A N D H R A P R A D E S H R U R A L L I V E L I H O O D S P R O G R A M M E2 0 0 3

ANDHRA PRADESH RURAL L IVEL IHOODSPROGRAMME WATER AUDIT

Edited by:Dr M S Rama Mohan Rao,Dr C H Batchelor, Dr A J James,Dr R Nagaraja, Dr J Seeley,Dr J A Butterworth

Major contributors:Dr M Snehalatha (Dhone coordinator),Mr Sundar Raman (Kalyandurg coordinator)and Mr A K Singh (CSWCRTI Tank studies)

Accion Fraterna:Mr Y V Malla Reddy and Mr B Raghuram Reddy

Anantapur DDP:Mr Radha Krishnamurthy, Mrs Suvarna,Mr G V Reddy

CRIDA:Dr H P Singh, Dr P K Mishra,Dr G R Korwar, Dr M Osman,Dr C A Rama Rao

CSWCRTI:Mr M Chandrappa, Mr K K Reddy,Dr S K N Math, Mr R N Adhikari,Mr Ilango, Mr W Muralidhar

Kurnool DPAP:Mr Sanjay Gupta, Mr Dana Kishore,Mr P Laxmanaswamy, Mr D Subba Reddy

NRSA:Dr Rakesh Paliwal, Mr Shyam Sunder,Dr Rajiv Kumar

A N D H R A P R A D E S H R U R A LL I V E L I H O O D S P R O G R A M M E

W AT E R A U D I T

Contact details of editors:

Dr M S Rama Mohan RaoCentral Soil & Water Research & Training Institute,Hospet Road, Bellary 583 104, Karnataka, IndiaE-mail: MSRRao3@yahoo.com.Telephone: + 91 (0) 8392 244666

Dr C H BatchelorWater Resources Management Ltd,Apple Tree Cottage, Seymour Gate,Chipping Campden, Glos GL55 6BP, UKE-mail: wrmltd@aol.com.Telephone: + 44 (0) 1386 840230

Dr A J JamesEnvironmental & Natural Resource Economist,609B Hamilton Court, DLF City Phase 4,Gurgaon, Haryana 122 002, IndiaE-mail: ajjames@vsnl.net.Telephone: + 91 (0) 124 5051338

Dr R NagarajaNational Remote Sensing Agency,Balanagar, Hyderabad 500 037, IndiaEmail: nagaraja_r@nrsa.gov.in.Telephone: + 91 (0) 40 2388 4239

Dr J SeeleySchool of Development Studies,University of East Anglia, Norwich NR4 7TJ, UKE-mail: j.seeley@uea.ac.uk.Telephone: + 44 (0) 1603 451999

Dr J A ButterworthNatural Resources Institute,University of Greenwich, Chatham. Kent ME4 4TB, UKE-mail: j.a.butterworth@gre.ac.uk.Telephone: + 44 (0) 1634 883615

Copyright: Andhra Pradesh Rural Livelihoods Programme.Additional copies of the report may be downloaded from theAPRLP website (www.aplivelihoods.org)or WHiRL website (www.nri.org/WSS-IWRM/)

This publication was commissioned as part of theAndhra Pradesh Rural Livelihoods Programmeand undertaken by DFID-supported consultants.The views expressed in this publication do not necessarilyreflect official policies. This document has been producedwith the intention of promoting discussion between keystakeholders in the water sector in Andhra Pradesh.

Please reference the report as: Rama Mohan Rao, MS,Batchelor, CH, James, AJ, Nagaraja, R, Seeley, Jand Butterworth, JA. 2003. Andhra Pradesh RuralLivelihoods Programme Water Audit Report. APRLP,Rajendranagar, Hyderabad 500 030, India.

The photographs were takenby Charles Batchelor.

This publication has been producedby Write-Arm, Bangalore.

E-mail: writearm@vsnl.com.

Shri S RaySpecial Chief Secretary,

Panchayati Raj and Rural Development,Government of Andhra Pradesh

Foreword

The Water Audit has highlighted the need forgreater accuracy in official statistics heldby different line departments. As thesestatistics underpin decision-making at alllevels, increased effort is also required tomake water-related information accessibleto those who need it.

Finally, the APRLP Water Auditmakes recommendations for policies andinterventions that take a long-term approachto developing and managing water resources.These recommendations include practicalsuggestions for ways in which Andhra Pradeshcould introduce and pilot integrated waterresource management systems that have theaims of protecting drinking water supplies,promoting equitable access to water forproductive uses and enhancing the efficiencyand productivity of water use at all scales.Given that demand and competition for waterresources in Andhra Pradesh is almost certainlygoing to intensify in coming years, thecontribution of the APRLP Water Audit toongoing debates and policy formulation iswelcomed.

During the last 25 years, Andhra Pradesh hasbeen at the forefront of watershed developmentand other innovative programmes aimed attackling the challenges of poverty alleviation andenvironmental sustainability. The DFID-supportedAndhra Pradesh Rural Livelihoods Programme(APRLP) is providing the Andhra PradeshDepartment of Rural Development with anotherexcellent opportunity to develop and assess novelapproaches to helping rural communities andhouseholds overcome the complex constraints andimpediments that impact on all aspects of theirlivelihoods. The APRLP is also providing anopportunity to identify and build upon thesuccessful components of existing programmesand, where relevant, to rethink and modify thoseapproaches to development that have been lesssuccessful or that have resulted in unintendedconsequences.

The APRLP Water Audit working in partnershipwith relevant line departments has shown thatongoing water-related policies and programmeshave and are producing a range of benefits.However, the Water Audit has also identifiedsituations in which current programmes run therisk of being victims of their own success (e.g. incases in which intensive treatment of drainagelines is reducing tank inflows and the overallutility of traditional tank systems).

Contents

Executive summary 6

1 Introduction 141.1 Objectives of the APRLP 14

Water Audit1.2 Selection of study mandals 151.3 Institutional arrangements 16

2 Methodology 172.1 Background 172.2 Sources of information 182.3 Collection and quality control of 19

“terrestrial” physical data2.4 Analysis of remotely-sensed data 212.5 Collection and analysis of social 23

and institutional information2.6 Economic data collection and analysis 242.7 Water balance calculations 242.8 Runoff analysis 25

3 Physical characteristics of 27study mandals3.1 General information 273.2 Agro-climate 283.3 Geology and geomorphology 303.4 Soils 303.5 Drainage 323.6 Gully control structures 333.7 Tanks 333.8 Land capability 333.9 Land use 353.10 Vegetation 383.11 Land holdings 38

4 General resource status of 40Dhone and Kalyandurg4.1 Groundwater resources 404.2 Surface water resources 494.3 Impact of water harvesting on 51

patterns of water availability and use4.4 Annual water balances 56

5 Access to water resources 575.1 Status of rural water supplies 575.2 Water-related social discrimination 615.3 Water-related functionality of 62

village-level institutions

5.4 Functionality of SHGs 645.5 Participation of the poor and women 65

in community decision-making5.6 Access to irrigation water 695.7 Profitability of irrigated cropping 71

6 Constraints and risks 726.1 Water-related constraints 726.2 Water-related risks 74

7 Water resource planning and 76option selection7.1 Impacts of watershed development 767.2 Adaptive water resource 78

management7.3 Integrated water resource 78

management7.4 Water-related trade-offs and 79

win-win options

8 Recommendations 808.1 Summary of main 80

recommendations8.2 Recommendation 1 – 81

Wider range of options8.3 Recommendation 2 – 91

Targeting of options8.4 Recommendation 3 – 94

Village-level participatory planningwithin a wider District PlanningFramework (DPF)

8.5 Recommendation 4 – 95Shift from supply to demandmanagement of water resources

8.6 Recommendation 5 – 97Use of GIS-linked participatoryassessments as part of the M&E ofrural water supply programmes

8.7 Recommendation 6 – Reassessment 97of “official” water-relatedstatistics

9 Proposed follow-up activities 98

Acknowledgements 99References 100Glossary 101Abbreviations and Acronyms 103

Background

The Andhra Pradesh Rural Livelihoods Programme (APRLP) is beingimplemented in five southern districts of Andhra Pradesh, namely,Anantapur, Kurnool, Mahabubnagar, Nalgonda, and Prakasam.Taking watershed development as an entry point, the APRLP has theprime objective of developing effective and sustainable approaches toeliminating poverty in drought-prone areas of Andhra Pradesh.Livelihoods in these drought-prone areas are intimately linked with theavailability of water resources and, in particular, access and entitlementsto water for domestic and productive purposes. In recent years, there havebeen dramatic changes in the surface and sub-surface hydrology of theproject districts primarily as a result of inappropriate resource managementand increased groundwater extraction for irrigation. Some rivers that wereonce perennial sources of water are now dry for long periods of the year(e.g. Pennar River, Anantapur District); in-flows to many tanks are reduced;drinking water supplies in many towns and villages are becoming increasinglyunreliable; and, in general, the ability of livelihood systems to withstandthe shock of drought is deteriorating. Hence, it is in a context of demandfor water outstripping supply that the APRLP is operating.

Executive summary

6

APRLP Water Audit

The primary aims of the APRLP Water Auditwere to: 1) Assess the current status of waterresources in two selected mandals (i.e. sub-districts); 2) Evaluate water-related demandtrends in these mandals; 3) Study patterns ofwater-related access and entitlements of, inparticular, poorer social groups; 4) Considergender and social exclusion issues surroundingaccess and entitlements to water for bothdomestic and productive purposes; 5) Assess thefunctionality of water-related policies andinstitutions at the village, district and state levels;6) Identify improved resource managementpractices and policies; and, 7) Provide a practicalframework for more productive, sustainableand/or equitable use of water resources. Thisreport summarises the main findings andrecommendations of the Water Audit.

General Approach

The approach adopted by the APRLP WaterAudit built on experiences gained during a wateraudit that was carried out as part of the KarnatakaWatershed Development Project (KAWAD).Similar to the KAWAD study, the APRLP WaterAudit used as its starting point the fact that, inIndia, large amounts of physical, institutional andsocio-economic data have been and are beingcollected routinely in rural areas. Unfortunately,these secondary data are not always easilyaccessible and their quality is usually quitevariable. A major feature of the study was theconsolidation of data from a wide range ofsources onto a GIS database followinggroundtruthing, gap filling and quality control.Analysis focused on using water balancetechniques to assess current patterns of wateravailability and use as well as current and futuredemands for water. Information on access andentitlements to water, the functionality of village-level institutions and village-level views ondomestic water supplies were elicited using aparticipatory assessment technique modified tomeet water auditing information requirements.Albeit with specialist support from a number oforganisations and individuals, the bulk of theAudit’s work was carried out by relevantgovernment line departments and by the NGOsalready working in the Audit mandals.

Innovative methodologies

Although it was not a research project, theAPRLP Water Audit developed and refined a

number of methodologies some of which werefirst used during the KAWAD Water Audit.These included: 1) water-related participatoryassessments that produce outputs suitable forGIS analysis; 2) water auditing that combinesterrestrial and remotely-sensed data; 3) a simpleExcel-based modeling technique for assessingthe impact of water harvesting structures ondownstream water resource availability;4) decision trees that use social, and institutionalinformation along with physical information fortargeting project interventions and activities; and5) a simple GIS-based participatory assessmentmethodology for M&E of rural water supplies.

Characteristics of the“audit” mandals

The APRLP Water Audit was carried out intwo representative mandals, namely Dhone andKalyandurg. These mandals are located inKurnool and Anantapur Districts respectively.The total area of the two mandals is just under1,000 km and the total population isapproximately 200,000. The climate of bothmandals is semi-arid with potential evaporationexceeding rainfall in all but a few months in anyyear. Mean annual rainfall in Dhone andKalyandurg is 560 and 525 mm respectively.However, there is considerable inter- and intra-annual rainfall variability and droughts and yearsof relatively high rainfall are common. As a ruleof thumb, out of every ten years, five are droughtyears, two of which are severe and one of whichis catastrophic. At the time of writing this report,both mandals were experiencing a catastrophicdrought which was impacting severely onagricultural production and, in many villages, ondomestic water supplies. Kalyandurg is located inan area with predominantly red soils (alfisols) thatare underlain by crystalline basement geologies(e.g. granites and gneisses). Dhone is also locatedin an area with predominantly red soils with someareas underlain by crystalline basement and othersby sedimentary geologies (e.g. shales andlimestone).

Land Use

The main land use in both mandals is rainfedarable cropping. In general, a single rainfed cropis grown each year and around 80% of the rainfedarable area is under groundnuts. Crops grownare a mix of cash and subsistence crops, withmost farmers growing some of each, often in thesame field (e.g. the groundnut and red gram

7

intercropping that is popular in many areas).Although official statistics suggest much lowerfigures, compelling evidence (based on remotesensing and survey data collected by the AndhraPradesh Groundwater Department) points toDhone and Kalyandurg having around 8% and12% of net land area under irrigation respectively.Or, put more succinctly, all the indications arethat the actual net irrigated area is 4-5 timeshigher than the official figures.

In the last 10-15 years, there has been adramatic increase in groundwater-based irrigation.Even in the command areas of tanks, mostfarmers have shifted from using tank releases forirrigation to using pumped groundwater. In mostcases, the tank sluices have been blocked, therebyconverting the tanks into percolation tanks.Access to irrigation enables farmers to grow morethan one crop regardless of rainfall conditions.This said, frequent power cuts and, in some areas,rapidly increasing competition for groundwaterare putting limits on the total area underirrigation. Good returns are possible whereirrigation is feasible and access to irrigable land isan important determinant of the level andsecurity of livelihoods amongst the ruralpopulation.

Around 20% of Dhone is under forest, muchof which is highly degraded, whereas less than 5%of Kalyandurg is under forest. Although precisestatistics are not available, indications are that,with the exception of poor quality land, commonlands in both mandals have been heavilyencroached.

Profits per hectare from irrigated agricultureare, on average, twice that from rainfed farming:the lowest profit from an irrigated crop is oftenhigher than the highest profit from a rain fedcrop. Cultivation of some crops, however, is noteconomic, and farmers earn income largelybecause they do not have to buy the inputs theyrequire. Although the large kharif areas underrainfed groundnut are justified by the high relativeprofits, areas under other crops do not generallycorrelate with the net returns per hectare.Crops with high returns (e.g. mulberry, onions,vegetables) are grown on comparatively smallerareas because of local factors such as access tomarket, high cultivation costs, production risksand lack of local storage and processing facilities.Food crops (e.g. jowar), on the other hand, aregrown on larger areas than might be expected,given their relative profitability. This can beexplained by the decision-making of poorerfarmers that is geared towards maintaininghousehold food security.

Investment in borewells

Failed borewell investments as a result ofgroundwater depletion have become an importantcause of indebtedness and poverty. Growinginequity in access to groundwater is also fuelinga process of social differentiation which impactsdirectly on the livelihoods of some groups andcontributes to the consolidation of power relationswithin communities.

Tanks and other water bodies

Although there are few natural open waterbodies, there are 94 tanks in the two mandalssome of which date back several hundred years.Many tanks have been abandoned or are in a stateof disrepair with broken bunds and silted bedswhich are now cropped. The majority of the tanksthat are still functioning, have been converted topercolation tanks such that irrigation in commandareas relies on groundwater instead of surfacewater releases from the tank. Inflows to manytanks have declined in recent years as a result ofincreased water harvesting and groundwaterextraction in the tank catchment areas. In manycases, this has had a severe impact on the utility,biodiversity and cultural value of the tanks. Inextreme cases, reduced tank inflows have adverselyaffected the reliability of domestic water supplies.Such severe negative impacts occur when tanks arean important source of recharge for the aquifersused for urban supply and when rainfall is belowaverage.Masonry check dam

8

General Resource Statusof Project Watersheds

Rainfall

Although a widely-held view is that averageannual rainfall has declined in dry areas ofsouth-western Andhra Pradesh, this view is notsupported by analysis of rainfall records. There is,however, an indication that there might be smallseasonal shifts and a small increase in rainfallvariability.

Groundwater

In both Dhone and Kalyandurg, there has beena dramatic increase in groundwater extraction forirrigation during the last 10 - 15 years. As a result,groundwater levels have fallen and, in Kalyandurgin particular, shallow wells have failed as deepborewells have been constructed and as extractionfrom deeper aquifers has become the norm.Although the number and density of wells inKalyandurg and Dhone are similar, levels ofgroundwater extraction are around 30% higherin Kalyandurg as compared to Dhone. Thisdifference helps to explain why 30% of the wellsin Kalyandurg are completely defunct or failroutinely. The figure for well failure in Dhone isaround 8%. Well surveys by the district-levelAndhra Pradesh Groundwater Departmentshowed that the actual number of wells in eachmandal is nearly double the official record.This finding has major implications for estimatesof the stage of groundwater development that arebased on well statistics (e.g. the GEC-97Methodology).

Net groundwater extraction for irrigation,domestic and livestock use for Kalyandurg andDhone was estimated at 11.0% and 8.4% ofmean annual rainfall respectively. As the AndhraPradesh Groundwater Department’s estimateof groundwater recharge in this area isapproximately 10% of annual rainfall, thissuggests that current levels of extraction inKalyandurg are not sustainable. Analysis ofvillagewise groundwater extraction showed largedifferences between villages in both mandals.These can be attributed in part to more favourablehydro-geological conditions in some villages andin part to variations in the historical pattern ofdevelopment of both ground and surface waterresources. This finding illustrates the fact thatthere is neither equitable access to irrigationbetween villages nor between households withinvillages.

Data from the automatic ground water levelrecorder installed in Kalyandurg as part of theNational Hydrology Project do not correlatewell with the findings of the Andhra PradeshGroundwater Department’s survey in thismandal. This can be explained by the fact thatthe recorder is sited in an area with a relativelylow level of groundwater extraction. Also theactual location of this recorder (as with manyothers) was chosen primarily on the basis ofsecurity. The broader implication is that automaticwater-level recorder information should be used

Villagers collecting water from an agricultural well asa result of failure of domestic wells, Dhone

9

with extreme caution when assessing regionalimpacts of, say, watershed development ongroundwater status.

Fluoride levels in groundwater in excess ofpermissible limits were found in many drinkingwater sources in both mandals. The Bureau ofIndian Standards set a maximum permissiblefluoride concentration in domestic water of1.0 ppm. However, concentrations of up to1.5 ppm are considered to be acceptable in theabsence of an alternative safer source. Although areasonable level of awareness of fluoride problemsexists at the village and district levels, a laxnesswas noted in the way in which fluoridepermissible limits are being used by departmentsresponsible for drinking water supplies. Aconsequence is that action is not taken even whenfluoride concentrations are well in excess of1.5 ppm. A subsequent, more detailed fluoridesurvey that was carried out by the WHiRL Projectin some villages in Kalyandurg indicated thatfluoride concentrations in domestic water sourcesare being under-reported in official statistics. Thisstudy also showed an average increase in fluorideconcentrations of around 30% in domestic watersources during the 2002/2003 drought.

The overall conclusion of the “groundwater”component of the Audit is that the scope fordeveloping additional groundwater resources inboth mandals is limited and, in much ofKalyandurg, groundwater is already severely over-exploited. This conclusion does not agree with arecent statewide groundwater survey carried outby the Andhra Pradesh GroundwaterDepartment. This difference of opinion can beexplained by the fact that the statewide surveyused “official” figures for irrigated area and wellnumbers that appear to hugely underestimate thesituation on the ground.

