALj = minimum area under the jth crop and AU

j =maximum area under the jth crop

The above optimization model was also solved using theLINGO 80 software package

Inputs of the optimization model

Inputs to the management model include the total area underhigh medium and low land maximum and minimum areasunder different crops water requirement of crops in differ-ent months in different land types under different scenarios(ie wet normal and dry) and net income from the cropsin different land types in each scenario The net incomesfrom the crops were computed from the potential cropyields price of the crops cost of cultivation and irrigationcost In addition the maximum permissible pumping ratesof each well were obtained from the developed hydraulicmanagement model and served as an input to the optimiza-tion model The areas under high medium and low landare 956 1081 and 408 ha respectively The maximum areaof each crop was fixed as 50 more than the targeted areaconsidering the practical feasibility and farmersrsquo interestThe minimum area under each crop was fixed as one quarterof the targeted area considering the preference of the localfarmers The cost of cultivation including cost of irrigationof the above-mentioned crops was computed by standardmethodology followed by the Commission for AgriculturalCosts and Prices (CACP) India The mean yield of thecrops and selling price of the crop produce in the marketwere obtained from agriculture statistics Directorate ofEconomics and Statistics Government of Odisha IndiaThe net irrigation requirement in different crops in differ-ent land types ie low medium and high land werecalculated based on the data obtained from the Departmentof Agriculture Government of Odisha Finally the netincomes from the crops under different land types for thethree scenarios were calculated Further an irrigation


Table III Net irrigation requirements of the rabi crops underdifferent scenarios


efficiency of 90 was considered in this study becauseonly groundwater is used in the study area for irrigation

Crop

Net irrigation requirement (mm)

Wet scenario Normal scenario Dry scenario

Paddy 875 938 999Sugarcane 308 458 529Potato 219 261 277Onion 224 266 282Groundnut 199 241 257Winter vegetables 184 237 255Summer vegetables 118 218 271Greengram 110 138 150Blackgram 115 146 158Horsegram 133 172 188Mustard 154 192 210

RESULTS AND DISCUSSION

Rainfall characteristics

Analysis of 20 years (1990ndash2009) of rainfall data in thestudy area indicated an average annual rainfall of 1650 376mm The highest mean monthly rainfall (403mm) witha standard deviation of 194mm was observed in the monthof August Though rainfall events are distributed throughoutthe year the rainy season usually starts from mid-June andlasts up to mid-October November to May is usuallycharacterized as a dry period The most reliable months forrainfall are July August and September The probabilityanalysis of the monthly rainfall data indicated that two-parameter log normal distribution is the best fit to the rainfalldata of January February March and December Pearsontype III distribution for April and November log Pearsontype III distribution for May June August and OctoberGumbel type 1 extremal distribution for the month of Julyand normal distribution for the month of September Figure 4shows the temporal variation of monthly rainfall during wetnormal and dry scenarios Clearly there is a considerablevariation in the amount of monthly rainfall during wet anddry scenarios

Crop water requirement

The net depth of irrigation requirements of different cropsfor the wet normal and dry scenarios are shown in Table IIIThe net irrigation requirements of paddy in the wetnormal and dry scenarios are 875 938 and 999mmrespectively The increase in net irrigation requirementof crops from the dry to normal scenario varied from a

0

100

200

300

400

500

600

Jan Feb Mar Apr May June

Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month

Figure 4 Monthly variation of rainfall du


minimum 72 in the case of paddy to a maximum of848 in the case of summer vegetables However theincrease in net irrigation requirement of crops from thenormal to wet scenario varied from a minimum of 6in the case of onion to a maximum of 243 in the caseof summer vegetables

The total crop water requirements in the months of No-vember December January February March April andMay were estimated at 48 105 90 105 265 105227 105 247 105 222 105 and 81 105m3 re-spectively Similarly the total water requirements for therabi paddy sugarcane potato onion groundnut wintervegetables summer vegetables greengram blackgramhorsegram and mustard were determined as 739 10575 105 58 105 06 105 22 105 107 10550 105 33 105 52 105 30 105 and 08 105m3 respectively