Surface Runoff

Data from gauging stations operated by theCentral Water Commission indicate that annualsurface runoff at the large watershed scale inAnantapur and Kurnool is somewhat lower thanis often reported or than accepted wisdom wouldsuggest. Although there is large inter-annualvariation, average annual runoff as a percentage ofrainfall is in the range 1% to 8% for theChinnahagari, Pennar, Hundri, Chitravati andVadavathi rivers. Although runoff for individualor sequences of rainfall events is often higher, as isrunoff at the plot and field scale, this findingindicates that, in the absence of inter-basin

transfers, the scope for augmenting waterresources in these mandals by damming, divertingor pumping water from local rivers is quitelimited. The low values of runoff are notsurprising given the physical characteristics of theregion and the large number of tanks, check damsand nala bunds that existed before the APRLPstarted watershed development work in thesedistricts.

Status of domestic water supplies

Participatory assessments were made of thestatus of the 225 and 438 domestic water points(hand pumps and public taps) in Dhone andKalyandurg respectively. The following factorswere considered when classifying each water pointas being satisfactory or as having a problem:functionality (i.e. no technical problems),distance to water point (i.e. less than 1.6 km),crowding (i.e. less than 250 people using thewater point), adequacy of supply (i.e. 40 lpcdavailable 365 days/year), peak summer availability(i.e. similar time and effort needed to collectwater in peak summer), accessibility (i.e. no socialexclusion), and water quality (i.e. acceptable fromusers’ viewpoint). These factors and permissiblelimits are similar to those used by the RajivGandhi National Drinking Water Mission. Therather startling finding was that 51% and 24% ofdomestic water points in Kalyandurg and Dhonerespectively were classified, by the users, as havinga problem. In addition to showing the widevariety of problems faced by domestic water users,these figures compare starkly with officialstatistics which suggest that, at any one time,there are few problem water points in thesemandals.

Participation of the poor andwomen in meetings

Assessment of the participation of the poor invillage-level meetings showed that, in Kalyandurg,the poor were much more likely to attendmeetings and influence decisions affecting themthan was the case in Dhone. The high level ofempowerment of the poor reflects well on thework of an NGO, namely the RuralDevelopment Trust, which has been working inthis mandal for more than 25 years. Lessencouraging was the fact that the participation ofwomen in meetings appears to be weak in bothmandals. This finding implies that a lot of effortwill be required if women are to be become,under APRLP and similar programmes, effective“agents of change”.

10

Self-Help Groups

Participatory assessment of the water-relatedfunctionality of self-help groups showed that,while they have been successful at empoweringmembers and in thrift and credit activities, theyhave little influence on or interest in water-relateddecision making.

Performance ofGram Panchayats

Participatory assessment of the water-relatedperformance of panchayats showed that there isconsiderable variability in their effectiveness.Interestingly, an inverse relationship betweenpanchayat effectiveness and community actionwas observed in that community action was moreprevalent wherever panchayats were weak and/orineffective.

Social discrimination

Water-related social exclusion is a deeply-ingrained fact of life in many villages in thesemandals. Although the government has installedseparate water points for different social groups inan attempt to minimise this problem,discrimination persists, especially when waterpoints go dry during peak summer months. Thediscrimination identified in participatoryassessments falls into two main categories. Firstly,prohibition on Scheduled Castes and ScheduledTribes directly accessing water from open wellsand, secondly, forced preference of ‘upper’ castesat water points meant for ‘lower’ castes.

Information from the participatoryassessments also showed that, in many villages,a large amount of additional effort is requiredto collect even reduced amounts of water duringpeak summer. The additional effort and associateddrudgery, in the form of longer queuing times andgreater distances walked, has a disproportionateimpact on the lives of women and children as inmost households they take primary responsibilityfor fetching and carrying water from water points.

Challenges and constraints

As a sustainable livelihoods project, theAPRLP is attempting to develop and promotepolicies and practices that meet the sometimesconflicting challenges of improving productivity,sustainability and equity. Key challenges includeidentifying and promoting resource managementpractices and policies such that:

· There is an overall increase in productionin project watersheds and a concurrentimprovement in the livelihood assets ofpoorer social groupings;

· Resource extraction and use do not lowerthe future availability of resourceendowments (e.g. as a result of deterioratingwater quality, increased groundwaterextraction in the areas in which over-exploitation is already taking place,upstream development of resources at theexpense of downstream users or vice versa);

· Access, entitlements and patterns ofresource extraction meet basic needsfor drinking water as well as providingopportunities for land-based and non-land-based income generation;

· Vulnerability of the poor is diminished;in particular, their vulnerability to externalshocks to their livelihoods, due to theenvironment (e.g. droughts and floods)and due to social and political change.

Surveying Battuvani Palli tank, Kalyandurg

11

Raising awareness at all levels of the real natureof water-related challenges in the region needs tobe given a high priority. Current watersheddevelopment publicity is often misleading in thatit suggests that there are quick fixes to water-related problems in semi-arid areas (e.g. check damconstruction, contour bunding and tree planting).While these activities have an important role toplay, on their own they can only have a limitedand mainly localised impact. More importantly,when used inappropriately, these activities cancause negative impacts that may be socially,politically and/or environmentally unacceptable.Unlike higher rainfall areas, the semi-arid areas ofsouthern Andhra Pradesh now only have limitedadditional resources that can be developed.The challenge, therefore, is to make better useof existing water resources bearing in mind thatthe majority of changes in resource use ormanagement involve negative trade-offs.

Recommendations

The main recommendations of the WaterAudit are as follows:

1. Wider range of options. Although theAudit findings suggest a rather gloomystate of affairs, one positive conclusion isthat there are a large number of watermanagement options that could bepromoted by APRLP (over 40 have beenidentified and listed in the report).All these options have the potential toincrease the water use productivityand/or to improve equitable access to waterresources. The fundamental need is toconsider the trade-offs associated with eachoption (or sequence of options) and toselect options that maximise the social andeconomic benefits. In most cases, thismeans giving domestic water supplies thehighest priority and then allocating waterto uses that have the next highest socialand economic value.

2. Targeting of options. In general, watersheddevelopment programmes and similarinitiatives do not target activities toappropriate physical, social andinstitutional settings. Or put another way,such programmes tend to use a “one-sizefits all” approach. An alternative approach,which is described in the report, is to usedecision-trees to match interventions andactivities to appropriate settings.

This approach can and should use acombination of physical, social, economicand institutional data in the decision-making process. Although potentially timeconsuming, this approach can becomesimple and rapid if data are readilyavailable (e.g. once a water audit has beencompleted and a GIS database created)and, preferably, once the approach isincorporated into a more generalmanagement information system.

3. Village-level participatory planning within awider District Planning Framework. The aimof proposed District Planning Frameworksshould be to tackle issues not covered invillage-level planning (e.g. upstream-downstream equity, sustainability ofdomestic water supplies, protection of rarehabitats, drought-proofing, pollutioncontrol etc.) and, ideally, they should bebased on principles of integrated andadaptive water resource management.Ideally also, consideration should be givento adopting and adapting novel approachesto integrated water resource planningand management that are currently beingpioneered in South Africa and arerecommended by the Global WaterPartnership and others. This recommendationmay seem to many like a retrograde returnto top-down master planning, however,

Livestock around Chapari tank, Kalyandurg

12

“official” statistics and the reality on theground with regard to area underirrigation, well statistics, stage ofgroundwater development and status ofdomestic water points. Hence it isrecommended that further checks onrelevant official statistics be carried out,possibly using procedures used by theWater Audit team. This work is neededurgently because these statistics underpinpolicies relating to: groundwaterdevelopment, NABARD loans, power sectorreform, protection of rural water supplies,river basin management and irrigationdevelopment. Findings from the Audit havealso shown that certain beliefs thatunderpin the watershed developmentprogrammes (e.g. annual runoff is 30-40%of annual rainfall) need to be re-evaluated.

Concluding Remarks

The Water Audit has shown that in the studyareas demand for water is outstripping supplyand that scope for augmentation is limited.Although the water resource situation in this areais extremely serious and arguably at crisis pointfor many villages, there is much that can be done.However, it cannot be stressed enough that thereare no quick fixes to the complex challengesfacing people in the semi-arid areas of AndhraPradesh. Policies and practices are needed that arebased on accurate information and that seeklong-term solutions. Finally, it is stronglyrecommended that a plan be developed andimplemented for the capacity building that isneeded at all levels if effective integrated waterresource management policies and programmesare to be developed and implemented in AndhraPradesh.

it is clear from this study that waterresource planning and management isneeded at levels above the village if manylarge scale equity and sustainabilitychallenges are to be addressed adequately.

4. Shift from supply to demand management ofwater resources. The results of the studyshow clearly that the focus of the APRLPand similar programmes should be onresource management as opposed toresource development. This said, demandmanagement is not going to be a panaceaand its introduction has many potentialunintended consequences, some of whichcould impact negatively on poorer socialgroupings. A whole range of demandmanagement options are listed in thisreport. One of the most important optionsrelates to the essential need to create thepolicy and legislative environment thatprovide incentives for more productiveuse of water by individual users anddisincentives for practices that are wastefulor lead to environmental degradation.However, changing current attitudes andbehaviour is not going to be an easy task,particularly as many recent and currentstate-level policies (e.g. grants for wellconstruction, subsidised electricity forpumping irrigation water, support pricesfor paddy rice) have had the unintendedconsequence of encouraging inefficientand inequitable use of water.

5. Use of GIS-linked rapid participatoryassessments as part of the M&E of rural watersupply programmes. The Water Audit hasrevealed a major disparity between officialstatistics and the users’ view of the statusof domestic water supplies. One reason forthis disparity is the strong emphasis ofcurrent M&E on the functioning ofinfrastructure and the meeting of supplynorms (i.e. 40 lpcd). Many of the otherproblems faced by users are ignored by theM&E systems. However, these additionalproblems can be considered, andappropriate steps can be taken if routineM&E systemstake users’ views into account.Incorporating and presenting thisinformation using a GIS database is oneway of processing and presenting therelevant information in a form that caneasily be assimilated and acted upon bydecision makers at all levels.

6. Reassessment of “official” water-relatedstatistics. Finally, survey work carried outprimarily by government departments hasshown huge discrepancies between Lifting a rainfed groundnut crop in Kalyandurg

13

1.1 Objectives of APRLP Water Audit

The APRLP Water Audit commenced inNovember 2000 with a series of inception workshopsthat provided the main stakeholders with anopportunity to discuss the Audit’s specific objectives,methodology and institutional arrangements.The agreed wider objectives of activities in the studyareas were:

i. To assess the status of water and othernatural resources at scales ranging from themicro-watershed to the macro-watershed,thereby, to inform decision-making andpolicy formulation at all levels in theAPRLP and in Andhra Pradesh’s watersheddevelopment programme.

ii. To assess pressures on groundwater and surfacewater resources and current trends in use anddemand (e.g. water demands for domesticpurposes, irrigation, non-land based activities).

iii.To assess the relative value of different wateruses in terms of productivity, equity andbasic human and environmental needs.

iv. Based on the above, to identify resourcemanagement practices and policies that areeconomically viable and that have the potentialto bring about more equitable, sustainable andproductive use of water in the long term.

v. To provide baseline data against which someresource-related indicators can be monitored.

vi. To build capacity of DPAP, DDP and NGO staff inthe APRLP districts so that they have theskills and confidence to undertake and updateresource audits.

Introduction1

14

Figure 1. Location of APRLP and KAWADWater Audit study areas

1.2 Selection of studymandals

The APRLP Water Audit wascarried out in two mandals, namely,Dhone mandal in Kurnool districtand Kalyandurg mandal inAnantapur district (see Figure 1).Criteria for selecting these mandalsincluded representativeness and thepotential availability of NGOsupport. Final selection was madefrom short lists provided by theDPAP and DDP Project Directors.The mandal scale (i.e. approximately400-500 km2) was selected because itwas felt that this would providefindings that would give a goodinsight into issues of scale (e.g.upstream-downstream competitionfor water). It was decided also thatthis would be a good scale to work atwhilst training and capacity buildinghad a high priority and whilst newdata collection procedures werebeing piloted.

Administrative units, rather thanphysical units (i.e. watersheds), werechosen as the main units for carryingout the Water Audit because muchof the data are available on avillagewise basis and much resource-related management and decision-making has to take place on the basisof administrative units. However,relevant analysis was carried outusing physical units whenevernecessary (e.g. analysis of surfacerunoff ).

Kocheruvu village, Dhone

15

Anantapur District

KurnoolDistrict

DoddahallaWatershed

UpparahallaWatershed

ChinnahagariWatershed

Dhone Mandal

Kalyandurg Mandal

ANDHRA PRADESH

Hyderabad

KARNATAKA

Bangalore

TAMILNADU

Chennai

MAHARASHTRA

Mumbai

KAWAD watersheds

APRLP water audit study

KERALA

Andhra Pradesh

India

1.3 Institutional arrangements

The APRLP began in the second half of 2000with the establishment of a Project Support Unitat the state level and District Capacity-BuildingCentres in each of the five districts1. Given theAPRLP’s emphasis on capacity building, it wasdecided that the Water Audit would beimplemented by GoAP line departments withsupport being provided by a specialist supportteam (see Figure 2). The logic was that activeinvolvement in the Water Audit of, in particular,district-level staff would build their capacity sothat they would be able to scale up the waterauditing to other mandals and districts. However,as will be discussed, this approach did not proveto be entirely successful and the bulk of the work,particularly during the latter stages of the Audit,was carried out by the specialist support team.This team was comprised of staff from theCentral Soil and Water Conservation Researchand Training Institute (Bellary), staff from theLand Use Section of the National Remote SensingAgency, Dr Snehalatha and Mr Sundar Raman(both of whom subsequently joined APRLPDistrict Capacity-Building Centres), Dr J Seeley(University of East Anglia), Dr J Butterworth(Natural Resources Institute), Dr A J James(Environmental Economics consultant) andDr C H Batchelor (Hydrological consultant).

Figure 2. Institutional arrangement of theAPRLP Water Audit

1 Up-to-date information on the programme can befound on: www.aplivelihoods.org

Dry season paddy cultivation in Dhone

Buffalo wallowing in urban waste water, Kalyandurg

State Level

Mandal Level

District Level

Kurnool WRAField Teams

Departmentof Rural

Development

WRA SpecialistSupport Team

Anantapur DPAPAudit Team

State-levelorganisations

Relevantdonor-fundedprogrammes

Kurnool DPAPAudit Team

APRLP SupportUnit

Anantapur WRAField Team

16

2.1 Background

Water audits, under various different names, are beingpromoted increasingly as a key step towards effective andsustainable integrated water resource management. For example,the International Water Management Institute (IWMI) hastaken a lead in advancing the case for water accounting and indeveloping relevant definitions and procedures (Molden, 1997;Molden et al, 2001; IWMI, 2002). Similarly, the Global WaterPartnership (GWP) has stressed the importance of water resourceassessments as part of integrated water resource management(GWP, 2000). Although there are some subtle differencesbetween the methodologies that are being promoted by differentorganisations, the overall objectives of the different approachesare similar (see Box 1).

Methodology2

17

Box 1. Why carry out a water audit?Because a water audit can:

· Identify the current status of waterresources at different scales andtrends in demand and use;

· Provide information on access andentitlements to water and thetrade-offs that have resulted or willresult from different patterns ofwater use;

· Provide information on social andinstitutional factors affecting accessto water and reliability of watersupplies;

· Help identify externalities whichonly become apparent when thepatterns of water use are consideredat the macro temporal and spatialscales;

· Provide information that is requiredfor assessing efficacy of existingwater-related policies;

· Identify opportunities for savingor making more productive and/orequitable use of water;

· Identify the effectiveness of currentdrought and flood coping strategies;

· Identify potential problems resultingfrom competing or multiple uses ofwater;

· Assess the accuracy of governmentstatistics;

· Identify the extent to which decision-making is based on hydrologicalmyths or misconceptions.

The concept of water auditing is based onthe argument that knowledge of the current statusof water resources and trends in demand and use isa precondition for successful water management.Equally important, an understanding of factorsaffecting patterns of access and entitlement towater resources is fundamental in any projects thatseek to improve and protect the livelihoods ofpoorer social groups. Effective water auditingimplies a holistic view of the water resourcessituation and its interaction with societal use. Thisincludes: 1) addressing the occurrence of surfaceand ground water, in space and time, and, inparticular, assessing levels of sustainable use andthe frequency of extreme events such as droughtsand floods; 2) providing a tentative assessmentof the demand trends for different uses;3) identifying the main driving forces influencingdemand and use (e.g. government policy, societalbehaviour); 4) assessing the functionality andeffectiveness of institutions charged withdeveloping and managing water resources; and,5) understanding factors that affect access andentitlements to water for both domestic andproductive uses.

2.2 Sources of information

Although the Water Audit used quality-controlled secondary information whereverpossible, primary data collection was carried outto fill gaps and to collect additional data.Agreement on the need for and the methods ofcollecting these additional data was reached duringthree inception workshops. Compared to manycountries, relatively large quantities ofhydrological, geological, agricultural, social andother information is collected routinely in Indiaat the national and state levels by governmentand non-government organisations. Unfortunately,these data are not always easily accessible orutilisable for reasons that include:

· Data are fragmented in that they are heldby different organisations and, in somecases, by different departments orindividuals within these organisations;

· Spatial and non-spatial data are storedin a wide range of formats (e.g. maps,remotely-sensed images, tables of figures,text, graphs, etc.) and media (e.g. in yearbooks, on computer disks etc.);

· Spatial and temporal scales at which datahave been collected are not at all consistent;

· Data quality is extremely variable.Survey team working with a villager from

Lakshmipalli, Dhone

18

Box 2. APRLP Water Audit designphilosophy

1. Make maximum use ofsecondary baseline information(e.g. soil maps, remotely-senseddata, etc.) and resource monitoringinformation (e.g. rainfall records,groundwater levels, agriculturalstatistics etc.) that has been oris being collected by governmentline departments and otherorganisations;

2. Adopt an approach that buildson experience gained during theKAWAD Water Audit and, inparticular, include systematiccollection of social andinstitutional data;

3. Use GIS software to consolidatespatial information and, wherenecessary, reconcile differencesbetween administrative andphysical boundaries;

4. Ensure maximum involvement ofspecialists that are either based inthe APRLP districts or have longexperience of working in this area;

5. Encourage the active involvementof GoAP line department staff.

In Kalyandurg, line department staff were notreadily available, hence field data collection wascarried out primarily by field teams assembled byMr Sundar Raman (Anantapur Water Resource Auditcoordinator) or by CSWCRTI staff. Each field teamconsisted of an Engineering Diploma holder, a fieldassistant and a local villager who was preferably thevillage thalarry or etti. Assistant geologists of theGround Water Department under the guidance ofMr G V Reddy (Deputy Director, Andhra PradeshGroundwater Department) carried out the wellsurvey. In Dhone, field data collection was carried outprimarily by line department staff with CSWCRTIstaff undertaking some specialist tasks. Thisarrangement, although potentially good for capacitybuilding, led to many delays (see Box 3).

Figure 3. General procedure for collecting, qualitycontrolling and processing information

Some cadastral maps of the villages were availablein Kurnool and Anantapur whereas others had to bepurchased from the Survey of India in Hyderabad.Photocopies of relevant cadastral maps were used bysurvey teams as base maps for marking the location ofpoint features such as wells and check dams. At thesame time, GPS handsets (Garmin 12XL, USA) wereused to record the latitude and longitude of thesefeatures.

2.3 Collection and qualitycontrol of “terrestrial”physical data

A major feature of the APRLP WaterAudit was the consolidation of spatial andnon-spatial data from a wide range of sourcesonto geographical information system(GIS) databases (see Figure 3). Somegroundtruthing and gap filling was carriedout during the collection process and furtherquality control checks were carried out oncethe database was established. A major partof the quality control process was the inter-comparison of data and statistics fromdifferent sources and analysis and discussionsaimed at understanding the reasons fordisparities. This is arguably the key step ina water audit as it also involves assessingwhether data support accepted wisdomrelating to the development and managementof water resources.

Produce base maps

Groundtruth and reconcile data

Enter data into GIS database

Data analysis

Consultation and identification ofoptions/strategies

Information made available

Produceremotely-sensedthematic maps

Collect spatialand non-spatial

data

Collect social,economic and

institutional data

19

Box 3. Lessons learnt during the information collection and groundtruthing phase

Many delays were experienced during the collection and groundtruthing of terrestrialdata. Reasons (in approximate order of importance) included:

· Delays in transfer of APRLP funds to the field teams. These funds were needed to pay fortransport, DAs, fieldwork equipment etc. As the Water Audit was one of the first APRLPfield activities, the modalities of getting funds from DFID to GoI to GoAP to theDRD and then right down to the field teams had not been tested previously;

· Delays in letting the contract for processing remotely-sensed data and digitising spatialterrestrial data. Internal problems affecting APSRAC meant the contract had to beswitched to NRSA. Delays were then caused because, at that time, APARD did nothave the authority to let contracts;

· Delays caused by the workload of DPAP/DDP staff. These staff can be particularly busywhen central funds are released. It should be noted also that Water Audit activitieswere given a relatively low priority by some individuals and groups;

· Availability and accessibility of secondary data. In some district-level departmentscomplete sets of maps, reports, non-spatial data are readily available but notin others;

· Perception that the Audit requires only primary data to be collected. Convincing some teammembers that the aim of the Audit was to groundtruth and use secondary datawherever feasible was not successful;

· Recruitment of a person to work in Anantapur and to take responsibility for the Audit.The recruitment of Mr Sundar Raman delayed work in Anantapur. Further delayswere then caused by Mr Sundar Raman being allocated regular DDP tasks;

· Delays in the purchase of 2 PCs and GPS equipment. This problem was also linked toteething problems related to the transfer of APRLP funds.

Although this list may give a rather negative impression, the positive message is that decisionsand actions taken at all levels meant that ways were found around all the problems listed. A lotof experience was gained that will help any follow-up audits and, possibly, the implementationof other APRLP activities at the district and mandal levels. This said, it is recommended thatdata collection and groundtruthing in any further audits is either carried out by line departmentstaff who are relieved from other duties until the work is finished or contracted out to theprivate sector.

and condition noted. Regarding the tanks orpercolation tanks, the conditions of bund, sluice,weir and feeder channel were collected during thesurvey along with year of construction, cost,potential command area, area irrigated during theyear 1999-2000 and water spread area. It waspossible to collect some of this information fromthe Panchayat Raj and Minor IrrigationDepartments. For each revenue village the timetaken was between 5 to 10 days based on the sizeof the village. A similar procedure was followed inDhone.

In Kalyandurg, the gully survey teams walkedalong each and every gully collecting data relatingto the gully dimensions every 200-300m. Foreach survey point, the latitude and longitude wererecorded, the point was marked on the relevantcadastral map and the nearest survey number wasrecorded on the relevant proformas. Thecondition of each gully such as whether it wasunder cultivation, levelled or flat was alsorecorded. All the water harvesting structures suchas masonry check dams, earthen bund or rock filldams were also surveyed and their effectiveness

20

scale were digitised and mosaics of these werecreated to extract the village, cadastral andmandal maps. Well locations, drainages patterns,check dams, gullies and embankments were alsodigitised from cadastral maps which had beenannotated by the field teams. Watersheds, sub-watersheds, mini-watersheds and micro-watersheds were delineated and digitised usingthe Survey of India topographical maps.

The land use/land cover themes were preparedusing the following data analysis procedures.Onscreen interpretation (based on tone/colour,texture, pattern, shape, size, location, shadow,association) of satellite image and its digitisationwas performed using Erdas Imagine Software.Before interpretation and analysis of the satelliteimagery, the satellite image was rectified andresampled using a 1st order polynomialtransformation model and a bilinear interpolationalgorithm respectively. To perform the same, theground control points were taken fromtopographical maps and used with relevant GPSdata sets to achieve sub-meter RMS accuracy. Forbetter interpretation, high resolution capability ofPAN data and multi-spectral advantage of LISS-III data were fused using Brovey TransformationModel and Bilinear Interpolation resamplingalgorithm to generate the merged image.This merged image was used for interpretation/analysis. The steps taken are presentedschematically in Figure 4.

For the well surveys, staff of the AndhraPradesh Groundwater Department (APGWD)with the help of villagers visited all the wells in thevillage area and collected data that included: typeof well, use of well, dimensions of well, year ofconstruction, area irrigated, pump capacity anddischarge of well, reliability of well, latitude andlongitude and survey number. Water sampleswere collected from each well and sent to theCSWCRTI or APGWD laboratories in Kurnoolin the cases of Kalyandurg and Dhone respectively.Follow-up water sampling and fluoride analysiswere carried out in four villages in Kalyandurgby staff involved in the WHiRL Project.

In the main, cropping statistics were collectedfrom the Mandal Revenue officer. Details of landholdings were collected from the NationalInformatics Centre. Rainfall data were collectedfrom the Andhra Pradesh GroundwaterDepartment and Chief Planning Officers inKurnool and Anantapur.

2.4 Analysis ofremotely-sensed data

Cloud-free satellite data of February-March2001 of Indian Remote Sensing Satellite (IRS)-1C/1D (PAN and LISS-III) were used as the mainsource of remotely-sensed data. Once analysed,these data were reconciled with the relevantterrestrial spatial data. Final mapping was done on1:50,000 scale. Cadastral maps at the 1:8,000

Village meeting in Battuvani palli, Kalyandurg

21

Eight digit alphanumeric symbolic codes wereallocated for all polygons by adopting analphanumeric system as per the Natural ResourceInformation System (NRIS) guidelines.

The field data and GPS collected data werelinked to the respective spatial database of Recordof Right (ROR) and relevant field data. To linknon-spatial data to the spatial database, theprimary database key (linking code) for wells,check dams, gullies and embankments wasdefined with respect to mandal villages. Afterdefining the primary key, the non-spatial databasewas linked using GIS database relational utility.

Watersheds were delineated using thedelineation procedure adopted by All India Soiland Land Use Survey (AIS & LUS) and NationalRemote Sensing Agency (NRSA) for the projecton wastelands mapping of India. The coding usedan 8-stage hierarchical approach starting fromwater resource region (average size 550 lakh ha.),basin (95 lakh ha.), catchment (30 lakh ha.), sub-catchment (7 lakh ha.), watershed (1 lakh ha.),sub-watershed (0.15 lakh ha.), mini-watershed(0.05 lakh ha.), micro-watershed (0.005 lakh ha.).The delineation of different boundaries wascarried out by considering factors such as rivermorphology, drainage and slope/elevation.

22

Figure 4. Procedure for processing and reconcilingspatial data

Ancillary data(Field data, GPS data etc)

Cadastral maps, otherresource maps,topomaps etc

Digitisation, editing, labeling,mosaicing and data linking etc

Collateral dataSatellite data

IRS PAN IRS-IC/ID LISS-III

Geometric Correction

Enhancement

Data merging (PAN+LISS-III)

Image interpretation

Land use/Cover map

Outcropsmap

Cropmap

Urban and ruralarea map

Rivers andtanks map

Data Base

Spot height

mapContour

mapWatershed

mapDrainage

mapBasemap

Lineamentsmap

Capabilitymap

Groundwater map

Soilmap

Geologymap

Mandalmap

Villagesmap

Cadastralmap

Wellsmap

Gulliesmap

Checkdams map

Maps derived from satellite imagesMaps digitised from other resource maps

Vegetationmap

Box 4. Lessons learnt using GPShandsets

GPS handsets were used by fieldteams as one of the methods fordetermining the location offeatures such as check dams andwells. Unfortunately, the datacollected proved to beinadequate for most Auditpurposes. Primarily, this wasbecause the system’s“antispoofing” system wasoperative during the WaterAudit. Consequently, theaccuracy of the GPS handsetreadings was of the orderof ± 10-20 m. Shortage ofhandsets also resulted in theKalyandurg team using GPShandsets produced by differentmanufacturers. This introducedadditional uncertainty in themeasurements. It isrecommended in future thataudits use differential GPS toimprove the accuracy ofmeasurements and equipment ofthe same specification purchasedfrom only one manufacturer.

2.5 Collection and analysisof social and institutionalinformation

Quantified Participatory Assessment(QPA) was used as the main method ofcollecting water-related social andinstitutional information. This methodcollects qualitative information usingconventional rapid-rural appraisal tools(e.g. transect walks and focus groupdiscussions) and uses descriptive ordinalscoring to convert this information intonumbers. In both mandals, followingpre-testing and training, a field teamcomprising both men and women spentone day in each village. After acommunity-leaders’ meeting in themorning, the field team of 6-12 peoplesplit up and undertook group work.In the smaller groups, villagers undertooktasks that included drawing a detailedsocial map (marking all poor and non-poor households, and water points in thehabitation); assessing the functionalityof all SHGs in the village; filling in a

questionnaire on household water use; and, confirmingthe accuracy or otherwise of the information providedby the community leaders. At the end of the day, theentire team presented the findings at a Gram Sabha,where the purpose of the assessment was re-iterated andthe collected information was verified.

Reasons for choosing this form of participatoryassessment included:

· Large sample size: All 100 habitations of Dhoneand Kalyandurg mandals had to be covered in ashort space of time (i.e. 3 months) at anacceptable cost. A low cost was required becauseit was envisaged that assessment would berepeated as part of the APRLP’s M&E programme.

· Both qualitative and quantitative informationwere required.

· GIS-compatible information was needed:The qualitative and quantitative informationcollected were to be combined in the GIS databasewith the more “technical” information that wasbeing collected by other people and groupsinvolved in the Water Audit.

Following the fieldwork, numerical information wasentered either into Excel spreadsheets or into an Accessdatabase and supporting qualitative information wasrecorded in a village report. Considerable discussionthen went into developing different formats forpresenting the data as part of GIS layouts andincorporating it in the development of decision-makingtools such as decisions trees.

Using GPS equipment to mark the location ofvillage boundary stone, Dhone

23

Box 5. Lessons learnt during fieldwork

It would have been better if the gullysurvey, well survey and other “technical”fieldwork had been carried out inconjunction with the QPA surveys.This would have improved theparticipation and awareness at thevillage level and generated higher-quality information. Although thishad been the original plan, logisticalproblems and delays in the transfer offunds resulted in a fragmented approachbeing taken. However it is stronglyrecommended that a combinedcoordinated approach be used in anyfurther audits.

2.6 Economic data collectionand analysis

Proformas were developed for collecting dataon the profitability of both rainfed and irrigatedcrops and a sample of farmers was interviewed ineach mandal. The resulting data were then used toundertake both financial and economic analysis ofdifferent crops and cropping systems. Additionalsupporting information was collected on therelative profitability of non-landbased activities,the role of livestock, proximity to markets anddrought coping strategies.

Simple economic analysis was also carriedout on the ability of farmers to service loans onborehole investments and the levels of risk that,in particular, small and marginal farmers havebeen taking when developing new wells.

of the water balance were difficult to estimate,establishing water balances was possible using datathat were readily available. Results were cross-checked with the often qualitative observationsand experiences of specialists and local peopleworking and living in the areas of interest. Cross-checks were also made by comparing results withthe results from research studies that had beencarried out in the region (e.g. the ChinnatekurWatershed Study [Rama Mohan Rao et al, 1995]).

Separate water balance calculations weremade for: groundwater extraction for irrigation,domestic and livestock uses; actual evaporationfrom irrigated, rainfed arable, wasteland andforested areas; and, groundwater recharge.In the case of groundwater recharge the GEC 97methodology and the methodology describedin Box 6 were used.

As a caveat, it must be stated that there areuncertainties in the absolute values presented inthis report and it can be anticipated that theseuncertainties will be reduced as more reliabledata become available. However, as the samemethodologies were used when working witheach domain type (e.g. annual water balance ofa village area), relative differences can be usedwith a reasonable level of confidence.

2.7 Water balance calculations

Water balance calculations were used toassess the status of water resource availability,with particular attention being given toevaluating the impacts of land use change,groundwater extraction and water harvestingstructures on temporal and spatial patterns ofwater resource availability and use. The initialstep in performing the water balances was tospecify the spatial and temporal boundaries ofthe domain of interest. For example, individualtank systems bounded by the catchment andcommand area were considered to be a domainbounded in time by a particular growing season.Conservation of mass requires that, for thedomain over the time period of interest, inflowsare equal to outflows, plus any change of storagewithin the domain. Although many components Silt excavation from Chapari tank, Kalyandurg

24

Box 6. Estimation of annual average recharge

Recharge estimates were made using the simple water balance equation:

R = Qp + Qa + Rd + Et + Bf - s

Where: R was annual groundwater recharge, Qp was groundwater pumped from wells,Qa was aquifer throughflow, Rd was recharge to deep aquifers, Et was groundwaterextraction by deep-rooting vegetation, Bf was the baseflow contribution to streamsand s was change in aquifer storage.

For the domains over which recharge estimates were made, it was assumed that Qawas very small. This assumption was based on a knowledge of aquifer characteristicsand levels of groundwater depletion. Rd was also assumed to be negligible becauseextraction is already taking place from the deep aquifers in these domains. Et and Bfwere assumed to be negligible because of low groundwater levels. s was alsoassumed to be small because the tendency during recent years has been for farmersto pump wells until they fail. Hence, recharge was estimated as being equivalent togroundwater extraction for irrigation which, in turn, was estimated using areasunder different crops and information gathered locally on actual irrigationapplication rates for these different crops. In each watershed, drainage wasestimated to be 20%, 20% and 10% of water applied during the kharif, rabi andsummer seasons, respectively.

Gauging River/ Site name River Start of End of CatchmentStation Basin record record area (km2)Code

12A Chinnahagari at Amkundi Bridge Krishna Jun-80 May-98 2206

11A Vadavathi at Bhupasamudram Krishna Dec-78 May-98 15326

95 Hundri at Lakshmipuram Krishna Jun-89 May-98 3592

8N Pennar at Nagalamadike Pennar Jun-78 May-98 4814

41 Pennar at Tadapatri Pennar Jun-71 May-98 12558

2N Chitravathi at Singavaram/ Ellanuru Pennar Jun-80 May-95 6301

93 Kunderu at Alladapalle Pennar Jun-85 May-98 8736

catchments. Stations with long periods ofmissing data were excluded from the analysis.Records with only a few months of missingdata were patched using data from the nearestadjacent station. A total of 15 stations were used.Areal rainfall values were determined from theindividual rainfall records using the Thiessenmethod.

2.8 Runoff analysis

River flow data were collected from theCentral Water Commission and other officialgovernment sources for seven river basins thatdrain the south-eastern parts of Andhra Pradeshand neighbouring Karnataka. These data wereconsolidated into continuous monthly and annualriver flow records of 9 to 27 years duration (seeTable 1). Contiguous monthly rainfall data werealso collated for rainfall stations within the study

Table 1. Details of hydrological stations used in runoff analysis

25

The impact of intensive water harvestingalong drainage lines on tank inflows wasestimated using a simple “bucket-type” waterbalance model. The main aim of this model beingto assess the affects of the additional storagecreated by new water harvesting structures andthe relationship between runoff attenuation anddifferent patterns of rainfall. During the WaterAudit, runoff analysis was carried out using datafrom five representative tank catchment areas insouthern Andhra Pradesh and north-easternKarnataka. Runoff estimates were made using aversion of the SCS method that has beenmodified for Indian conditions (see Box 7).The main findings of this analysis were cross-checked against the perceptions and knowledgeof villagers and NGO staff living and workingnear to the five tank systems that were studied indetail. Subsequent to the Water Audit the modelwas refined by the WHiRL Project and anExcel-based version was developed based on adaily time step. This model was then used toanalyse the affects of water harvesting on patternsand availability of runoff in the catchment areasof an additional six tanks in Kalyandurg.

Box 7. Runoff estimation usingthe SCS method

The SCS method for estimating runoffcan be applied to small agriculturalcatchments. Although it wasdeveloped originally using dataobtained in the US, the method hasbeen modified to suit Indianconditions (e.g. Singh et al., 1990).The equation governing the relationsbetween rainfall and runoff used inthe study reported here was:

Q = (P – Ia)2 / (P – Ia) + S

Where,Q = actual runoff (mm)P = run-off generating rainfall(mm)S = maximum potential rainfallretention (mm)Ia = Initial abstraction (0.1S and0.2S for black and red soilsrespectively)

The maximum potential rainfallretention was calculated by theequation:

S = (25400 / CN) -254

Where,CN is the curve number taken fromSingh et al (1990). (CN falls inthe range 1-100. When CN = 100all the rain runs off and S = 0 andP = Q).

Runoff generating rainfall wascalculated using the followingequation:

P = 0.7R – 21.2Where,

R = daily rainfall (mm).

Leaking taps in Kocheruvu village, Dhone

26

3.1 General information

Table 2 provides general facts and figures relating to the two study mandals

General information Dhone Kalyandurg

Longitude 770 39’ - 780 02’ E 770 05’ - 770 22’ ELatitude 150 15’ - 150 29’ N 140 18’ - 140 26’ NAltitude (m) 360 - 650 460 - 780Average annual rainfall (mm) 560 525Mandal area (km2) 487 488No. of revenue villages 16 15No. of hamlets 31 36Population (2001 Census) 102,018 81,105No. of males/females 51850/50168 41278/39827No. of literate males/females 31916/19467 24589/16059Child population (0-6 years) 14085 10604District decadal pop. growth (%) 18.1 14.3

Table 2. Summary statistics

Physical characteristics ofof study mandals3

27

adoption of farming systems that both cope withperiods of low rainfall, bearing in mind the factthat meteorological drought is a natural andrecurring phenomenon, and capitalise on yearsof above-average rainfall. The general perceptionis that in every ten year period, there will be fivedroughts of different intensities. Two of thesedroughts will be moderate, two will be severe andone will be catastrophic.

Although a widely-held view is that annualaverage rainfall has been declining in dry areas ofsouth-western Andhra Pradesh, statistical analysisof 100 years data from 13 stations in Anantapur,revealed that, if anything, average annual rainfallhas been increasing, albeit by around 25 mm,throughout the district since the mid-1970s(Hill, 2001). The only station to show a declineduring recent years is Kalyandurg (Hill, 2001).Comparison of two periods (1901-51 and1951-2001) revealed a slight decreasing trend invariability over the whole Anantapur District,with seven of the thirteen stations witnessingdecreasing variability. However, decadal analysisindicated that during the most recent decades of1991-2001 and 1981-91, nine and elevenstations, respectively, faced increasing variability(Hill, 2001).

2 Long-term rainfall data records for Dhone andKalyandurg are not in good order. In particular,the IMD Kalyandurg rainfall records for the period1901-1950 are clearly for another station.

3.2 Agro-climate

The climate prevailing in the two mandals issemi-arid to arid. Rainfed agricultural productionin south-western Andhra Pradesh is far from easyas monsoon rains are often unevenly distributedand droughts are common. In fact, Anantapurhas been listed as the second most drought-pronearea in India in terms of rainfall (Hill, 2001).Although Kurnool has generally higher rainfallthan Anantapur, Dhone ranks as the driestmandal in Kurnool district. Far from the east-coast, this part of Andhra Pradesh does notreceive the full benefits of the north-eastmonsoon (October to December); and being cutoff by the Western Ghats, the south-westmonsoon (June to September) is also preventedfrom fully reaching the district. The south-westmonsoon and north-east monsoon rainfallcontribute about 57% and 27% of the totalrainfall for the year, respectively (Hill, 2001). Inclimatic terms, the area does not have distinctkharif and rabi seasons as, rainfall permitting,cropping takes place continuously throughoutthese periods (except on deep black soils).