July Aug Sept Oct Nov Dec

nth

ring wet normal and dry scenarios


372 S MOHANTY ET AL

Model validation results

The scatter diagram along with 1 1 line 95 interval linesand 95 confidence interval lines for the validation isshown in Figure 5 It is obvious that the 1 1 line lies withinthe 95 confidence interval lines which indicates satisfac-tory validation of the developed groundwater-flow modelThe residual mean absolute residual mean SEE RMSEnormalized RMSE and correlation coefficient betweenobserved and simulated groundwater levels were found tobe 0044m 0489m 002m 0632m 6527 and 0958respectively (Figure 5) These goodness-of-fit criteria showthat there is a reasonably good calibration and validationof the groundwater-flow simulation model

Maximum permissible groundwater withdrawal

The value of maximum permissible pumpage was highest inthe months of November and December thereafter gradu-ally reducing to a minimum in the months of April andMay From the month-wise maximum permissible pumpingrates in the 69 tubewells the total monthly maximum allow-able pumpage from the river basin was estimated at 163168 160 145 160 147 and 152 million m3 in themonths of November December January February MarchApril and May respectively It should be noted that theactual pumping from a well in each month will depend on

11 Line ------ 95 Interval Line------ 95 Confidence

Interval Line

Figure 5 Scatter diagram of observed versus simulated groundwater levelswileyonlinelibraryco


the cropping pattern followed by the farmers Therefore itis essential to determine the optimal cropping pattern andoptimal pumping schedule for the study area

Furthermore it was found that the optimized maximumpermissible pumping rates in different months are mostlydetermined as the maximum assigned discharge of thewells in a particular month Thus the optimum pumpingrates of each well have been limited due to the capacityof the pumps installed which is lower than the actual wellyield Using the optimized pumping rates as inputs themodel was run to simulate groundwater levels Thedifference between simulated drawdowns and specifieddrawdown constraints (ie maximum allowable drawdown)varied from 14 to 26m

Optimal cropping pattern and groundwater allocation

The optimal allocation of different crops to different landtypes the optimal allocation of groundwater (ie net irriga-tion requirement) to different crops and the net annualincome from the optimal cropping patterns in the wetnormal and dry scenarios are presented in Tables IV (andashc)respectively The crops such as sugarcane potato onionwinter and summer vegetables have been allotted the maxi-mum area possible in all three scenarios because their netincomes are higher The greengram crop has also been allot-ted the maximum area possible in all three scenarios

for the validation period This figure is available in colour online atmjournalird


Table IV Optimal cropping pattern net irrigation requirement and net annual income for the (a) wet scenario (b) normal scenario and(c) dry scenario

(a) Allocated land (ha)

Crop Low land Medium land High land Total Net irrigation requirement (m3)

Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105

Winter vegetables 104 496 - 600 94 105

Summer vegetables - - 248 248 29 105

Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105

Net annual income =Rs 818 106

Note 1 US$=Rs 55 (approximately)

(b) Allocated land (ha)

Net irrigation requirement (m3)Crop Low land Medium land High land Total

Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105


(c) Allocated land (ha)


Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105



whereas the groundnut crop is allotted the maximum areapossible only in the wet and normal scenarios On the otherhand the area under paddy crop has been limited to 304 ha


in the wet scenario and a minimum area of 177 ha in the nor-mal and dry scenarios It should be noted that even thoughpaddy has a higher net income than crops like groundnut


374 S MOHANTY ET AL

and greengram its area has been limited due to higher waterrequirement This finding suggests that paddy cultivation bygroundwater should be minimized by adopting crop diversi-fication Such a strategy is necessary to conserve groundwa-ter resources and protect them from depletion whileensuring enhanced income to the farmers

Tables IV (andashc) show that total cropped area under highlands in the wet normal and dry scenarios are 1204 995and 575 ha respectively However as the cropping periodsof potato onion groundnut greengram blackgram andhorsegram under high lands are exclusive of the cropping pe-riod of summer vegetables the total maximum area underhigh lands at a particular time in the wet normal and dryscenarios are 956 747 and 327 ha respectively These figuresare obtained by deducting the area under summer vegetables(ie 248 ha) from the total area under high land in the respec-tive scenarios [Tables IV (andashc)] Therefore all the areasunder low land (408 ha) medium land (1081 ha) and highland (956 ha) have been covered under crops in thewet scenario whereas in the normal and dry scenariosonly the entire low and medium land have been coveredunder crops The decrease in total cropped area underhigh land in the normal scenario (747 ha) and dry scenario(327 ha) is due to the relative shortage of irrigation waterduring these two scenarios

It is worth mentioning that the optimization modelsuggests total cropped areas of 2693 2484 and 2064 ha inthe wet normal and dry scenarios respectively as com-pared to the total cropped area of 2423 ha targeted by theDepartment of Agriculture Government of Odisha in thestudy area Thus based on the optimal cropping patternsthere is a reduction of 75 in the area underpaddy and an increase of 50 in the area under sugarcanepotato onion winter vegetables summer vegetables and

0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )

Government targeted cropping p

Optimal cropping pattern

Rainfall S

Figure 6 Comparison of net irrigation requirements between government-targete


greengram in the post-monsoon season As the paddycultivation consumes a lot of water and its net income isnot so high it is strongly recommended to reduce the areaunder post-monsoon paddy so as to conserve groundwaterand improve livelihoods

The total net irrigation requirements (ie groundwater with-drawals) for the optimal cropping pattern in the wet normaland dry scenarios are 614 105 658 105 and663 105m3 respectively [Tables IV (andashc)] The net in-comes from the optimal cropping patterns for the wet normaland dry scenarios are estimated at Rs 818 million Rs764 million and Rs 716 million respectively It is apparentthat from the wet to the dry scenario there is a gradual increasein net annual irrigation water use and a decrease in net annualincome from the crops Figure 6 shows the comparison of netirrigation requirement between the government-targetedcropping pattern and the optimal cropping pattern in differentscenarios The net irrigation requirements for the government-targeted cropping patterns are 851 105 101 105 and111 105m3 for the wet normal and dry scenarios respec-tively Similarly Figure 7 shows the comparison of net annualincome between the government-targeted cropping patternand the optimal cropping pattern in different scenarios Thenet annual income from the government- targeted croppingpatterns are Rs 638 million Rs 622 million and Rs 612million for the wet normal and dry scenarios respectivelyThus if the optimal cropping patterns are adopted by thefarmers in the study area there will be saving in irrigationwater requirement of 237 105m3 (278) in the wetscenario 352 105m3 (349) in the normal scenario and447 105m3 (403) in the dry scenario Additionally therewill be an increase in total net income of Rs 180 million(282) Rs 142 million (228) and Rs 103 million(169) in the wet normal and dry scenarios respectively

scenario Dry scenario

attern

cenario

d cropping pattern and optimal cropping pattern under different scenarios


0

10

20

30

40

50

60

70

80

90

100


Government targeted cropping pattern


Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)

Figure 7 Comparison of net annual income between government-targeted cropping pattern and optimal cropping pattern under different scenarios


CONCLUSIONS

The present study was undertaken to demonstrate the efficacyof combined groundwater-flow simulation and resource optimi-zationmodels in the sustainablemanagement of groundwater inwell-based command areas The desired goal was achievedthrough a case study area located in Odisha eastern India Atransient groundwater-flow simulation-cum-optimizationmodelwas developed for the study area using Visual MODFLOW(groundwater-flow simulation tool based on finite differencemethod) and the response-matrix technique to maximizepumping from existing tubewells Finally optimization modelsusing the linear programming method were developed todetermine optimal cropping patterns for the wet normal anddry scenarios