Using data from Anantapur2, Figure 5 showsthat there is considerable variation in annualrainfall around the average value. In some yearsduring the period 1901-2001, rainfall was asmuch as 475 mm higher than the average and inothers it was more than 400 mm below theaverage. Extreme inter- and intra-annual rainfallvariability is also an important characteristic ofthe agro-climate of the study mandals. A majorchallenge facing farmers in this area is the

Figure 5. Anantapur deviation in annual rainfall from the long-term mean

28

-500

-400

-300

-200

-100

0

100

200

300

400

500

Dev

iatio

n fro

m m

ean

ann.

rain

fall

(mm

)

1901-021910-11

1919-201928-29

1937-381946-47

1955-561964-65

1973-741982-83

1991-922000-01

Figure 6 compares the meanmonthly rainfall at Kurnool andAnantapur with potentialevaporation3. It can be seen thatAnantapur’s mean monthly rainfallis below 100 mm for every monthexcept September. This figure alsogives an indication of the relativelyhigher rainfall at Kurnool duringthe early south-west monsoon andrelatively higher rainfall inAnantapur during the north-eastmonsoon. Mean monthly rainfall isless than potential evaporation forboth locations for every monthexcept September.

Figure 7 shows the probabilityof monthly rainfall exceeding 20,40, 60 or 100 mm at Anantapur. Itcan be seen that the highestprobability of monthly rainexceeding 60 mm is duringSeptember-October. Hence, this isthe period during which largevolumes of runoff are most likely tobe generated.

Figure 8 presents the meannumber of rain days per month forAnantapur. As the number of raindays is not well correlated with therainfall, this also indicates thatrainfall events tend to be smallerduring the early and late parts ofthe south-west and north-eastmonsoons respectively.

3 Potential evaporation can be definedas the rate of evaporation from anextensive surface of 8 – 15 cm tall, greengrass cover of uniform height, activelygrowing, completely shading the groundand not short of water. The ratiobetween rainfall and potentialevaporation provides a good indicationof the aridity of an area.

Figure 6. Mean monthly rainfall and potentialevaporation (ETp) (FAO)

Figure 7. Probability of different monthly rainfallamounts in Anantapur (CSWCRTI)

Figure 8. Mean number of rain days for Anantapur

29

0

50

100

150

200

250

Jan Feb Mar Apr May Jun Jul Aug Se p Oct Nov Dec

Rai

nfal

l/ETp

(mm

)

Ana ntapur Kurnool ETp

-20

0

20

40

60

80

100

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Prob

abili

ty (%

)

20 mm 40 mm 60 mm 100 mm

0

2

4

6

8

Rai

nfal

l (da

ys/m

onth

)

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 9. Geology of Kalyandurg mandal

underlain by granites are undulating with numerousrock outcrops. The areas underlain by thesedimentary rocks are hilly with a steeper terrain.Figure 10 shows Dhone’s main geomorphic units.4

3.4 Soils

Approximately 95% of the soils in Kalyandurgare alfisols with the remaining area covered by blackclayey soils (vertisols) (APGWD, 1999). The soilsin Dhone are also predominantly red sandy loamsoils with a depth in the range 0.3 - 1 m (APGWD,2000). Table 3 provides information on the lengthof growing period for each APRLP district based onrainfall frequency analysis and soil water holdingcapacity. The relatively short growing period inAnantapur gives an indication of the risk involvedin rainfed cropping in this district. The inherentadvantages of the vertisols, in terms of soil wateravailability, can also be seen.

Shallow and very shallow gravelly soils occur atthe base of foothills with 3-5% slope in Dhone and4-8% slope in Kalyandurg. The gravel content ofthese soils ranges from 50 to 85% and is a majorlimitation on arable cultivation. Moderately deepsoils occur in gently sloping lands and deep to verydeep soils occur in valleys.

3.3 Geology and geomorphology

Kalyandurg has an undulating topographywith a general slope towards north and east.The geology of the mandal (see Figure 9) iscomprised of Archean rocks which consist ofgneisses, schists, younger granites, quartz veins andbasic dykes. The Archaen rocks have suffered aconsiderable degree of tectonic disturbanceas a result of which the rocks have beenmetamorphosed and recrystallised (APGWD,1999). The granite rocks may be separated into twodistinct groups, namely, the massive and foliatedtypes. The massive grey granites give rise to elevatedfeatures while foliated rocks are found below theplains and also occur as low dome hillocks. Dharwarrocks, occurring as a linear schist belt within thegneissic complex, form linear hills in the easternpart of the mandal. Numerous basic dykes, whichare essentially dolerite in composition, traverse theolder rocks and these are exposed as long narrowand generally persistent ridges. The length of thedykes varies from 0.5 to 9 km and the width from50 to 500 m. In addition to the above, alluvium of1-7 m thickness occurs along the course of thePennar river and in the vicinity of minor streamsand tanks.

Located on the periphery of the Cuddapahbasin, Dhone has a geology that comprises bothcrystalline and sedimentary rocks. Granites andassociated rocks with intrusive bodies like dykes andquartz veins are encountered to the north and westof the mandal and sedimentary rocks such aslimestones and shales with intercalation of chertbands are found to the south and east. The areas

4 Despite considerable effort, obtaining a 1:50,000groundtruthed map of Dhone proved to be impossibleduring the water audit. Copies of this map could notbe found either in Kurnool or Hyderabad.

30

4000 0 4000 8000 (m)

Closepet GraniteDoleriteGravelGrey GraniteHornblendePink GraniteQuartz, Reef

Figure 10. Dhone’s geomorphic units

also common in the mandals particularly fromriver bed areas and groundnut fields. Both thesoils are universally deficient in nitrogen.Red and black soils are poor in phosphorus,while potassium may become a limiting factorin light soils dominated by the clay mineralkaolonite. Deficiency of zinc, sulphur andcalcium is widespread in the case of red soils thatare under continuous groundnut cropping.Similarly in black soils, zinc deficiency is a majorlimitation for crop production under intensivefarming. Boron toxicity is also reported to inhibitcrop growth in these mandals.

District Soil Type First moist Humid Second Length ofperiod period moist period growing(days) (days) (days) period

(days)

Anantapur Sandy Alfisol 49 7 56 112

Kurnool Sandy Alfisol 77 35 42 154Vertic Soil 77 35 42 154

Mahabubnagar Vertisols 27 141 21 203Vertic Soils 27 141 14 189Sandy Alfisol 27 141 14 182

Nalgonda Sandy Alfisol 42 112 28 196

Prakasam Vertic Soils 71 91 18 180Orthids 70 96 18 186

As a result of erosion, clays are transported tovalleys where they form calcareous clayey soilswhich are at times saline. Red soils suffer from thephysical limitations of crusting and compact subsoils, while black soils suffer from crusting andlow infiltration rates due to high exchangeablesodium percentage (>7.0). Soil salinity is a seriousproblem in the case of black soils and in red soilswhen irrigated with saline waters.

Soil depths vary as a result of differentialerosion. High levels of erosion continue to occuron many of the steeper slopes with soil lossesranging from 4 to 10 t ha-1 y-1. Wind erosion is

Table 3. Length of crop growing period for main soil types in thefive APRLP districts (Source: ICRISAT)

31

Chinnahagari river following heavy rains

Krishna river during October 2001

3.5 Drainage

The eastern part of Kalyandurg is drained bythe Pennar river which flows from south to northacross the mandal (see Figure 11), whilst thewestern part of Kalyandurg drains northwardstowards the Chinnahagari river. The mandal canbe divided into eleven sub-watersheds and thedrainage density is 1.79 km/km2. Dhone mandaldrains into both the Krishna and the Pennarsystems. This mandal can be divided into sevensub-watersheds that have streams that areephemeral. Taking the mandal as a whole theaverage drainage density is 2.63 km/km2.

Table 4. Water harvesting structures

Type Dhone Kalyandurg

Masonry 225 132check dams

Rockfill dams 28 19

Earthen dams - 22

Nala bunds - 6

Total 253 179

32

Figure 11. Kalyandurg’s water harvestingstructures and drainage network

3.6 Gully control structures

Figures 11 and 12 show the location of waterharvesting and gully control structures that havebeen constructed as part of the DPAP, DDP andother government programmes. Table 4 providesmore details on the type and number ofstructures. The totals in Table 4 can be comparedwith the APGWD figures of 83 check dams inKalyandurg and 164 gully control structures inDhone. One reason for such large discrepancies(i.e. 50% and 115%) is the fact that the officialfigures do not include structures that were notfunded by the DPAP and DDP programmes. Itshould be noted that a more detailed survey ofgully control structures by the WHiRL Project infour villages in Kalyandurg showed that manystructures were also missed by the Water Auditsurveys. If all smaller structures are included thetrue numbers of gully control structures may bearound double the figures reported here.

Gully control structures are relatively moreuniformly distributed across Kalyandurg. InDhone, gully control structures are concentratedprimarily in the red soil areas. Many gullies inother areas of Dhone that have steep slopes havebeen treated in recent years as part of the Neeru-Meeru programme’s continuous contourtrenching activities.

3.7 Tanks

The use of tanks for catching runoff is atraditional practice in the study mandals that ismore prevalent in the red soil areas underlain by

crystalline basement geologies. The tank surveysshowed that there are 85 tanks in Kalyandurgand 18 tanks in Dhone. Official statistics showthat in Kalyandurg there are 13 minor irrigationtanks having a total command area of 367 ha.Although some of Kalyandurg’s tanks date backto the early 18th Century, the majority wereconstructed during the second half of the 19th

Century or the first half of the 20th Century.The survey showed that approximately 25% ofthe tank bunds and weirs were in poor condition.As part of GoAP policy, all but four of thetanks are now used as percolation tanks(i.e. groundwater recharge structures) rather thansources of water for surface irrigation. Farmers inthe command areas now rely on groundwater asa primary source of irrigation water and, in somecases, seepage water from the tank or waterflowing through damaged sluices or weirs.In the majority of tanks, siltation has occurredand, as a consequence, dead storage and averagedepth have been reduced. Issues surroundingthe widely-held view that inflows to tanks havedecreased in recent years are discussed inSection 4.3.

3.8 Land capability

The areas of the two study mandals withdifferent land capability classes are presentedin Table 5. The soil and land characteristicassociated with each capability class and sub-class can be found in Table 6.

33

Figure 12. Dhone’s water harvesting structuresand drainage network

Table 6. Land capability soil and land characteristics

Land Soil and land characteristicsCapabilityclass

IIs-IIIe Moderately deep, well drained, good cultivable lands having erosion and soil problems due toheavy texture. Class also includes fairly good cultivable lands having erosion and soil problems.Some areas need simple soil and water conservation measures; all climatically adapted crops canbe grown.

IIes-IIIes Moderately deep, well drained, good cultivable lands associated with minor to moderate problemsof erosion and heavy textures and gravelliness; need appropriate soil and water conservationmeasures; all climatically adapted crops can be grown.

IIs-IIIes Moderately deep to deep, well drained, good cultivable lands with minor soil problems of heavycracking clays; associated with moderate problems of erosion and soil problems; needsimple soil and water conservation measures; all climatically adapted crops can begrown.

IIIes Moderately deep to shallow, well drained. Good cultivable lands with problems oferosion, shallow rooting depth, gravelliness and stoniness and gentle slopes; needintensive soil and water conservation measures; all climatically adapted crops can begrown.

IIIs Moderately shallow, well drained, moderately good cultivable lands having soil problems ofshallow rooting depth, gravelliness, slightly eroded and gentle slopes; need simple soil and waterconservation measures; all climatically adapted crops can be grown.

VIe-VIIes Shallow to moderately shallow, well drained uncultivable lands with very severe limitations oferosion, shallow rooting depth, gravelliness and stoniness, steep to moderate slopes; needintensive soil and water conservation measures; suitable for forestry, pasture and silvipasture.

VIes-VIII Shallow to very shallow, excessively drained, uncultivable rock lands having severe erosion andsoil problems associated with fairly good cultivable lands with very severe limitations of erosion,shallow rooting depth, gravelliness, stoniness and steep slopes; need intensive soil and waterconservation measures; suitable for quarrying, mining, as habitat for wild life, pasture andforestry.

34

Table 5: Areas under different land capability classes in Kalyandurg and Dhone mandals

Kalyandurg Dhone

Class Area (ha) Class Area (ha)

IIs to IIIes 6,253 IIs to IIIe 15,245IIIs 9,335 IIes to IIIes 5,325IIIes 31,809 IIIes 11,708VIe to VIIes 1,438 VIes to VIII 16,456

Total 48,835 Total 48,644

Figure 13. Land use based onremotely-sensed data

3.9 Land UseKalyandurg

Dhone

35

20%

8%

60%

6% 5%

0%

1%

ForestIrrigatedRainfed arableFallowWastelandTanks/riversUrban

Figures 13 and 14 present remotely-sensedinformation on current land uses in the twostudy mandals. There are a number of importantpoints that relate to this information. Firstly, themain land use in both mandals is currentlyrainfed arable cropping. The precise areacropped and the cropping system varies fromyear to year in direct response to the onset of thesouth-west monsoon and the subsequent rainfallpattern. However, in most years, only one cropcan be grown. Groundnut or a groundnutintercrop system is preferred. Secondly, withinliving memory, there has been a major decline inforested areas in both mandals and in thebiodiversity associated with undisturbed forest.Currently, Dhone has a relatively larger areathan Kalyandurg under Forest Departmentresponsibility. In both mandals, forested areasare generally degraded and devoid of vegetationother than scrub. Also in both mandals, goodquality forested land, other than land underForest Department control, has become heavilyencroached in recent years. Participatoryassessments indicated that fuel wood shortageshave also increased during recent years. Thirdly,substantial areas of fallow and waste land exist inboth mandals.

During the last 10-15 years there has been asubstantial increase in groundwater-basedirrigation in both mandals. According toremotely-sensed information, 8 and 12% of thenet land area of Dhone and Kalyandurgrespectively was under multi-cropped irrigationduring the 2000-2001 crop season. Thesefigures were supported by independent estimatesbased on well survey data collected by theAndhra Pradesh Groundwater Department aspart of this Water Audit. However, the “official”irrigation statistics supplied by the MandalRecords Office (i.e. the Revenue Dept) and theDepartmentof Agriculture (see Figure 15)suggested that the total net irrigated area ismuch smaller. During discussions involvingdistrict-level specialists and line departmentstaff, it became clear that gross under-reportingof “official” irrigation statistics is commonknowledge (see Box 7a). The lack of faith inirrigation statistics at the district level did notappear to be reflected at the state level where“official” irrigation statistics are used tounderpin a whole range of important policy-related decisions.

0

1000

2000

3000

4000

5000

6000

7000

Dhone Kalyandurg

Irrig

ated

are

a (h

a)

AP Revenue Department NRSA APGWD/CSWCRTI

Figure 14. Main land uses percentagesbased on remotely-sensed data

Figure 15. Comparison of irrigation statisticsfrom different sources

Kalyandurg

Dhone

36

Dhone

5%12%

64%

10%4% 3% 2%

Box 7a. Collection of irrigationstatistics

The basic source of primary dataas far as number of wells, landownership, cultivated area andirrigated area is concerned is thevillage record, or Adangal. Thisis maintained for every revenuevillage by the concerned VillageAdministrative Officers (VAOs)working under the RevenueDepartment. Data from revenuevillages are aggregated toprovide mandal, district andstate statistics. Thus theaccuracy and reliability ofirrigation statistics dependsentirely on the standard of workof individual VAOs not leastbecause data checkingmechanisms are basic andgenerally just a formality. Evenconscientious VAOs tend not toput much effort into recordinggroundwater-based irrigationstatistics and the reality is thatnew wells often go unrecordedand areas irrigated per well arerarely entered into the Adangal.One reason being that there ismore compulsion for accuracyregarding land use pattern, landownership and area undersurface irrigation sources asthey are directly used forcollection/assessment of taxes(Janakarajan, 2000).

Continuous contour trenching near Dhone town

Irrigated floriculture in Erraguntla, Dhone

Salt-affected area near Lakshmipalli, Dhone

37

3.10 Vegetation

The region was endowed with diversified landuse systems ranging from agriculture to forestry.However, in the last 50 years, community barrenlands and large parts of forest areas abuttingvillages have been encroached upon by villagers.One consequence is that access to fuel, fodder andpasture lands has become a problem in many ofthe villages in the study mandals.

Table 7 summarises the results of a qualitativevegetation survey that was carried out in Dhoneand Kalyandurg. Although the lists obtained werequite encouraging, the main observation of thesurvey was the high prevalence of prosophisjuliflora in wasteland areas, along gullies andalong roadsides. It was assumed that this wasdue in part to degradation and overgrazing.

3.11 Land Holdings

Cadastral information was collected andupdated in digital format. In addition toproviding a valuable cross-check of land use,village/mandal area and well statistics, these dataprovided an important source of information foranalysis relating to levels of land fragmentation,economics of well construction and identification

of the main beneficiaries of water harvesting andgroundwater-based irrigation. Figure 16 is acomposite layout of spatial and non-spatialcadastral data that is presented to give an exampleof the level of detail available in the cadastraldatabase. It should be noted that this information,which is held and updated by the RevenueDepartment, is rarely used by developmentprogrammes. The villagewise size distributionof land holdings on patta land in Kalyandurg ispresented in Figure 17. It can be seen that theaverage size of holding, size of holding distributionand number of land owners varies enormouslyfrom village to village. In both Dhone andKalyandurg, the average size of holding isdeclining as a result of inheritance arrangementsthat involve splitting land between sons.However, extended families often farm theirholdings as one unit or, at the very least,cooperatively. Consequently, the size of individualholdings is not always a reliable indicator oflevels of poverty. It was found also that, in bothmandals, cooperative use of groundwater(i.e. sharing of wells) was common by membersof the same family.

Table 7. Tree, shrub and grass species found in Dhone and Kalyandurg

Tree species Shrubs Grass species

Acacia leucophloes Carissa carandus Apluda aristata

A. sundra C. spinorum Cymbopogan coloratus

A. arabica Cassia auriculata Cynodon dactylon

Albizia amara Dodonia viscose Digitaria longifolia

A. lebbek Caparis hispida Imprata cylindrica

Azadirachta indica Randia dimmatorum Heteropogan contortus

Dalbergia sissoo Euphoria tirucalli Sehima nervosum

Ziziphys Zylopyalis E. piverlia

Morinda tinctoria Jatropha carcus

Soymidia febrifuga

Pterocarpus santalinus

Wrightia tinctonia

Delonixelata

Hardwickia binata

Feronia elephantum

Gyrocarpus jacquini

38

0

100

200

300400

500

600

700

800900

1000

Hulik

al

Chap

iri

Mud

igal

lu

Gol

la

Man

irevu

Thim

mas

amud

ram

Mud

inay

anam

palli

Varli

East

Kod

ipal

li

Kaly

andu

rg

Gar

udap

uram

Krub

hara

halli

Bedr

ahal

li

Pala

voi

Durd

akun

ta

No. o

f far

mer

s

< 1 acre 1 - 2.5 acres 2.5 - 5 acres 5 - 10 acres > 10 acres

Figure 17. Villagewise land holding sizedistribution in Kalyandurg

Figure 16. Example of spatial and non-spatialcadastral information accessible from GISdatabase

39

4.1 Groundwater resources

Well statistics

Figures 18 and 19 show the locations of wells inKalyandurg and Dhone respectively. These data werecollected as part of the well survey carried out by theAndhra Pradesh Groundwater Department. The wellssurveyed included open wells, inborewells and borewells.Wells that were defunct but still intact were included inthe survey, whereas open wells that had been completelyabandoned and borewells that no longer had headworkswere not included in the survey. These figures show thepreference for siting wells near to drainage lines. In general,the absence of wells in certain areas can be explained interms of either unfavourable geological conditions or poorgroundwater quality.