The maximum allowable pumpage of the 69 tubewellsobtained by the simulation-optimization model were found tobe highest in the month of November and it gradually reducesto a minimum in the months of April and May The totalmonthly maximum allowable groundwater withdrawals fromthe river basin were estimated at 163 168 160 145 160147 and 152 million m3 in the months of NovemberDecember January February March April and May respec-tively The optimization model suggested a decrease in the areaunder post-monsoon paddy by 75 and an increase in the areasunder crops like sugarcane potato onion winter vegetablessummer vegetables and greengram by 50 over the existingareas The net annual income from the optimal cropping pat-terns for the wet normal and dry scenarios were estimated atRs 818 million Rs 764 million and Rs 716 million respec-tively By adopting the optimal cropping patterns obtained fromthe optimization model the net annual irrigation water require-ment can be reduced by 237 105m3 and the net annualincome can be enhanced by Rs 180 million for the wet sce-nario Similarly there is a potential to decrease the net annualirrigation requirements by 352 105m3 and 447 105m3


and increase the net annual income by Rs 142 million andRs 103 million for the normal and dry scenarios respectively

Overall the findings of this study are very useful to theplanners and decision makers concerned and based on theresults of this study management strategies could be formulatedfor the efficient utilization of water and land resources in thestudy area The methodology demonstrated in this study beinggeneric it is also useful for other river basins

NOTE1(Rs = Indian rupee 1 rupee=00182 US$ price level 2012)

REFERENCES

Ahlfeld DP Gemma BM 2008 Solving unconfined groundwaterflow management problems with successive linear programmingJournal of Water Resources Planning and Management ASCE134(5) 404ndash412

Ahmed I Umar R 2009 Groundwater flow modeling of Yamuna-Krishniinterstream a part of central Ganga Plain Uttar Pradesh Journal ofEarth System Science 118(5) 507ndash523

Allen RG Pereira LS Raes D Smith M 1998 Crop evapotranspira-tion guidelines for computing crop water requirements Irrigationand Drainage Paper No 56 Food and Agriculture Organization(FAO) Rome Italy

Alley WM Leake SA 2004 The journey from safe yield to sustainabilityGround Water 42(1) 12ndash16

Al-Salamah IS Ghazaw YM Ghumman AR 2011 Groundwatermodeling of Saq Aquifer Buraydah Al Qassim for better watermanagement strategies Environmental Monitoring and Assessment173(1-4) 851ndash860

Anderson MP Woessner WW 1992 Applied Groundwater Modeling Simula-tion of Flow and Advective Transport Academic Press Inc San DiegoCalifornia

Asghar MN Prathapar SA Shafique MS 2002 Extracting relatively freshgroundwater from aquifers underlain by salty groundwater AgriculturalWater Management 52 119ndash137


376 S MOHANTY ET AL

Barlow PM Ahlfeld DP Dickerman DC 2003 Conjunctive managementmodels of streamndashaquifer systems Journal of Water Resources Planningand Management ASCE 129(1) 35ndash48

Belaineh G Perlata RC Hughes TC 1999 Simulationoptimization model-ing for water resources management Journal of Water ResourcesPlanning and Management ASCE 125(3) 154ndash161

Biswas AK Tortajada C Izquierdo R (eds) 2009 Water Management in2020 and Beyond Springer Berlin Germany 268 pp

Central GroundWater Board (CGWB) 2006 DynamicGroundwater Resourcesof India (as at March 2004) Ministry of Water Resources New Delhi India

Doorenbos J Pruitt WO 1977 Guidelines for predicting crop waterrequirements FAO Irrigation and Drainage Paper 24 Food and AgricultureOrganization (FAO) of the United Nations Rome Italy

Garg NK Ali A 2000 Groundwater management in lower Indus basinAgricultural Water Management 42(3) 273ndash290

Hallaji K Yazicigil H 1996 Optimal management of coastal aquifer inSouthern Turkey Journal of Water Resources Planning and Manage-ment ASCE 122(4) 233ndash244

Hiscock KM Rivett MO Davison RM (eds) 2002 Sustainable Ground-water Development Special Publication No 193 Geological SocietyLondon UK

Jonoski A Zhou Y Nonner J 1997 Model-aided design and optimizationof artificial recharge pumping systems Journal of Hydrological Sciences42 (6) 937ndash953

Kumar R Singh RD Sharma KD 2005 Water resources of India CurrentScience 89(5) 794ndash811

Lin YC Medina MA 2003 Incorporating transient storage in conjunctivestreamndashaquifer modeling Advances in Water Resources 26(9) 1001ndash1019

Maddock III T 1972 Algebraic technological functions from a simulationmodel Water Resources Research 8(1) 129ndash134

Mall RK Gupta A Singh R Singh RS Rathore LS 2006Water resources andclimate change an Indian perspective Current Science 90(12) 1610ndash1626

Mohanty S 2012 Groundwater flow and management modeling in a riverisland of Orissa PhD thesis AgFE Department Indian Institute of Tech-nology Kharagpur Kharagpur West Bengal

Mohanty S Jha MK Kumar A Jena SK 2012 Hydrologic andhydrogeologic characterization of a deltaic aquifer system in Orissaeastern India Water Resources Management 26(7) 1899ndash1928


Peralta RC Datta B 1990 Reconnaissance level alternative optimalgroundwater use strategies Journal of Water Resources Planning andManagement ASCE 116(5) 676ndash692

Raul S Panda SN Hollander H Billib M 2011 Integrated water resourcesmanagement in a major canal command in eastern India HydrologicalProcesses 25 2551ndash2562

Rejani R Jha MK Panda SN 2009 Simulation-optimisation modeling forsustainable groundwater management in a coastal basin of Orissa IndiaWater Resources Management 23(2) 235ndash263

Rodell M Velicogna I Famiglietti JS 2009 Satellite-based estimates ofgroundwater depletion in India Nature 460 999ndash1002

Safavi HR Darzi F Marino MA 2010 Simulation-optimization modelingof conjunctive use of surface water and groundwater Water ResourcesManagement 24(10) 1946ndash1969

Sarwar A Eggers H 2006 Development of a conjunctive use model toevaluate alternative management options for surface and groundwaterresources Hydrogeology Journal 14 1676ndash1687

Shah T Molden D Sakthivadivel R Seckler D 2000 The Global Ground-water Situation Overview of Opportunities and Challenges IWMIColombo Sri Lanka

Sharma BR Gupta SK 1999 Regional salt and water balance modeling forsustainable irrigated agriculture in India ICID Journal 48(1) 45ndash52

Sophocleous MA 2005 Groundwater recharge and sustainability in the HighPlains aquifer in Kansas USA Hydrogeology Journal 13(2) 351ndash365

Theodossiou NP 2004 Application of non-linear and optimization modelsin groundwater aquifer management Water Resources Management 18125ndash141

Ting CS Zhou Y Vries JJ de Simmers I 1998 Development of a prelim-inary groundwater flow model for water resources management in thePingtung Plain Taiwan Ground Water 35(6) 20ndash35

Uddameri V Kuchanur M 2007 Simulation-optimization approach toassess groundwater availability in Refugio County TX EnvironmentalGeology 51 921ndash929

Zektser IS 2000 Groundwater and the Environment Applications for theGlobal Community Lewis Publishers Boca Raton Fla

Zume J Tarhule A 2008 Simulating the impacts of groundwater pumpingon streamndashaquifer dynamics in semi-arid north-western Oklahoma USAHydrogeology Journal 16 797ndash810


364 S MOHANTY ET AL

particles and requires no or little treatment have made it themost important and preferred source of water for domesticagricultural and industrial uses It also serves as the onlyreliable source of water supply during emergency periodsHowever aquifer depletion worldwide due to over-exploitation and the growing pollution of groundwater arethreatening the sustainability of water supply and ecosys-tems on the earth (Shah et al 2000 Zektser 2000Sophocleous 2005 Biswas et al 2009) Hence a seriousconcern is how to maintain a long-term sustainable yieldfrom aquifers (eg Hiscock et al 2002 Alley and Leake2004) in the face of looming climate change and socio-economic changes