General resource statusof Dhone and Kalyandurg4

40

Figure 19. Location of wells in Dhone

Figure 18. Location of wells in Kalyandurg

41

of the century are, therefore, misleading and thereality was that the rate of well construction wasmore uniform during these decades. In contrast,peaks during the last 25 years were relatedprimarily to government grants and loans thatencouraged sudden spurts in well construction.

Figures 20 and 21present the number ofwells of different types constructed during the20th Century in Kalyandurg and Dhonerespectively. As information on date ofconstruction of wells was elicited in part fromlocal villagers, information for the first part ofthe century is presented on a decadal basis. Thepeaks in well construction during the first part

Figure 20. Well construction timeline for Kalyandurg

Figure 21. Well construction timeline for Dhone

42

0

50

100

150

200

250 W

ell c

onst

ruct

ion

(wel

ls/y

ear)

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Openwell Inborewell Borewell

0

50

100

150

200

250

Wel

l con

stru

ctio

n (w

ells

/yea

r)

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

Open well Inborewell Borewell

The following points relate to the history andpattern of well construction in these mandals:

. There has been a dramatic increase in wellconstruction and groundwater extractionduring the last 15 years primarily forgroundwater-based irrigation;

. Official statistics on well constructionunderestimate the current number of wellsby a factor of two. This finding has majorimplications for estimation of groundwaterdraft that is based on well statistics(e.g. the GEC-97 Methodology);

. At the mandal level, there was reasonableagreement between water audit wellstatistics and statistics provided by Transcoon the number of pump connections;

. In recent years, groundwater extraction perwell has increased substantially as a resultof the availability of submersible pumpsand electricity;

. The shift in well construction from openwells to borewells, in Kalyandurg inparticular, represents a shift fromgroundwater extraction that exploitedshallow regolith aquifers to extraction fromdeeper bedrock aquifers.

Box 8. Driving forces behind wellconstruction

Factors that have encouraged the increasein well construction include:

. Markedly higher and more reliablereturns that farmers get fromirrigated cropping as opposed torainfed cropping;

. Government programmes that haveprovided grants or soft loans for wellconstruction;

. Government policies that have led tothe cost of electricity being subsidised;

. Improved drilling technology andcompetition between contractors.Both have ensured that the cost ofborewell construction is relatively low;

. Competition for water betweenfarmers accessing the same aquifer.This has led to competetivedeepening of wells and/or constructionof new borewells.

Figure 22 presents the status of wells surveyed in Kalyandurg. It can be seen that in early 2001, 31% ofthe wells surveyed had either failed completely or were unreliable. The wells that were completely defunctwere predominantly open wells. Similar analysis of survey data for Dhone indicated that less than 1% ofwells were defunct and 7% of wells failed routinely. Although the number and density of wells inKalyandurg and Dhone was similar, this difference in well failure was explained by the fact that levels ofgroundwater extraction for irrigation, domestic use and livestock were around 30% higher in Kalyandurgas compared to Dhone. In both mandals indications were that competition for water between farmers wasintensifying. The prognosis is therefore for an increase in well failure in both mandals and a more markedshift from open well to borewell construction in Dhone.

Figure 22. Well failure ratesin Kalyandurg

43

Domestic water use

Net groundwater extraction for irrigation,domestic and livestock use for Kalyandurg andDhone was estimated at 11.0% and 8.4% of meanannual rainfall respectively. As the Andhra PradeshGroundwater Department estimate ofgroundwater recharge in this area is approximately10% of annual rainfall, this suggests that currentlevels of extraction in Kalyandurg are notsustainable. Analysis of villagewise groundwaterextraction showed large differences betweenvillages in both mandals. These were attributed inpart to more favourable hydro-geologicalconditions in some villages and in part tovariations in the historical pattern of developmentof both ground and surface water resources. Thisfinding, coupled with the results of analysis ofcadastral information, illustrates the fact that thereis neither equitable access to irrigation betweenvillages nor between households within villages.

Figure 23 gives an indication of the number ofwells per revenue village in Kalyandurg that werebeing used as sources of water for irrigation anddomestic supply. As might be expected, thenumber of “irrigation” wells now far exceeds thenumber of “domestic supply” wells. Annualgroundwater extraction for domestic and livestockuse was estimated at 0.6% and 0.8% of annual

rainfall for Kalyandurg and Dhone respectively.If annual domestic and livestock use areconsidered in terms of average annualgroundwater recharge, it is apparent thatcurrently 6% and 8% of average annual rechargeis being used to meet these needs in Kalyandurgand Dhone respectively. Taking a populationgrowth rate of 2.5% and assuming that currentper capita levels of water use are maintained, itcan be estimated that groundwater extraction fordomestic and livestock use will double in the next30 years to 12% and 16% of average annualgroundwater recharge. In many village areas,meeting this increased demand will not bepossible without a commensurate reduction ingroundwater use for irrigation. There isconsiderable variability in domestic water usewith some revenue village areas (e.g. Dhone town,Kalyandurg town and Duradakunta) havingcurrent levels of use well in excess of 20% ofaverage annual recharge.

Figure 23. Use of wells in Kalyandurg

44

Figure 24. Automatic groundwater level datafrom Dhone

watershed development programmes.Information showing: changes in access andentitlements to water, changes in time taken bywomen and children to fetch and carry water,changes in the number of villages requiringtankers or with water markets and changes inirrigated cropping intensity provide morecompelling evidence as to whether or notprogrammes are successful.

Groundwater levels

Figure 24 presents data from two of the threeautomatic groundwater level recorders installed inDhone as part of the National Hydrology Project.Although these provide a useful insight into thedynamics of groundwater level change, the data arenot easy to interpret without accurate informationon levels of groundwater extraction in theimmediate vicinity of the observation wells. Atbest, groundwater levels are a crude and unreliableindicator of the localised success (or otherwise) of

45

0

50

100

150

200

250

300

Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02

Mon

thly

rain

fall

(mm

)

Data from the groundwater levelrecorders at Golla and Kalyandurg(see Figure 25) do not correlate wellwith the findings of the GroundwaterDepartment’s survey in this mandal.In the case of Golla, this can be explainedby the fact that the recorder is sitedin a locality with a relatively low level ofgroundwater extraction. Also the actualsite of this recorder (as with many others)was chosen primarily on the basis of thesecurity of the equipment. The widerimplication is that water-level recorderinformation should be used with extremecaution when assessing regional impactsof, say, watershed development.

Box 9. Electricity tariffs forirrigation wells

One-off connection chargesinclude: a Rs.25 application fee,a Rs.100 security deposit anda development fee of Rs.1,000per HP of the pump. Onceconnected, the electricity tariffis an annual charge thatdepends only on the rating ofthe pump. Pumps with ratingsof up to 3 HP, 3-5 HP, 5-10 HPand more than 10 HP haveannual tariffs of Rs.225,Rs.375, Rs.475 and Rs.575respectively. Farmers with smalland marginal holdings can optfor unit payment or optionaltariff of Rs.0.2 per unit upto 2,500 units and Rs.0.5 forconsumption in excess of2,500 units.

Figure 25. Groundwater level data from Golla andKalyandurg (Source: Andhra Pradesh GroundwaterDepartment)

Small tank near Gosanapalli, Dhone

Goat drinking from a hand pump

46

0

50

100

150

200

250

300

Jan-97 Jul-97 Jan-98 Jul-98 Jan-99 Jul-99 Jan-00 Jul-00 Jan-01 Jul-01 Jan-02 Jul-02

Mon

thly

rain

fall

(mm

)

Figure 26 presents stage of groundwater development in Kalyandurg as estimated by the AndhraPradesh Groundwater Department using data collected during the Water Audit survey. This figure showsthat groundwater development has reached the “critical” or “over-exploited” stage in two thirds of thevillages in this mandal. Subsequent and more detailed water-budget investigations carried out by theDFID-supported WHiRL project in four villages in Kalyandurg supported the findings presented here.

Box 10. Stage of groundwater development terminology

Safe: A sub-unit is categorised as “safe” with potential for future groundwater developmentif one of the following two criteria is fulfilled: i) the stage of groundwater development isless than or equal to 70% and the water table during at least one of the two intervals(either pre-monsoon or post-monsoon) does not show a falling trend and ii) the stage ofgroundwater development is greater than 70% but less than or equal to 90% and the watertable during both pre-monsoon and post-monsoon intervals does not show a falling trend.

Semi-critical: A sub-unit is categorised as “semi-critical” with caution to be executed forfuture groundwater development if the following criterion is fulfilled: the stage ofgroundwater development is greater than 70% but less than or equal to 90% and the watertable during only one of the two intervals (either pre-monsoon or post-monsoon) showsa falling trend.

Critical: A sub-unit is categorised as “critical” with only very marginal scope for futuregroundwater development if one of the following criteria is fulfilled: i) the stage ofgroundwater development is more than 90% and the water table during only one of the twointervals (either pre-monsoon or post-monsoon) shows a falling trend and ii) the stage ofgroundwater development is equal to 100% and the water table during both pre-monsoonand post-monsoon intervals shows a falling trend.

Over exploited: A sub-unit is categorised as “over exploited” with practically no scopefor any future groundwater development if the following criterion is met: the stage ofgroundwater development is more than 100% and the water table during both pre-monsoonand post-monsoon intervals shows a falling trend.

Source: Andhra Pradesh Groundwater Department

Figure 26. Stage of groundwaterdevelopment in Kalyandurg

47

a laxness was noted in the way in which fluoridepermissible limits are being used by departmentsresponsible for drinking water supplies. Aconsequence being that action is not taken evenwhen fluoride levels are well in excess of Indianpermissible limits (i.e. 1.5 ppm).

Although not such a severe problem as fluoride,the common occurrence of saline groundwater inboth mandals was found to be a problem in somedomestic supplies and, in some areas, a majorconstraint on agricultural use. In Dhone, fifty of thesources of domestic supply were found to exceed thepermissible limit for total dissolved solids.

Conclusion

The overall conclusion of the “groundwater”component of the Audit is that the scope fordeveloping additional groundwater resourcesin both mandals is limited and, in much ofKalyandurg, groundwater is already severely over-exploited. This conclusion does not agree with arecent statewide groundwater survey carried out bythe Andhra Pradesh Groundwater Department. Thisdifference of opinion can be explained by the factthat the statewide survey based estimates ofgroundwater draft on “official” figures for irrigatedarea and well numbers that appear to hugelyunderestimate the situation on the ground.

Figure 27. Sampled wells in Dhone with fluorideexceeding domestic water permissible limits

Groundwater quality

The current WHO permissible limit for thefluoride concentration of drinking water supplies is1.5 ppm with the added recommendation that“climatic conditions, volume of water consumedand intake from other sources should be consideredwhen setting national standards”.In 1993, the Bureau of Indian Standards set amaximum permissible fluoride concentration of1.0 ppm, although concentrations of up to1.5 ppm are considered to be acceptable in theabsence of an alternative safer source.

Figure 27 shows the number of wells that weresampled in Dhone that have fluorideconcentrations in excess of 1.5 ppm. It can be seenthat 468 of the wells sampled had fluoridelevels in excess of 1.5 ppm. Of these, 62 werebeing used solely as a source of domestic supply.Preliminary water sample analysis for Kalyandurgsuggested a relatively lower prevalence of wellswith fluoride in excess of 1.5 ppm as comparedto Dhone. However, follow-up analysis bythe WHiRL Project, indicated that high fluoride indomestic water supplies is a major issue inKalyandurg. The routine monitoring of wells bythe WHiRL Project also indicated a widespreadincrease in fluoride concentrations of around30% during the 2002/2003 drought. Althougha reasonable level of awareness exists at the villageand district levels of the risks to human healthrelated to high levels of fluoride ingestion,

48

4.2 Surface water resources

Runoff at the regional and largewatershed scale

Analysis of gauging data from the CentralWater Commission (CWC) showed that annualsurface runoff, at the macro-watershed or basinscale, is somewhat lower than is often reportedor than accepted wisdom would suggest(see Table 8). With the exception of one river,the Kunderu at Alladapalle, the average annualrunoff values are within the range 4-35 mmwhich is equivalent to between 0.8 and 7.5% ofrainfall. The main reason for the apparent higherrunoff in the Kunderu is the fact that this river isused for inter-basin transfer of Krishna Riverwater. The relatively low runoff values observedat most gauging stations challenge thewidespread assumption that runoff in this regionis always in the range 30-40% of annual rainfall.Although significant volumes of water can draininto the Bay of Bengal during periods or yearswith exceptionally high rainfall, in general,runoff represents a relatively small loss from theregion.

The runoff figures presented in Table 8agree with the low runoff values reported in theKAWAD Water Audit Report (Batchelor et al.,2000) for large watersheds. They are alsoconsistent with the characteristics of the area(e.g. semi-arid climate, extensive developmentof water harvesting structures).

Figure 28 compares annual rainfall andrunoff at the CWC Tadapatri gauging stationon the Pennar River. This figure shows theconsiderable inter-annual variation in rainfalland surface runoff and that there have been longintervals of virtually no surface flow during this25 year period (e.g 1990-1996).

Figure 29 compares monthly rainfall andrunoff at Tadapatri. In addition to indicatinglevels of variability, this figure shows that onaverage 80% of annual runoff occurs during thethree months September to November.

In conclusion, the main points from therunoff analysis using CWC data are:

. Throughout southern Andhra Pradesh,average annual runoff values at the largewatershed scale, in the absence of inter-basin transfers, are in the range 1% and8% of average annual rainfall.

. Rainfall and runoff are extremely variableboth within and between years.Coping with seasonality in surface wateravailability and droughts and floods istherefore a key issue.

. Although scope for further developmentof surface water resources along withregulation of river flows is possible insome areas, the scale of this developmentis much lower than suggested by someGoAP programmes.

River Average Average Averageannual runoff catchment annual runoff(mm) rainfall (mm) (as % rainfall)

Chinnahagari at Amkundi Bridge 6 451 1.3Vadavathi at Bhupasamudram 4 450 0.8Hundri at Lakshmipuram 35 468 7.5Pennar at Nagalamadike 12 573 2.0Pennar at Tadapatri 28 551 5.0Chitravathi at Singavaram/ Ellanuru 20 549 3.6Kunderu at Alladapalle 133 617 21.6

Table 8. Average runoff and rainfall forstudy catchments

Check dam near Golla, Kalyandurg

49

Figure 28. Comparison of annual rainfall andrunoff, Pennar River at Tadapatri

Figure 29. Comparison of monthly rainfall (a)and runoff (b), Pennar River at Tadapatri

(a)

(b)

50

0

100

200

300

400

500

600

700

800

900

1971

/72

1972

/73

1973

/74

1974

/75

1975

/76

1976

/77

1977

/78

1978

/79

1979

/80

1980

/81

1981

/82

1982

/83

1983

/84

1984

/85

1985

/86

1986

/87

1987

/88

1988

/89

1989

/90

1990

/91

1991

/92

1992

/93

1993

/94

1994

/95

1995

/96

1996

/97

1997

/98

Rai

nfal

l/Run

off (

mm

)

Runoff Rainfall

0

50

100

150

200

250

300

350

Rai

nfal

l (m

m)

0

20

40

60

80

100

120

140

160

Jun-

71

Jun-

72

Jun-

73

Jun-

74

Jun-

75

Jun-

76

Jun-

77

Jun-

78

Jun-

79

Jun-

80

Jun-

81

Jun-

82

Jun-

83

Jun-

84

Jun-

85

Jun-

86

Jun-

87

Jun-

88

Jun-

89

Jun-

90

Jun-

91

Jun-

92

Jun-

93

Jun-

94

Jun-

95

Jun-

96

Jun-

97

Jun-

98

Jun-

99

Jun-

00

Jun-

01

Runo

ff (m

m)

Runoff at the micro-watershed andfield scales

Micro-watershed and field-scale studies ofrunoff have been carried out by the CSWCRTIin Bellary and by CRIDA in Anantapur. Thesestudies have shown that average annual runoff istypically 2% to 15% of average annual rainfalland highly dependent on such factors as: soiltype, slope, land use and/or vegetative cover andpresence of in-field soil and water conservationmeasures. Although average annual runoff figuresare low, it must be emphasised that runoffresulting from individual or sequences of rainfallevents will often be much higher. Runoff will alsotend to be relatively high from small plots orfrom certain features in the landscape (e.g. rockoutcrops, roads, urban areas, areas of hard panetc.). However, it is the annual average runofffigure that provides an indication of the scope fordeveloping additional surface water resources atany given point in a watershed.

Table 9 summarises runoff findings frommicro-watershed experiments that were carriedout by CSWCRTI (Bellary) at off-stationlocations. The information from Chinnatekuris particularly relevant because this experimentalwatershed is located in Kurnool Districtapproximately 40 km from Dhone. It can beseen from these findings that mean annualrunoff as a percentage of mean annual rainfallis less than 10% regardless of the nature of theland use or land surface conditions. It can alsobe seen that runoff is relatively lower fromforested areas and from agricultural areas thathave received soil and water conservationtreatment.

4.3 Impact of water harvestingon patterns of water availabilityand use

For centuries, tanks have played an importantrole in maintaining rural livelihoods in the semi-arid areas of South India. Until recently, theyprovided a reliable source of water for irrigationand, in many cases, they helped recharge shallowaquifers that met drinking water needs during dryseasons and periods of drought. As has been welldocumented elsewhere, a shift towards centralisedgovernment contributed to the breakdown oftraditional management systems as governmentdepartments took responsibility for decisions andtasks that had originally been the responsibility oflocal people and/or elites. As a result of relativelyrecent state government policy, most tanks in thetwo study mandals have been converted intopercolation tanks (i.e. sluices have been blockedand surface water is no longer released forirrigation in the command area).

In recent years, people in the study mandalshave observed a reduction in both inflows tomany tanks and the frequency of spillage orsurplusing. In discussions, the reasons that areoften given for these changes are: i) A decline inrainfall and ii) Deforestation in the tankcatchment area. Neither of these explanations isconvincing because: i) There has not been ameasurable decline in rainfall and ii) Althoughdeforestation has taken place, this occurred beforethe phenomena of reduced inflows was observedand, in any case, deforestation is more likely toincrease tank inflows.

Using the methodology described in Section2.8, the impacts of water harvesting on fivetraditional tanks systems were assessed. Chapariand Yapadinne tanks are located in Kalyandurgand Dhone respectively, whereas Gundlur,Inchigeri and Anabur tanks are located in north-east Karnataka in areas with similar agro-climaticconditions to the study mandals. Physiographicinformation relating to the tanks and theircatchment areas can be found in Table 10.

Experimental Mean No. Runoff (% mean annual rainfall)watershed annual of

rainfall years Forest Treated Untreated Barren(mm) Catchment agricultural Catchment area

Catchment

Chinnatekur 501 5 2 5 9G R Halli 569 7 - 2 - 9

Table 9. Summary of runoff from twoexperimental micro-watersheds

51

Gundlur Anabur Inchigeri Yapadinne Chapari

1990 2002 1987 2002 1990 2002 1987 2002 1986 2002

Wells:catchment 14 92 71 147 15 89 112 308 84 217command 8 8 58 134 34 72 1 1 27 98

Irrigation:catchment 22 143 86 178 11 241 186 536 132 343command 32 32 65 158 54 156 44 44 83 132

Structures:command 4 23 5 26 8 37 5 48 20 29

Irrigator 8 93 87 147 14 33 111 308 115 315farmers incatchment area

Tank inflow 61 35 138 93 40 20 285 213 70 63(ham)

Spillage 9/11 7/11 5/11 1/11 6/15 2/15 14/15 12/15 1/15 1/15frequency

Av. Spillage 25 10 4 2 9 1 187 135 7 7(ham)

Table 11 presents information on recent increases in the number of wells, irrigated area and irrigatorfarmers in the tank catchment areas. Table 11 also presents information on the increase in the number ofwater harvesting structures and the estimated impact that these have had on annual tank inflows, frequencyof years in which surplusing takes place and average annual spillage. These figures show that the additionalstorage created along drainage lines in the tank catchment areas in recent years has led to a decline in tankinflow, spillage frequency and average spillage in almost all cases. The reason is that runoff that wouldotherwise have run into the tanks is captured by the water harvesting structures. Once captured, the bulkof this water either evaporates or infiltrates and recharges aquifers which are pumped as sources of water forirrigation. In these semi-arid areas, water harvested by structures only makes a small contribution to baseflow which itself tends to be captured by downstream structures.