In India the demand for water has already increased man-ifold over the years due to growing urbanization increasingpopulation agriculture expansion rapid industrializationand economic development and it has an increasing trendin all sectors (Kumar et al 2005 Mall et al 2006)Roughly 52 of irrigation consumption across the countryis extracted from groundwater and there are several partsof the country that face water scarcity due to intensivegroundwater exploitation (Central Ground Water Board(CGWB) 2006) and mismanagement Excessive groundwa-ter exploitation has led to an alarming decrease in ground-water levels in several parts of the country such as TamilNadu Gujarat Rajasthan Odisha West Bengal Punjaband Haryana (Sharma and Gupta 1999 Mall et al 2006CGWB 2006) In recent studies analysis of GRACE satel-lite data has revealed that the groundwater reserves in thestates of Rajasthan Punjab and Haryana are being depletedat a rate of 177 45 km3 yr1 (Rodell et al 2009) Itwas also found that between August 2002 and December2008 the above-mentioned north-western states of India lost109 km3 of groundwater which is double the capacity ofIndiarsquos largest reservoir Wainganga and almost triple thecapacity of rsquoLake Meadrsquo the largest man-made reservoir inthe United States (Rodell et al 2009) Thus the depletionof groundwater resources has increased the cost of pumpingcaused seawater intrusion in coastal areas and has raisedquestions about sustainable groundwater supply as well asenvironmental sustainability Therefore efficient and judiciousutilization of surface and groundwater resources is essential toprotect vital groundwater resources as part of sustainable landand water management strategies

The state of Odisha in eastern India is no exception andhas its own share of water problems with diverse situationsin different parts such as the recurrence of drought inwestern parts pockets of saline water in the coastal tract andacute water scarcity in many other parts Because of theuneven nature of rainfall and its capricious distribution thereis an increasing dependence on groundwater resources tomeetthe growing water demand from the agriculture industrial anddomestic sectors The overexploitation of groundwater has


resulted in declining groundwater levels in several areas andseawater intrusion in coastal areas (CGWB 2006) Eventhough sufficient water is available in coastal areas in themonsoon (wet) season there is a water shortage for irriga-tion in the post-monsoon (dry) season Therefore thereis a need to develop optimal groundwater managementstrategies to increase the area under post-monsoon seasoncrops and thereby sustain agricultural productivity andlivelihoods

Quantitative techniques are required to best satisfy thecompeting water demands in a basin Simulation models inconjunction with optimization models constitute a powerfulset of tools for maximizing utilization of available land andwater resources minimizing adverse impacts on the envi-ronment and for providing cost-effective managementgoals In the last four decades groundwater simulationmodels (Ting et al 1998 Asghar et al 2002 Lin andMedina 2003 Sarwar and Eggers 2006 Zume andTarhule 2008 Al-Salamah et al 2011) and groundwatersimulation-optimization models (Peralta and Datta 1990Hallaji and Yazicigil 1996 Jonoski et al 1997 Belainehet al 1999 Barlow et al 2003 Uddameri and Kuchanur2007 Ahlfeld and Gemma 2008 Safavi et al 2010) havebeen widely used for developing optimal groundwater man-agement strategies in different parts of the world Howeverin developing countries like India basin-wide groundwatermodelling studies are still in the initial stage due to severalsocio-scientific factors such as the lack of adequate andgood-quality field data financial resources infrastructureand proper technical expertise As a result as yet very fewstudies on basin-wide groundwater-flow modelling (Ahmedand Umar 2009 Raul et al 2011) and groundwatersimulation-cum-optimization modelling (Garg and Ali2000 Rejani et al 2009) have been carried out in India ingeneral and eastern India in particular

The present study was conceived in order to demonstratethe efficacy of combined groundwater-flow simulation andresource optimization models in developing efficientstrategies for the sustainable management of groundwaterin command areas dominated by well irrigation To achievethis objective a study area the KathajodindashSurua inter-basinlocated in Orissa state in eastern India was selected Thepresent study is first of its kind in the study area

STUDY AREA

Location and climate

The study area the KathajodindashSurua inter-basin issurrounded on both sides by the Kathajodi River and itsbranch the Surua The study area is located around the con-fluence of the Mahanadi River with the Bay of Bengal alongthe eastern coast of India (Figures 1 and 2) It is located













366 S MOHANTY ET AL















Tablecrops

Sl N

1234567891011

Copy






o Crop Area (ha)










368 S MOHANTY ET AL




Copy

skn frac14XNPW

ifrac141

Xn

jfrac141









Copy

MaxQ frac14X69

ifrac141

X7

jfrac141

qijdj (4)




X69

ifrac141

X7

jfrac141





the study area


A 678 5 572 28 408

X69

ifrac141


B 740 6 680 29 408C 767 7 580 30 474D 680 8 676 31 474E 703 9 680 32 463F 676 10 582 33 457G 572 11 682 34 463H 703 12 703 35 457I 680 13 703 36 463



J 582 14 703 37 392K 463 15 703 38 392L 408 16 740 39 416M 457 17 703 40 464


N 392 18 740 41 474O 474 19 680 42 464P 416 20 740 43 423Q 464 21 678 44 416R 423 22 678 45 464S 464 23 680 46 4161 767 24 703 47 4162 767 25 576 48 4643 767 26 408 49 4644 767 27 463 50 464





k at 69 well sites






370 S MOHANTY ET AL



Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)





X11

jfrac141




X3

ifrac141

X11

jfrac141




Wk frac14 X69

ifrac141




ALj le

X3

ifrac141











Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL











































366 S MOHANTY ET AL















Tablecrops

Sl N

1234567891011

Copy






o Crop Area (ha)










368 S MOHANTY ET AL




Copy

skn frac14XNPW

ifrac141

Xn

jfrac141









Copy

MaxQ frac14X69

ifrac141

X7

jfrac141

qijdj (4)




X69

ifrac141

X7

jfrac141





the study area


A 678 5 572 28 408

X69

ifrac141


B 740 6 680 29 408C 767 7 580 30 474D 680 8 676 31 474E 703 9 680 32 463F 676 10 582 33 457G 572 11 682 34 463H 703 12 703 35 457I 680 13 703 36 463



J 582 14 703 37 392K 463 15 703 38 392L 408 16 740 39 416M 457 17 703 40 464


N 392 18 740 41 474O 474 19 680 42 464P 416 20 740 43 423Q 464 21 678 44 416R 423 22 678 45 464S 464 23 680 46 4161 767 24 703 47 4162 767 25 576 48 4643 767 26 408 49 4644 767 27 463 50 464





k at 69 well sites






370 S MOHANTY ET AL



Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)





X11

jfrac141




X3

ifrac141

X11

jfrac141




Wk frac14 X69

ifrac141




ALj le

X3

ifrac141











Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL




































Tablecrops

Sl N

1234567891011

Copy






o Crop Area (ha)










368 S MOHANTY ET AL




Copy

skn frac14XNPW

ifrac141

Xn

jfrac141









Copy

MaxQ frac14X69

ifrac141

X7

jfrac141

qijdj (4)




X69

ifrac141

X7

jfrac141





the study area


A 678 5 572 28 408

X69

ifrac141


B 740 6 680 29 408C 767 7 580 30 474D 680 8 676 31 474E 703 9 680 32 463F 676 10 582 33 457G 572 11 682 34 463H 703 12 703 35 457I 680 13 703 36 463



J 582 14 703 37 392K 463 15 703 38 392L 408 16 740 39 416M 457 17 703 40 464


N 392 18 740 41 474O 474 19 680 42 464P 416 20 740 43 423Q 464 21 678 44 416R 423 22 678 45 464S 464 23 680 46 4161 767 24 703 47 4162 767 25 576 48 4643 767 26 408 49 4644 767 27 463 50 464





k at 69 well sites






370 S MOHANTY ET AL



Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)





X11

jfrac141




X3

ifrac141

X11

jfrac141




Wk frac14 X69

ifrac141




ALj le

X3

ifrac141











Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL

































368 S MOHANTY ET AL




Copy

skn frac14XNPW

ifrac141

Xn

jfrac141









Copy

MaxQ frac14X69

ifrac141

X7

jfrac141

qijdj (4)




X69

ifrac141

X7

jfrac141





the study area


A 678 5 572 28 408

X69

ifrac141


B 740 6 680 29 408C 767 7 580 30 474D 680 8 676 31 474E 703 9 680 32 463F 676 10 582 33 457G 572 11 682 34 463H 703 12 703 35 457I 680 13 703 36 463