Table 10. Physiographic features of the five tanks

Table 11. Summary of changes in characteristics of the tank systems

Physiographicfeatures Gundlur Anabur Inchigeri Yapadinne Chapari

Altitude (m) 536-880 600-711 460-641 480-600 450-525

Rainfall (mm) 472 576 573 585 580

Soil type Alfisol Alfisol Vertisol Alfisol Alfisol / Vertisol

Catchment area (ha) 1234 2476 527 6423 2900

Average slope (%) 0.5-2.5 1.5-3.0 1-8 0.8-4.5 0.8-3.5

Tank capacity (ham) 27 144 34 42 68

Water spread (ha) 15 85 15 42 28

52

Bearing in mind the fact that some of thesetanks are more than 200 years old, Figure 30compares potential mean annual tank inflows(i.e. the naturalised runoff that would occur ifthere were no structures in the tank catchmentarea), mean annual tank inflows approximately 15years ago and mean annual tank inflows in 2002.It can be seen that in all cases there has been areduction in mean annual inflows during both theperiod from construction up until 15 years agoand from 15 years ago up until 2002. In the caseof Chapari, the biggest reduction took placeduring the period up until 1986 and, whereas forGundlur, the reduction was more marked duringthe period 1990-2002. Table 12 summarises tankinflow information that is specific to each tank.

Tank Specific characteristics relating to tank inflows

Chapari Major decline in tank inflow occurred approximately 100 years ago when two tanks wereconstructed in catchment area to meet the domestic water needs of Kalyandurg town.

Yapadinne Although there has been a large reduction in tank inflows, this is not leading tosignificant problems because the catchment area is very large in comparison to tank capacity .

Gundlur Sand extraction along feeder canals and stone quarries in catchment have createdadditional storage in the catchment area.

Anabur Original design of the tank relied on two feeder canals to fill the tank to full capacity.Unfortunately, one feeder canal was blocked by water harvesting structures and theother was poorly constructed and never functioned.

Inchigeri Inchigeri has a relatively small catchment area. Runoff from this area was reduced asa result of farmers blocking feeder canals with large field bunds. These have enabledthese farmers to cultivate paddy and sugar cane. Reduced inflows to the tank and theresulting reduced percolation to groundwater appears to be directly linked to the severedrinking water problems now faced by people living in Inchigeri town.

Table 12. Summary of tank inflow information specific to each

The findings presented in Table 11 are entirelyconsistent with other studies in the region thathave also reported reduced runoff as a result ofwater harvesting along drainage lines and withinagricultural and forested areas (e.g. Rama MohanRao et. al. 1993 and 1995). Soil waterconservation and construction of a large numberof check dams along the drainage lines is clearly amajor cause for reduced runoff. Groundwaterextraction in the vicinity of the drainage lines andwater harvesting structures is probably anadditional contributory factor because theresulting groundwater depletion allows a largervolume of water to infiltrate and be stored in theaquifer. However, the relative importance of thesetwo causes and possible interactions between thetwo could not be differentiated in this study.

Figure 30. Comparison of tank inflows with no structures incatchment, and the situation 12-16 years ago and in 2002

53

0

50

100

150

200

250

300

350

400

Gundlur Anabur Inchigeri Yapadinne Chapari

Tank

inflo

w (h

am)

No structures 12-16 years ago 2002

Figure 31. Gundlur tank surface water balance components

catchment areas during the last 12-16 years.It can be seen that there has been a big increase ineach case with Yapadinne having the largest netexpansion in irrigated area. Figure 32 also showsthe portion of the increased irrigation that couldhave come from increased groundwater rechargethat resulted from increased water harvesting inthe tank catchment areas. It can be seen that all theincrease in irrigation in Anabur could, in theory,be met by the increased groundwater recharge,whereas in Chapari virtually all the additionalirrigation must rely on “natural” groundwaterrecharge.

Figure 31 presents information on Gundlurtank’s surface water balance components duringan eleven year period. This figure gives anindication of the variability in these components.It also shows that the potential runoff, tankinflow and spillage is only loosely correlated withannual rainfall. Subsequent analysis carried out bythe WHiRL project has shown that the impact ofintensive water harvesting on tank inflows is mostmarked in percentage terms in years with lowrainfall.

Figure 32 compares the increase ingroundwater-based irrigation in the tank

Figure 32. Possible sources of water for increased irrigation in tankcatchment areas during the last 12-16 years

54

0

100

200

300

400

500

600

700

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998

Rai

nfal

l/Run

off (

mm

/yea

r)

Rainfall Runoff Tank inflow Tank spillage

0

50

100

150

200

250

300

350

400

Gundlur Anabur Inchigeri Yapadinne Chapari

Cha

nge

in c

atch

men

t irr

igat

ion

(ha)

Groundwater Structures

Increased groundwater development in thestudy mandals during the last 12-16 years hasresulted in many hundreds of farmers havingaccess to irrigation. Given the relatively highprofitability of irrigated agriculture in this region(Batchelor et al., 2000), the additional and morereliable income generated will have had asignificant impact on the livelihoods of thesefarmers and their households. Although there isnot a direct causal link between water harvestingand increased groundwater development(i.e. increased groundwater development is takingplace even in areas without water harvestingactivities), construction of structures and waterharvesting does increase the volumes of wateravailable in the tank catchment. Thereby, waterharvesting is contributing to the dramatic changesthat have taken place in recent years in terms ofpatterns of water availability and use.

Interestingly, reduced inflows to tanks havenot led to a widespread reduction in the total areairrigated in the tank command areas becausefarmers that used to rely on tank releases now usegroundwater as a water source. From theirrigation perspective and assuming that currentlevels of groundwater extraction are sustainable,changes during the last 12-16 years have beenentirely positive not least because these changeshave benefited poor and marginal farmers as wellas relatively richer farmers. However, if the non-irrigation uses of the tanks are considered, itbecomes obvious that the “irrigation” benefitshave come at a social and economic cost. In thelast 12-16 years the utility of many tanks hasdeclined (Yapadinne being an exception) foractivities such as washing, bathing, wateringlivestock and pisciculture. In some cases(e.g. Inchigeri), tanks are no longer a perennialsource of recharge for wells that meet thedomestic water needs of the village.

In many cases, the quantity of watersurplusing has been reduced and, hence, lesswater is available to downstream users. Whetherthe benefits resulting from a changed pattern ofwater use warrant the negative trade-offs dependson the specifics of the tank system and on whichsocial group is considered. It also depends on thescale considered. At a wider scale, the increasedagricultural production and potentially moreproductive water use may be more acceptablethan might be the case at the household scale.

In theory, it is possible to increase inflows intotanks without impacting on the incomes of thenew irrigator farmers in the catchment area.

This could be achieved in part by clearing feederchannels that are currently clogged and that arenot important sources of groundwater rechargeand, in part, by promoting more efficient andproductive use of water in the catchment area.However, without the removal of or modificationto larger structures and major changes in thesystem of incentives and disincentives underwhich farmers make decisions, it is unlikely thatany “additional” water resulting from productivityimprovements would flow into the tank. It ismore likely that individual farmers will use thiswater to irrigate a larger part of their holdings orto grow additional summer crops. Alternativelythis water will be accessed by farmers in thecommand area who do not currently have accessto water for irrigation. Promotion of communityor affinity group groundwater and surface watermanagement also has the potential to lead tomore productive and equitable water use. In somecases, this could involve the fitting of gates onlarger structures and the development ofoperating rules that involve letting early runoffevents flow to tanks. However, as the tank andtank command area are often located in differentvillage areas, with the exception of improvingfeeder channels, it seems that there are no win-win options that will see the tanks return toanything near their pristine state. Changes ingovernment policy and a shift towards demandmanagement could help. However, theprofitability of irrigated agriculture is such thatmany demand management measures will not beaccepted readily by farmers whose incomes arebased on using large volumes of water. This issuehas been studied by the APWELL Project whichhas also piloted novel approaches to participatoryhydrological monitoring (APWELL 2003a and2003b).

Although there is a tendency to considergovernment programmes aimed at increasingwater harvesting (e.g. watershed developmentprogrammes, source protection component of

Livestock near Gosanapalli tank, Dhone

55

3% 1% 9%

52%

17%

17%1%

Runoff

Urban water use

ET (irrigated areas)

ET (rainfed arable areas)ET (non-arable areas and ephemeral water bodies)

ET (forest areas)

GW recession

regard water harvesting as a totally benigntechnology. In contrast in semi-arid areas,downstream communities are becomingincreasingly aware of the problems causedby intensive drainage-line treatment inupstream areas.

4.4 Annual water balances

Figure 33 presents indicative estimates of thecomponents of the annual water balance forDhone and Kalyandurg at the macro-watershedscale. These estimates have been produced on theassumption that, on average, storage terms will beinsignificant. This figure shows that evaporationfrom rainfed arable areas is the largest componentof the water balance. In both mandals,evaporation from different surfaces or land uses isthe fate of approximately 95% of the rainfall.These figures contrast with the statewide waterbalance figures of the Andhra Pradesh WaterConservation Mission (Anon, 2003) whichsuggest that the fate of annual rainfall is asfollows: evapotranspiration – 41%, surface runoff– 40%, percolation to groundwater bodies – 9%and retained as soil moisture – 10%.

Figure 33. Annual macro-watershed or basinscale water balances

rural water supply programmes) to be entirelybenign, it is clear they can have a big impact onthe viability and utility of traditional tank systemsand on patterns of water availability and usewithin a tank catchment and command area.This impact appears to be most marked in lowrainfall years and when increased water harvestingin the tank catchment area is coupled withincreased groundwater extraction. The changedpattern of water use and associated changes inaccess to water for domestic and productivepurposes results in trade-offs and distinct winnersand losers. Although these trade-offs might beacceptable and, in some circumstances highlydesirable, they should be considered explicitly instrategic-level and village-level decision makingprocesses which, ideally, should be based onprinciples of integrated and adaptive waterresource management (Batchelor et al, 2001).It is clear also that programmes of tankrehabilitation should consider the multiple trade-offs both within the individual tank system(i.e. catchment area, command area and the tankitself) and the larger macro-catchment beforedecisions are taken on whether rehabilitation ofindividual tanks is needed. It is clear also thatprogrammes of tank rehabilitation must becareful to differentiate between symptoms andcauses of decline in tank utility and to be sure toaddress causes such as those affecting tankinflows.

The following points provide a summary of themain findings of analysis relating to theimpacts of water harvesting on patterns ofavailability and use:

. In the study mandals, water harvesting andgroundwater-based irrigation have inrecent years had a major impact on patternsof access and use of water for irrigation.This has led to major improvements inthe livelihoods of many households oncethey have paid off the debt incurred in theprocess of becoming irrigator farmers(e.g. borewell construction).

. In many areas, intensive water harvestingcoupled with over-exploitation ofgroundwater is impacting on downstreamwater availability and, in particular theutility of tank systems. Hence, significantnegative trade-offs are often associatedwith the changed pattern of use.

. With a few notable exceptions, governmentand NGO watershed development or ruralwater supply source protection programmes,

Kalyandurg

Dhone

56

3% 1%13%

55%

23%

4% 1%

5.1 Status of rural watersupplies

Assessments were made of the statusof all 225 and 438 domestic water pointsin Dhone and Kalyandurg respectively.The methodology used is described inbrief in Section 2.5. A more detaileddescription of the methodology and thefindings can be found in James (2002a)and James and Snehalatha (2002).

The following criteria and acceptablelimits were considered when classifyingeach water point as being satisfactory(i.e. `OK’) or as having a problem:

· Functionality: No major technicalproblems with pumps, standpipesor reticulation system.

· Distance to water points:The water point is not more than1.6 km away.

· Accessibility: Access to waterpoints is not denied as a result ofsocial exclusion. Subtle water-related discrimination betweenScheduled Caste (SC) users andnon-SC users was found to bewidespread; however, this did notnecessarily result in SC usersbeing denied access to water points.

· Water Quality: Acceptable from theusers’ view in terms of taste or abelief (based on experience)that the point supplied water withan acceptable level of fluoridecontamination.

5Access to waterresources

57

Fluoride problems in Battuvanipalli

In Battuvanipalli village nearKalyandurg town, villagers complainedof the symptoms of fluoridecontamination (pain in the joints,falling teeth, bleeding gums and brittlebones), even though the fluoridetesting of the water source by the RWSshowed a concentration of only 1.7 ppm(i.e. in excess of the permissible limitof 1.5 ppm). Action was not taken sincethe “operational” limit being used was2.0 ppm. Even worse, subsequentanalysis of this source by the WHiRLProject showed that fluorideconcentration was actually in excess of4 ppm.

· Adequacy of supply: Adequate supply ofwater for domestic uses, from the users’point of view.

· Peak summer availability: Water collectiontime and effort not much more during peaksummer months than in other months.

· Over-crowding: No more than 250 usersper water point.

Using these criteria and acceptable limits,Figures 34 and 35 summarise the users’ view ofthe status of water points in Kalyandurg andDhone respectively. A water point was classifiedas being a “problem” water point if, in the view ofthe users, it failed to meet the acceptable limits ofany one of the criteria. This assessment indicatedthat, although no water point was more than1.6 km away or had more than 250 users, 51%and 24% of water points in Kalyandurg andDhone respectively were either not functional orfailing to meet the adequacy of supply and waterquality norms of the Rajiv Gandhi NationalDrinking Water Mission (see Box 11). Thesefigures contrast starkly with official figures for“problem” water points that showed only a smallpercentage of water points as having a problem.One reason for this is the fact that the officialstatistics concentrate almost entirely on thefunctionality of water supply systems and onwhether or not a supply of 40 lpcd can beprovided. Also, the official statistics do not givemuch attention to problems of peak summeravailability and the problems experienced bywomen particularly in queuing for water duringthese lean times.

Washing clothes in the Pennar river

25 year old man suffering from skeletal fluorosis,Battuvani Palli, Kalyandurg

58

59

Figure 34. Users’ view of status of domesticwater points in Kalyandurg

Figure 35. Users’ view of status of domesticwater points in Dhone

Box 11: Criteria for Identification of Problem Habitations

A habitation which fulfils the following criteria may be categorised as a Not Covered (NC)/No SafeSource (NSS) Habitation:

(a) The drinking water source/point does not exist within 1.6 km of the habitations in plains or 100m elevation in hilly areas. The source/point may either be public or private in nature. However,habitations drawing drinking water from a private source may be deemed as covered only when thewater is safe, of adequate capacity and is accessible to all.

(b) Habitations that have a water source but are affected with quality problems such as excesssalinity, iron, fluoride, arsenic or other toxic elements or biologically contaminated.

(c) Habitations where the quantum of availability of safe water from any source is not enough to meetdrinking and cooking needs. [Estimated at 40 litres per person per day, including drinking(3 litres), cooking (5 litres), bathing (15 litres), washing utensils and house (7 litres) and ablution(10 litres)]

Hence in the case of a quality affected habitation, even if it is fully covered as per the earlier norms itwould be considered as a NSS habitation if it does not provide safe water at least for the purpose ofdrinking and cooking.

Habitations which have a safe drinking water source/point (either private or public) within 1.6 km. inthe plains and 100 m in hilly areas but where the capacity of the system ranges between 10 lpcd to 40lpcd, could be categorised as “Partially Covered (PC)”. These habitations would, however, be consideredas “Safe Source (SS)” habitations, subject to the water quality parameter.

All remaining habitations may be categorised as “Fully Covered (FC)”.

Source: RGNDWM (2000).

Figures 36 and 37 provide information on the most common water point problems encounteredby users in Kalyandurg and Dhone respectively. It can be seen that unacceptable water quality andnon-functional water points were the problems that were most frequently cited in Kalyandurgand Dhone respectively.

Figure 36. Types ofdomestic water supplyproblems,Kalyandurg Mandal,October 2001

Figure 37. Number ofwater points withdifferent types ofproblems,Dhone Mandal,March 2001

60

22

74

156

27

Non-functiona l Inadequate supply Poor quality Multiple problems

37

5

20

7

1

6

Non-functiona l Overcrowding Inadequatesupply

Unpredictable Poor quality Multipleproblems

The users’ assessments in Kalyandurg alsoconsidered a number of “minor” problems which,following appropriate capacity building, could besolved locally. These were:

· Leakage: Many pumps, standpipes, tapsand reticulation systems had leaks;

· Malfunctioning hand pumps: Many handpumps were malfunctioningas a result of inadequate maintenance;

· Unsanitary conditions around water points:Inadequate drainage around many waterpoints was the main cause.

Misleading Figures:The Case of Pathacheruvu

On paper, there is only one functioninghand pump for all 45 households ofPathacheruvu. A visit however showedup two agricultural bore wells near thevillage settlement (closer than the handpump) with sufficient water to meet alldomestic needs of those households,besides providing irrigation andlivestock needs. However, the handpump that was used most had a fluorideconcentration in excess of 2 ppm.

· The nature and intensity of problems varynot only across villages but also withinvillages. Some households were moreaffected than others since the nature of theproblems vary from water point to waterpoint. Moreover, detecting problems withdomestic water supply can be quitecomplicated. Water from a water point maybe used for different purposes (such as forlivestock, domestic uses, and irrigation)depending on the quality and quantity ofwater. Thus, a large number of water pointsin a village and/or adequate quantities ofwater at all water points may concealproblems with water quality (e.g. fluoridecontamination) or social discrimination.

.5.2 Water-related socialdiscrimination

Social restrictions on use of drinking anddomestic water sources by Scheduled Castes (SCs)and Scheduled Tribes (STs) were found in allvillages surveyed. These restrictions took twomain forms.

· SCs and STs cannot touch (‘contaminate’)open-well water, but can use public tapsand hand pumps. Wherever scarcity forcesvillagers to use open wells as a source ofdomestic supply, SCs and STs bring theirvessels to the well but cannot draw waterfrom it. They have to wait for upper castevillagers to fill their pots with water.

· In many villages, separate hand pumps orpublic taps have been set up in parts of thevillage where SCs and STs stay (commonlycalled ‘SC colonies’). When water is scarceand insufficient in the “upper caste” areasof the village, but available in the SCcolony, upper caste villagers come to filltheir vessels. The result is the SC and STfamilies have to wait till the upper casteshave taken their fill before collecting theremaining water from public taps or handpumps installed for their exclusive use.

In summary, important findings from theassessment of the status of domestic watersupplies included:

· Users’ views of the status of domestic watersupplies are not captured by the currentprocedures for monitoring rural watersupplies;

· There is a major disparity between theusers’ views of the status of rural watersupplies and official statistics;

Figure 38: Nature ofminor problems,Kalyandurg Mandal,October 2001

61

54

133

20

Major leakages Unsanitary surroundings Multiple problems

Nature of minor problem

Caste Restrictions in Kocheruvu

A large cruciform step well inthe centre of the old villagesettlement was being used fordrinking water in December2000 because the piped watersupply was not functioning as aresult of a major pumpbreakdown. Upper caste menand women walked down thesteps and filled their vesselsdirectly from the well. Lowercaste men and women, however,had to wait half-way down thesteps for an upper caste personto favour them by filling theirvessels. Not all upper castes didthis and sometimes the lowercaste person had to wait forsome time.

SC Colony in Manirevu

In December 2000, the newly-elected sarpanch had providedthe SC colony with 4 new publictaps so as to improve thecolony’s water supply that washitherto dependent on handpumps. But while public taps inthe main village are outlets fromsmall storage tanks that are, inturn, supplied by the pipedsystem, the new taps in the SCcolony are not connected tostorage tanks. This means that ifand when electricity fails andthe main pump of the pipedsystem stops working, the tapsin the SC colony run dry whilethe taps in the main villagecontinue to flow, using the waterstored in the storage tank.

5.3 Water-related functionalityof village-level institutions

Strong village-level institutions are a vital elementof the proposed reform of rural water supply andsanitation services coordinated by the Rajiv GandhiNational Drinking Water Mission (see Box 13). AndhraPradesh has an impressive history in forming women’sself-help groups (SHGs), and it is normal to expect thatvillage decision-making has been strengthened by thisdevelopment. The participatory assessments sought tofind out the extent to which village institutions like thepanchayat and SHGs, respond to water-relatedproblems.

In Dhone, assessments for the 42 villages surveyedindicated that in a majority of villages the panchayatwas unsympathetic to problems concerning watersupply (and sanitation). In Kalyandurg in contrast, itwas found that in almost half the villages the panchayatwas sympathetic and effective as far as water-relatedproblems were concerned. The results for Kalyandurgare summarised in Figure 39 along with the questionsthat were used as part of the ordinal scoring system.

As lower caste villagers are not allowed to take waterdirectly from the open well, a higher caste woman is filling

the pot of the lower caste woman, Kocheruvu, Dhone

62

Box 12: RWSS Sector Reform Project

This project is being implemented in 58 districts in 22 states all over the country, includingChittoor, Nalgonda, Prakasam and Khammam in Andhra Pradesh. This project advocates fourkey principles for sustainable RWSS systems:

1. Changing the role of government from provider to facilitator

2. Increasing the role of communities in the planning and management oftheir own facilities

3. Users paying all operation and maintenance costs, and at least 10% of the capital costsof supply

4. Promoting integrated water resource management

The institutional arrangements to implement the project include the formation of VillageWater and Sanitation Committees (VWSCs), as sub-committees of the Gram Panchayat, tomanage implementation and subsequent maintenance at the village level.

Source: Note on Andhra Pradesh Pilot Projects: Rural water supply and sanitation sectorreforms, Water and Sanitation Program – South Asia, September 2000.

QPA Ordinal Scoring Options Ordinal Scores Number ofvillages

Listens, but no action is taken 0 9Listens and acts, but no follow up and hence no result 25 11Listens and acts, but results come after a long time 50 8Listens and acts, gets quick results but not effective 75 5Listens, acts, and gets quick and effective results 100 21

Two main reasons were identified for the poor response by the panchayat. These were;. Apathetic leaders: These included sarpanches of the revenue village living in one of the

constituent habitations and paying relatively less attention to complaints from the otherhabitations.

. Fund constraints: Even when sarpanches were interested in taking action, funds wereoften a major constraint.

Figure 39. Panchayat Response to Water Related Problems, Kalyandurg Habitations, 2001

63

0

5

10

15

20

25

0 25 50 75 100

QPA ordinal scores (details in accompanying table )

No.

of h

abita

tions

Figure 40: Status of SHG functioning: Kalyandurg SHGs, 2001

5.4 Functionality of SHGs

The assessment of SHGs indicated that manywere being very successful in terms of thrift andcredit but few had any interest in or experienceof tackling water-related problems. This findingcasts some doubt on the proposals from somequarters to involve SHGs in water-relatedchallenges. In some villages, particularly thosewith weak panchayats, it was found thatspontaneous community action was leading tosolution of water-related problems. This action,however, even when it involved SHG members,was not organised by the SHGs.

Figure 40 presents the results of theparticipatory assessment of the functioning ofSHGs in Kalyandurg. This shows that the village-level perception is that 159 of the 268 SHGs (ornearly 60%) fall into category “Saving, effectiveutilisation of loans for at least one productivepurpose (with profits).” Figure 41 presents theresults of the assessment of SHG influence oncommunity decision-making in villages of Dhone.This shows that in 36 of the 42 villages (or 86%)surveyed SHGs were not effective in influencingwater-related community decision making. Thesefindings suggest that women’s SHGs will needconsiderable strengthening before they can take onthe role of change agents in the habitations in thetwo study mandals.

QPA Ordinal Scoring Options QPA Scores Number of SHGs

Only saving; no loans 0 15

Saving and loans only for consumption purposes 25 43

Saving, effective utilisation of loans & for at least 50 20one productive purpose (no results)

Saving, effective utilisation of loans & for at least 75 31one productive purpose (breakeven)

Saving, effective utilisation of loans & for at 100 159least one productive purpose (with profits)

Box 13: Some villagers’ view on theresponse of the panchayat to watersupply problems

“Since Sarpanch is residing in Chapiri,he has not taken any interest and noresponse, in spite of people’s request.”(Chapirithanda habitation, Chapirirevenue village).

“No response from panchayat, asSarpanch was from other village.”(Mallapuram habitation, Palavoyrevenue village).

“Panchayat listens, but no actiontaken, because of no funds topanchayat.” (Varli habitation, Varlirevenue village).

“When motor underwent repairs seventimes, on all occasions, the panchayattook immediate and effective action.”(Duradakunta habitation, Duradakuntarevenue village).

“Panchayat listens to problem and actsquickly in case of handpump repair atall times.” (Palavoy habitation, Palavoyrevenue village).

64

0

20

40

60

80

100

120

140

160

180

0 25 50 75 100

QPA ordinal scores (details in accompanying table)

No. o

f SHG

s

5.5 Participation of the poorand women in communitydecision-making

Even if a community decides to act on awater-related issue, the decision to act is notalways made collectively or democratically.The participatory assessments explored this issueand the results are summarised in Figures 42 and43. Figure 42 shows that in 91% of the villagesin Kalyandurg, the poor participate in meetingsand in 67% of the villages the poor bothparticipate and influence decisions affectingthem. Figure 43 shows that the situation inDhone is not so encouraging, in that the poorparticipate in only 50% of the villages, and thepoor both participate and influence decisionsaffecting them in just 16% of the villages.The relatively high figures for Kalyandurg can beattributed to the fact that an NGO, namely theRural Development Trust, has been working inKalyandurg for more than 25 years. One ofthe main thrusts of their work has beenempowerment of the poor.

Waiting to fill water pots, Kocheruvu, Dhone

QPA Ordinal Scoring Options QPA Scores Number ofHabitations

Groups do not play a role, it is left to individual members 0 10

Groups represent, but do not pursue, even to get assurances 25 8of future action

Groups represent pursue and get assurance of action, 50 18but no effective action occurs

Groups represent, pursue, and get effective action 75 3

Groups listen, act, and get quick and effective results 100 3

Figure 41. SHG influence on community decision-making, Dhone, 2001

65

0

2

4

6

8

10

12

14

16

18

20

0 25 50 75 100

QPA ordinal scores (details in accompanying table)

No.

of h

abita

tions

Box 14. Some villagers’ views ondecision-making involving the poor,Kalyandurg

“Because of disunity, disputes andmisunderstanding, nobodyparticipates in community decision-making.” (Balavenkatapuram).

“In spite of the large number of poorpeople, we are unable to influencethe decisions taken by the non-poor.”(Dodagatta).

“Since, it is a single community, all families are relatives.”(Pathacheruvu).

“Poor and non-poor participated indecision making and constructed tworatchakattas, one in the SC colonyand the other in the bus stand.”(PTR Palli).

“The whole village is participating inannual meetings (on every Shri RamaNavami) and they discuss and makedecisions about village problems andthe means to solve them.”(Vitlampalli).

“Poor could influence in gettinghousing scheme, pipeline, electricityetc.” (Bedrahalli).

“Non-poor give preference to poorin decision-making; and poor alsoaccustomed to give their opinionsto community. Repair work done toMariamma temple, ratchakattaconstructed in SC colony. Poorinfluence decisions concerningthemselves.” (Mudigal).

Hand pump in Obulapuram, Kalyandurg

Traditional practice of applying a mix of waterand dung to the ground in front of houses,

Pathacheruvu village, Dhone

66

Figure 42. Villagewise participation of the poorin community decision-making, Kalyandurg

Figure 43. Village wise participation of the poorin community decision-making, Dhone

67

Figure 45. Villagewise participation ofwomen in community decision-making,Dhone

Figure 44. Villagewise participation ofwomen in community decision-making,Kalyandurg

68

Women are not involved in community decision-makingWomen are in committes but do not attendWomen attend meeting but mostly let men take major ddecisionsWomen attend, participate but cannot influence major decisionsWomen attend, participate and are able to influence major decisionsNo data

Figure 44 and 45 summarise findings relatingto the participation of women in communitydecision-making in Kalyandurg and Dhonerespectively. These figures show that inKalyandurg women were only participating inmeetings in 17% of the villages and participatingand influencing decisions in 4% of the villages.In Dhone, women were participating in meetingsin 14% of the villages and participating andinfluencing decisions in 5% of the villages.For Kalyandurg in particular, there is a starkcontrast between the figures on the participationof the poor and of women in meetings.These findings confirm that, although poor menparticipate in meetings, women, and particularlypoor women, rarely participate in meetings.In summary, despite impressive advances inwomen’s empowerment through the SHGmovement in Andhra Pradesh, women still donot participate effectively in community decision-making. Even in a mandal like Kalyandurg,which has a good record in remedial actionby panchayats and community participation,participation by women in community decisionsis low. In a majority of the habitations (allexcept 2), women were either not involved incommunity decision-making, or did notattend meetings or influence major decisionson community issues relating to water andsanitation.

5.6 Access to irrigation water

Access to reliable water supplies for irrigation,livestock and other productive purposes isextremely important to the livelihoods of a largenumber of people living in the study mandals.Although there is some growth in the servicesector, the majority still rely heavily onagricultural production. There is a strong beliefamongst small and marginal farmers, based onexperience, that gaining access to water resourcesand becoming an irrigator farmer is a reliablemeans of escaping from poverty. Hence,considerable amounts of money have been andcontinue to be invested in borewell constructionin Dhone and Kalyandurg. Private borewellinvestments total around Rs. 4 crores in Dhoneand around Rs. 22 crores in Kalyandurg (basedon a cost of Rs. 35,000 per borewell at 2002prices). This level of investment reflects therelative profitability of irrigated cropping.However, over-exploitation of groundwater inboth study mandals has led to a situation wherebyinvestments in irrigation have become awidespread cause of poverty (see Box 16).

Overhead tank in Kocheruvu, Dhone

Box 15. Villagers’ views on women’sparticipation in decision-making

“Women are not included in communitymeetings. They are scared to participate.Women participate only for theirproblems, and are not involved incommunity problem solving.” (Manirevu).

“Men are not giving much importance towomen’s participation in communitymeetings.” (Obalpuram).

“Women are not encouraged to participatein community decisions, even thoughmost of them have minimum education.”(Kadadarakunta).

“Women are not being allowed and alsowomen are not interested inparticipating.” (Venkatampalli).

“Women mainly participate in VidyaCommittee meetings. But once they madea dharna for water problem at KalyandurgMRO office.” (Vitlampalli).

“Women participated in communitydecision-making; asked for communitylatrines. Decision taken and work alsodone.” (Mudigal).

“DWCRA groups participated in roadformation by shramadan. Women attendcommunity and Gram Sabha meetings andspeak about their problems.” (Golla).

69

Box 16. Mechanisms that link poverty to overexploitation and competition for groundwaterresources in crystalline basement areas

Failed borewell investments. Investment in well construction is a gamble with high risks,particularly in ridge areas and other areas with low recharge potential. Farmers who takeloans to construct borewells but are unable to find groundwater, will not be able to makerepayments and, in many cases, will quickly spiral into debt.

Higher borewell costs of latecomers. Latecomers to borewell construction often have to makelarger investments than firstcomers. This is because groundwater levels have already fallen,and siting a successful borewell involves drilling to greater depths. Also latecomers oftenhave smaller land holdings and, as a result, the scope for siting a successful well is morelimited. The net result is latecomers have to take larger loans and, consequently, are morelikely to default.

Competitive well deepening. Wells owned by rich farmers tend to be more productive and/orgenerate more income. If groundwater overexploitation takes place and water levels decline,richer farmers are more able to finance competitive well deepening. Also, as wealthy farmerstend to have established their wells before competitive deepening starts, they are in a muchbetter position to take new loans. As latecomers are often unable to finance competitive welldeepening, their wells fail and they are unable to repay loans and often have to sell theirland to moneylenders, who are in many cases the rich farmers.

Impacts on domestic water supplies. Overexploitation of groundwater for irrigation haslowered water tables in aquifers that are also sources of urban water supply. This has led toa reduction in supply, particularly, in peak summer and periods of drought. Collecting watertakes more time and involves carrying water longer distances. In some extreme cases,competition between agricultural and urban users is leading to complete failure of thevillage water supply. In these cases, villagers sometimes have to use water sources that arenot safe and suffer illness as a result. Illness usually represents both a loss of income andexpenditure on medical treatment.

Crop failure or low market prices. If crops should fail for any reason (e.g. major interruption inelectricity supply, wells running dry) or if there should be a steep fall in market prices ofproduce, farmers with large loans for borewell construction are extremely vulnerable. If theyshould fail to make repayments on loans, which typically have interest rates of 2% permonth in the informal money market, they can easily spiral into indebtedness with littlehope of recovery.

Falling groundwater levels. Falling groundwater levels in many areas have increased the riskof wells failing during periods of drought as there is no longer a groundwater reserve orbuffer to maintain supply during dry seasons and droughts. It is often poor and marginalfarmers that have borewells that are most likely to fail albeit temporarily during suchperiods.

Impact of intensive drainage line treatment. Intensive drainage line treatment as part ofwatershed development and other programmes can impact on the pattern of recharge.The net result is that, in semi-arid areas, some borewell owners can see the yields of theirborewells increase but others (usually located downstream) see their borewells becomeless productive.

Reduction in informal water vending. Informal markets for groundwater have emerged inrecent years in these parts of semi-arid India, as farmers with access to surplus supplies sellwater to adjacent farmers who either lacked the financial resources to dig their own wellsor had insufficient supplies in the wells they did own. Now as well yields decline, watermarkets are becoming less common as well owners keep all available supplies for theirown use.

Source: Batchelor et al. (2003)

70

5.7 Profitability of irrigatedcropping

Profits per hectare from irrigated agricultureare, on average, twice those from rainfed farming.The lowest profit from an irrigated crop is oftenhigher than the highest profit from a rainfedcrop. Using data from Dhone, Figure 46compares the profitability of rainfed and irrigatedcrops. Cultivation of some crops, however, is not`economic’, and farmers continue to cultivatelargely because they do not have to buy theinputs they require. Compared to areas underrainfed cropping, a relatively small proportionof the cultivable area of each mandal is irrigated.Although the large kharif areas under rainfedgroundnut are justified by the high relativeprofits, areas under other crops do not generallycorrelate with the net returns per hectare.Crops with high returns (e.g., mulberry, onions,vegetables) are grown on comparatively smallerareas because of local factors such as access tomarket, high cultivation costs, production risksand lack of local storage and processing facilities.Food crops (e.g., jowar), on the other hand, aregrown on larger areas than warranted by theirrelative profit because they provide other benefits(e.g. food security).

Water is not such a big constraint on non-agricultural activities such as brick-making,making pottery and operating tea stalls/kiosks.However, ready access to small volumes of waterat times when it is needed is still critical to thesuccess of such enterprises. More information onthe economics of different water use options canbe found in James (2000b).

Figure 46. Relative profitability of rainfed crops (brown bars) andirrigated crops (green bars), Dhone

Contour trenching

71

0

5

10

15

20

25

30

35

40

45

Seta

ria

Sorg

hum

Baj

ra

Cas

tor

Gro

undn

ut

Sunf

low

er

Red

gram

Cot

ton

Ben

galg

ram

Brin

jal

Tom

ato

Cot

ton

Padd

y

Chi

lli

Oni

on

Net

retu

rns

(tho

usan

ds R

s /h

a

6.1 Water-related constraints

Constraints that limit the number ofresource-focused options that can and shouldbe promoted by the APRLP and similarprojects include (not in order of importance):

· Climate: The climate of the study areasis semi-arid and, consequently, rainfallis extremely variable in time and space.Hence, groundwater recharge andrunoff into tanks is also extremelyvariable, as is the productivity ofrainfed arable and non-arable lands.

· Extreme events: Meteorological droughtand floods are recurring naturalphenomena that have a relativelygreater impact on the poor andvulnerable.

· Aquifer characteristics: Hardrockaquifers are prevalent in the studyareas. These aquifersstore onlylimited quantities of water in shallowweathered and deeper fractured layers.

· Water resource availability: In manyareas, current usage of water resourcesapproximates and even exceeds annualreplenishment. In the absence ofinter-basin transfers, the scope foraugmenting water resources bydeveloping additional ground andsurface water resources is limited.

Constraints and risks6

72

· Indebtedness: Small and marginal farmerstend to have high levels of debt. Debtrepayments and the cost of borrowingreduce their ability to use land andwater resources efficiently. Levels ofindebtedness amongst relatively richerfarmers have also risen for reasons that arelisted in Box 16. In some villages, this isleading to a consolidation of power withina relatively small elite group.

· Electricity supplies: Frequent and prolongedpower cuts, particularly during summermonths, encourage farmers to over-irrigatewhen power is available. Power surgesoften damage transformers and, thereby,cause shortages of pumped water fordomestic use and irrigation.

· Awareness: The Water Audit has shownthat, in general, there is a lack ofawareness at all levels, of the severity andcomplexity of water resource problems inthe study mandals.

· Short-termism: There is a belief, generatedin part by watershed developmentpropaganda, that there are quick fixes tothe water-related challenges facingcommunities in semi-arid areas of southernAndhra Pradesh.

· Water-related myths: A number of water-related myths have become acceptedwisdom at the policy level in AndhraPradesh (see Box 17). Subscription to thesemyths can lead to poor decision-making,ineffective policies and wastage offinancial and human resources.

· Government policy: Currently, thereare few incentives or disincentives toencourage individuals or groups ofwater users to maximise water useefficiency and/or productivity. This isdespite there being an active debate onthis topic.

· Specialist knowledge: Although specialistknowledge and experience exists, this isnot always readily available to, or usedby, villagers and the implementors ofwatershed development or rural watersupply programmes. Careful targeting ofinterventions to particular physical, socialand institutional settings is rarelypracticed. Consequently, the tendency isto fund and implement the sameinterventions everywhere, regardless ofthe setting or priority needs.

· Poor quality information: Much water-related decision-making is based on officialstatistics that are incorrect and/or out ofdate.

· Population increase: Year by year a largerproportion of groundwater recharge andsurface water storage is needed to meetdomestic and urban water requirements.As a consequence, the water available forother uses is reducing. In addition, withincreasing population and in the absence ofimprovements in water use productivity,there will be a decrease in the proportionof the population that can rely on land-based activities as a main source of income.

· Small and fragmented land holdings:In general, farmers have holdings thatare becoming increasingly fragmented.This makes good land husbandry and goodwater management difficult.