J 582 14 703 37 392K 463 15 703 38 392L 408 16 740 39 416M 457 17 703 40 464


N 392 18 740 41 474O 474 19 680 42 464P 416 20 740 43 423Q 464 21 678 44 416R 423 22 678 45 464S 464 23 680 46 4161 767 24 703 47 4162 767 25 576 48 4643 767 26 408 49 4644 767 27 463 50 464





k at 69 well sites






370 S MOHANTY ET AL



Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)





X11

jfrac141




X3

ifrac141

X11

jfrac141




Wk frac14 X69

ifrac141




ALj le

X3

ifrac141











Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL

































Copy

MaxQ frac14X69

ifrac141

X7

jfrac141

qijdj (4)




X69

ifrac141

X7

jfrac141





the study area


A 678 5 572 28 408

X69

ifrac141


B 740 6 680 29 408C 767 7 580 30 474D 680 8 676 31 474E 703 9 680 32 463F 676 10 582 33 457G 572 11 682 34 463H 703 12 703 35 457I 680 13 703 36 463



J 582 14 703 37 392K 463 15 703 38 392L 408 16 740 39 416M 457 17 703 40 464


N 392 18 740 41 474O 474 19 680 42 464P 416 20 740 43 423Q 464 21 678 44 416R 423 22 678 45 464S 464 23 680 46 4161 767 24 703 47 4162 767 25 576 48 4643 767 26 408 49 4644 767 27 463 50 464





k at 69 well sites






370 S MOHANTY ET AL



Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)





X11

jfrac141




X3

ifrac141

X11

jfrac141




Wk frac14 X69

ifrac141




ALj le

X3

ifrac141











Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL
































370 S MOHANTY ET AL



Copy

MaxZ frac14X3

ifrac141

X11

jfrac141

PjYj Cj Iij

aij (8)





X11

jfrac141




X3

ifrac141

X11

jfrac141




Wk frac14 X69

ifrac141




ALj le

X3

ifrac141











Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL



































Crop









0

100

200

300

400

500

600


Mo

Rai

nfa

ll (m

m)

Wet Month

Normal Month

Dry Month






nth



372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL
































372 S MOHANTY ET AL






Interval Line












Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL



































Paddy 304 ndash - 304 246 105

Sugarcane - 195 - 195 57 105

Potato - - 300 300 64 105

Onion - - 30 30 07 105

Groundnut - - 123 123 24 105



Greengram - - 315 315 32 105

Blackgram - 336 144 480 48 105

Horsegram - - 44 44 06 105

Mustard - 54 - 54 074 105

Total 408 1081 1204 2693 614 105





Paddy 177 - - 177 160 105

Sugarcane - 195 - 195 87 105

Potato - - 300 300 78 105

Onion - - 30 30 08 105

Groundnut - - 123 123 30 105



Greengram - 315 - 315 40 105

Blackgram - 193 92 285 39 105

Horsegram - - 202 202 35 105

Mustard - 9 - 9 02 105

Total 408 1081 995 2484 658 105




Paddy 177 - - 177 174 105

Sugarcane - 195 - 195 102 105

Potato - 62 238 300 83 105

Onion - 30 - 30 08 105

Groundnut - - 52 52 13 105



Greengram - 315 - 315 45 105

Blackgram - 101 - 101 15 105

Horsegram - - 37 37 07 105

Mustard 9 - - 9 02 105

Total 408 1081 575 2064 663 105







374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL
































374 S MOHANTY ET AL




0

2

4

6

8

10

12

14

Wet scenario Normal

Net

irri

gat

ion

req

uir

emen

t (m

illio

n m

3 )



Rainfall S






attern

cenario



0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL
































0

10

20

30

40

50

60

70

80

90

100




Rainfall Scenario

Net

an

nu

al in

com

e (m

illio

n R

s)



CONCLUSIONS







REFERENCES









376 S MOHANTY ET AL
































376 S MOHANTY ET AL
































optimal development of groundwater in a well …...optimal development of groundwater in a well...

Documents