Water tanker operating near Anantapur in March 2003

73

Box 17. Water-related myths

Water-related myths that were found to be relatively common during the Water Auditincluded:

. Water harvesting is a totally benign technology. Although water harvesting technologiescan produce huge benefits, intensive drainage line treatment, in particular, cansignificantly reduce water resource availability to “downstream” communities. In somecases, this negative trade-off does not matter; in others, severe hardship can result.

. Planting trees increases local rainfall and runoff. The reality is that forests exert asmall, almost insignificant influence on local rainfall and, notwithstanding a smallnumber of exceptions, catchment experiments generally indicate reduced runoff fromforested areas as compared to those under shorter vegetation.

. Runoff in semi-arid areas is 30-40% of annual rainfall. Although localised runoff andrunoff from individual storms can be high, annual runoff at the micro-watershedscale (or greater) in semi-arid areas tends to be much lower than 30-40%.

. Rainfall has decreased in recent years. With few exceptions, studies of long-termrainfall records, that have used data from a single set of rain gauges, have not showna significant decrease (or increase) in mean annual rainfall.

. Aquifers are underground lakes. The reality is that check dams and other water-harvesting structures usually only have localised impacts on groundwater levels andaquifers rarely behave like underground lakes. Or put another way, localised recharge inone place rarely leads to an immediate rise in groundwater levels many kilometres away.

. Water use of crops depends mainly on crop type. A common misconception is that thedaily water use of crops is directly related to the crop type and that evaporation rates aremany times higher from some crops as compared to others. The reality is that, assumingthat a crop is well supplied by water and has a full canopy (i.e. the crop completelyshades the ground), the daily rate of evaporation is driven primarily by themeteorological conditions (e.g. radiation, wind speed, dryness of the air).

. Aquifers once depleted stay depleted. A pessimistic view of aquifer depletion is that thisis an irreversible process. The reality is that, in most cases, aquifers can be re-establishedor replenished as long as the balance between recharge and extraction is swung towardsrecharge. This can occur as a result of increased recharge, decreased extraction or both.

6.2 Water-related risks

Experience in the region has shown that thereare a number of water-related risks associated withwatershed development and with the sourceprotection measures implemented as part of ruralwater supply programmes. Theseinclude:

. Increased borewell construction andincreased irrigation by individuallandowners: In most cases, the primemotivation of farmers to become involvedin soil and water conservation programmesand the construction of waterharvesting structures is not altruistic. It isto increase the water resources that areavailable to them for irrigation.At the

watershed scale, this is justified only if theresources that are being “harvested” do nothave a higher value if they are put toother uses.

· Increased borewell construction andincreased irrigation by “poor” landowners:Successful watershed development projectsoften improve the financial status ofrelatively poor farmers such that they areable to take loans for constructingborewells and installing pumps. As above,this is fine as long as the water they“harvest” does not have alternativehigher-value uses (e.g. as a source ofdomestic water supply).

74

· Deterioration in village water supplies:There is a high risk that projectinterventions will reduce runoff into tanksor other structures that are importantsources of recharge of the aquifers thatmeet urban water requirements. Projectinterventions can also lead indirectly toincreased pumping of groundwaterfor irrigation in urban and peri-urbanareas, thereby, increasing the risk offailure of village water supplies duringthe summer months. Finally, projectinterventions can lead to increasedconsumption of water: per household, bylivestock, by horticulture within the villageand by non-land-based activities.

· Conflicts within villages and betweenvillages: Some interventions, that involvechanging land use or patterns of wateravailability and use, result in distinctwinners and losers. If there is a risk of thishappening, conflicts should be managed byensuring that losers are compensated insome way. As above, decisions on whetheran intervention should take place shouldbe based on economic and social value.

· Reduction in net productivity: There is arisk that promotion of interventions witha high social value will lead to reductionsin net productivity at the village orwatershed scale. For example, use of waterfor irrigation on marginal lands (usuallyowned by poorer farmers) will tend to beless productive than use of the same wateron better quality land (usually owned byrelatively richer farmers). Ultimately,determining the balance betweenacceptable social and economic value isa political decision.

. Increase in environmental degradation:It is generally assumed that an increasein forestry equates to environmentalimprovement in watersheds and that thisis sufficient in terms of meetingenvironmental sustainability targets. Inmany cases, increased forestry will lead tosignificant improvements in biodiversity,particularly if indigenous tree species areplanted. There are risks, however, thatchanging patterns of land and water useand, hence, the hydrology of watershedswill lead to reduction in biodiversity inareas other than forested areas (e.g. inwetland areas, in ephemeral streams,in and around tanks). There is also a riskthat project interventions will adverselyaffect water quality (e.g. pollution resultingfrom: increased use of agro chemicals).Kocheruvu open well, Dhone

Waiting for water at a water point inYapadinne, Dhone

75

7.1 Impacts of watersheddevelopment

Arguably, “traditional” watersheddevelopment has become synonymouswith the construction of check damsand other water harvesting structures.Soil and water conservation activitiesare seen as an initial step towardssustainable development and as acost-effective means of disbursingsubstantial amounts of funding.Box 18 summarises the positive andless positive aspects of “traditional”watershed development. Therecommendations that follow inSection 8 are aimed at addressing theless positive aspects of currentapproaches whilst not negating thepositive aspects and benefits.

Water resource planningand option selection7

76

Box 18. Positive and less-positive aspectsof “traditional” watershed development

Results from the Water Audit indicatethat positive aspects of the ongoingwatershed development programmeinclude:

· Increases in net agriculturalproduction on arable andnon-arable lands;

· Development of village-levelinstitutions;

· Substantial improvements inthe livelihoods of some socialgroupings;

· Implementation of an approachthat has widespread politicaland public support.

The less positive aspects of theprogramme in dryland areas include:

· Certain groups capture waterresources, often at the expenseof the poor;

· New village-level institutions areusually outside government and,consequently, they often have a shortlifespan and minimal political orlegislative support for any actions ordecisions that they might take;

· Protecting drinking water suppliesis not seen as an integral part ofwatershed development;

· Emphasis is on development of waterresources (i.e. on increasing watersupplies by constructing check dams,rehabilitating tanks, etc.) and not onmanagement of water resources(i.e. on managing demand and onmaximising the social and economicvalue of water).

· As planning takes place at the village-level, a whole range of wider issuesis ignored (e.g. upstream-downstream equity, inter-villageequity, flood protection, droughtpreparedness, pollution of watercourses, biodiversity and protectionof rare habitats etc.)

· Watershed development publicity orpropaganda (e.g. wall paintings,street plays etc.) is often misleadingin that it suggests that there arequick fixes to water-related problemsin semi-arid areas.

After: Batchelor et al. (2003)Masonry check dam

77

7.2 Adaptive water resourcemanagement

The history of water management is, in manyways, a history of searches for universallyapplicable solutions (Moench, 1999).The “development era” focused on constructionof large-scale infrastructure projects such as damsand municipal supply schemes. In a similarmanner, many now advocate economic pricing,basin approaches, integrated planning, thedevelopment of participatory water managementinstitutions and demand management as “the”solution to management needs. The principle ofadaptive management turns many current debateson resource management on their head. Instead ofgetting bogged down debating whether or notdecentralised participatory approaches are betterthan centralised ones for water management, theprinciple of adaptive management involvesworking from the specifics of a given problemoutwards, to the best solution for addressing it.Some problems may be best addressed throughdecentralised participatory approaches; others mayrequire more centralised forms of planning andintervention (Moench, 1999, Batchelor 2001).

Adaptive management is also a process bywhich management actions and directions areadjusted in the light of new information oncurrent and likely future conditions. It recognisesthe limitations of current knowledge and thepotential impacts of current resource managementstrategies and also recognises that there is adisjuncture between the inherent variability ofnatural and social systems and the tendency formanagement approaches to cluster around afew management models. New managementparadigms emerge in response to the limitationsin earlier ones. When these new paradigmsbecome dominant, their inherent limitationsemerge and they are gradually discarded infavour of new “better” paradigms.

The rapid pace of technical, social andeconomic change in Andhra Pradesh is such thatit is very difficult to predict future patterns ofwater supply and demand with any greatconfidence. The possible impacts of short andlong term climate changes add furtheruncertainty. This being the case, resourcemanagement systems are required that are flexibleand able to adapt to new challenges.

7.3 Integrated water resourcemanagement

In many areas, competition and equitableaccess to water resources was not an issue as longas the size of the “cake” was growing. Now,however, competition over access is emerging as amajor issue in semi-arid areas and one that hasthe potential to have a serious long-term impacton the livelihoods of people living in these areas.

Lack of integration between the sectorsinvolved in water resources planning andmanagement is a fundamental impediment toimproved water resource management. In general,there is little integration between the watersheddevelopment, water supply and sanitation (WSS)and power sectors despite the fact that all threesectors impact hugely on the availability andaccess to water resources and, hence, on thelivelihoods of urban and rural communities. Forexample, in semi-arid India, an unintendedimpact of watershed development and RWSSsource protection measures is an increase ingroundwater extraction for irrigation that canlead to an increasing risk of resource-relatedfailure of village water supplies. A potentialunintended impact of power sector reform is alsoa big increase in groundwater extraction that mayalso impact adversely on rural drinking watersupplies. Given that the profitability of“commercial” irrigated cropping is so high, newtariffs are unlikely to limit groundwaterextraction by larger farmers. However, they arelikely to make pumping costs too high for poorand marginal farmers that are growing subsistencecrops. In contrast to the previous two examples,in many areas, the livelihoods of all farmers(i.e. the relatively rich and the poor and marginalfarmers) are being affected by the reallocation ofwater resources, that they were using forirrigation, to meet the ever increasing demands ofurban areas.

Regardless of official policy (or norms), accessand allocation issues tend to have major politicaland economic ramifications. Furthermore themechanisms used for allocation, touch deepcultural, ethical and often religious sensitivities.Clearly integrated water resource management isneeded that involves identification of the multipleimpacts and trade-offs resulting from current orproposed sectoral policies. Once these impactsand trade-offs are identified (by water auditing orother procedures) it is then up to the politicalprocess to make informed decisions.

78

7.4 Water-related trade-offsand win-win options

Opportunities for improving water useefficiency or productivity in the studymandals fall into four categories(after Seckler, 1996):

· Increasing output per unit ofevaporated water;

· Reducing pollution anddegradation that diminishesthe value of usable water;

· Reallocating water from lowerto higher value uses;

· Reducing losses of usable waterto sinks.

All the options listed in Section 8 fallinto one or more of the categories listedabove. Wherever possible APRLP shouldpromote win-win options. However, when arange of scales is considered, there are fewoptions that can be classified as being win-win, regardless of the physical and socialsetting. Hence, minimising trade-offs bymatching options to a given setting iscrucially important.

Using IWMI’s terminology, most of theriver basins in southern Andhra Pradesh arenow approaching the status of being closedbasins whereby utilisable outflows are, onaverage, fully committed. Hence, there isonly limited justification for attempting toaugment water resources by creatingadditional storage in the form of, say, checkdams with high crest heights, nala bundsand small dams. There is, however, a strongargument for ensuring that water isimpounded where it provides the highestvalue range of goods and services. In somecases, achieving this objective might involveinstalling new structures along drainagelines and removing existing structures,lowering their crest heights or installingsluice gates that are left open during at leastpart of the rainy season.

Vegetable market

Watering livestock from a domestic water point inManirevu, Kalyandurg

79

8.1 Summary of main recommendations

Analysis of information from the study mandals and subsequent discussions at thedistrict and state levels with senior government officers, line department specialists,senior researchers, NGO staff and APRLP staff led to the identification of six mainrecommendations. These are summarised below and discussed in greater detail insubsequent sub-sections.

Recommendation 1. Projects involving watershed development-typeactivities should promote a wider range of options/interventions and,in particular, options/interventions aimed at: protecting drinking watersupplies, improving the access of poor households to water for productivepurposes and reducing the impacts of droughts on rural livelihoods.

Recommendation 2. Projects involving watershed development-typeactivities should target and match interventions to the specific physical,social and institutional settings.

8 Recommendations

80

adopt and build capacity at all levels in adaptiveand integrated water resources managementprinciples and methodologies.5

8.2 Recommendation 1 –Wider range of options

Although the study findings suggest a rathergloomy state of affairs in terms of the currentand future status of water resources in the studymandals, one positive conclusion is that there area large number of water management optionsthat could be promoted by APRLP. These aresummarised in Tables 13-19. All these optionshave the potential to increase the productivity ofwater use and/or to improve equitable access towater resources. However, matching of optionsto particular physical, social and institutionalsettings is crucial. As part of this targetingprocess, particular consideration should be givento the trade-offs associated with each option orsequence of options.

5 GWP (2000) provides a good description of theprinciples of integrated water resources management.

Options Specific details Potential trade-offs

Increasing the . Selection of appropriate crops Drainage and, hence,productive use . Use of good genetic material groundwater recharge mayof water . Seed priming be reduced as a result of

. Good crop nutrition increased vegetative cover

. Good weed and pest control and healthier deeper-

. Minimising post-harvest losses rooting crops

. Intercropping systems

. Making use of specialist advice

Reducing soil . Planting early in Kharifevaporation . Maintaining crop cover and/or

mulches during rainy periods. Managing crops and cropping

systems in response to specificweather patterns (see Figure 47)

In-situ soil and moisture . In-field soil moisture techniquesconservation matched to the different soil type,

slope and other factors(see Figures 48 and 49)

Table 13. Rainfed arable cropping options

Recommendation 3. Village-level water-related participatory planning should takeplace within a wider district planningframework.

Recommendation 4. There should be amuch greater emphasis on waterresource management and a shift ofemphasis from supply to demandmanagement of water resources.

Recommendation 5. As part of a largerprogramme of making better use of water-related information in decision-makingprocesses at all levels, GIS-linkedparticipatory assessments should becomepart of routine M&E of rural water supplyand sanitation programmes.

Recommendation 6. A major effort isneeded to update and improve the qualityof the water-related information that isbeing used to underpin water-relateddecision making at all levels. Effort shouldalso be directed towards makinginformation more accessible to potentialusers, particularly at the district level.

The recommendations listed above shouldbe viewed within the overall recommendationthat the APRLP and similar programmes should

81

Table 14. Rainfed non-arable options

Options Specific details Potential trade-offs

Making more · Alternative land use systems Changing land use andproductive use of · Dryland horticulture management of common-water on CPR lands · Silvo-pastoral systems pool resources may result

· Income generation, in distinct winners andparticularly, during droughts losers. The rights offrom timber and non-timber existing usersforest products (e.g. livestock owners, gatherers

of fuelwood) may not becatered for in newmanagement arrangements.In some cases, soil andwater conservationmeasures and improvedvegetative cover mayreduce recharge and runoffto tanks

Improved management · Joint forest managementof CPR land · Community grazing schemes

· Community fuelwood schemes

Making more · Planting of grassesproductive use of and fodder legumeswater on privately- · Application of fertilisersowned non-arable · Establishment of energy coppicesland · Dryland horticulture

Concentrating rainfall · Directing runoff to arablewhere it can be used areas and/or to tanksproductively · In-situ soil moisture

conservation (see Figure 49)

Reducing soil · Maintaining cover withevaporation vegetation with high economic,

social environmental value· Concentrating rainfall

wherever soil evaporationlosses can be minimised(e.g. near useful vegetation)

82

Figure 47. Cropping systems for improved rain water management

83

Cropping system for improved rain water management

“Normal”rains

Delayed onsetof monsoon

Early withdrawalof monsoon

Black soil Red soil Black soil Red soilBlack and

red soil

Shallow Medium Deep Shallow DeepShallow

and deepAll

depths

Shallow DeepMedium

. Bajra,sesamemouthbeanand spreadinggroundnut

. Spreadinggroundnut +red gram

. Rabi.sorghum,sunflower,safflower,bengalgramand coriander

. Safflower +bengalgramjowar +bengalgram

. Greengramfollowed byjowar,sunflower orbengalgram

. Seteriafollowed bysunflower

. Groundnutfollowed byhorsegram

. Redgram,castor

. Groundnut +redgram,sorghum +redgram,ragi + redgram,ragi + dolichos,cowpea +castor

. Jowar,bajra, ragi,groundnut,sunflowerfollowed bycowpea,horsegram oronion, maizefollowed byhorsegram

. Avoid cashcrop and sownormal rabicrop

. Sow spreadinggroundnutand relayjowar

. Groundnut,hybrid bajra,sunflower +seteria,redgram,cowpea,horsegram

. Sow improvedjowar varietyupto 10 October,local jowarupto 30 Octoberand after 30October jowarfor fodder

. After 15 Julyany crop butgroundnut(eg bajra,sunflower orredgram aspure standor ragicombinedwith theabove)

. Ratoon cerealcrops ifpossible

. Uprootsensitivecropcomponentsin mixed orintercroppingsystems

. Reduce plantpopulation byremovingalternate orevery thirdrow in caseof determinatecrops

. Supplementalirrigation iffeasible

84

Figure 48. Soil and water conservation measures

Figure 49. In-situ soil and moisture conservation

Soil and water conservation measures ( rainfall < 600mm)

Black Soils Red Soils

Shallow ShallowMedium Deep Deep

Contourbunds withopen ends orwaste weirs

10m widecontourborders

Contourbunds withopen ends orwaste weirs

Zinggterraces withraised wasteweirs

Levelling oflower 30% ofslope

Gradedbunds, cont.borders;Zinggterracesdependingupon intakerates

10m widelevelledcontourborders

Contourbunds withopen ends orwaste weirs

Dead furrowswidth:40-60 cmdepth: 15 cm

If WHneeded,gradedbunds

Deadfurrowswidth:40-60 cmdepth: 15 cm

In-situ soil moisture conservation ( Inter-terraced areas)

Shallow Medium Deep DeepMediumShallow

Red Soil Black Soil

Formation of contour borders

Summer deep ploughing Deep tillage

Soil crusting controlled by periodic tillage or byincreasing organic matter

Set-row system. Crop rows in same position each yearto improve soil organic matter

Dead furrows every 2-3m planted with intercrops.

Application of P fertiliser

Ridge and furrow system

Dust mulching and/or organic mulching

Compartmental bunding: 0-0.5% slope no bunding,0.5 to 1% slope 6x6 m, 1-2% slope 4.5x4.5m, 2-3% slope3x3m

If infiltration rates are high, tied ridges up until sowing.Width 0.45m, spacing 1-2m

Broad bed 1.5x1.5m and furrow 0.45m up until harvest

Rubble bunds on contour for slopes >0.5%. 0.3m high,0.5m base width and 0.3m vertical interval

Dust or organicsurface mulch

Surface and vertical organicmulches

Dead furrows every 2-4mplanted with intercrops

Figure 50. Treatment of non-arable lands

Treatment of non-arable lands ( rainfall < 600mm)

Construct a trapezoidal diversion drain with 0.2-0.5% bed slope large enough todivert runoff from upstream area. Spoil on downstream side of drain. Runoffdiverted to recharge pond and used to establish a community irrigation scheme

Hills (betta)(slope >10 %)

Mounds (Dibbe)(slope 5-10 %)

Wastelands(slope 0-5 %)

Gully lands(halla)

Exclude biotic influences (e.g. cattle) by establishing a social fence

Construct adiversion bund.0.4-0.6m2 x- section.Vegetativeprotection required

Easen the sideslopes and plantwith trees

Construct smallearthern bundsacross the gully1m wide and 0.15mhigh

10-20m intervalplant bundswith suitablevegetation

Silvi-pastoralsystems

Contour trenchesor crescent shapedpits 4-10m spacing

Plant trees in pitsor trenches andsuitable grasses inbetween pits ortrenches

Apply 20-25 kgDAP/ha

Gradonis 5-10mwide depending onthe slope

Beds of Gradonisplanted withappropriategrasses and treesplanted on thebenches

Apply 20-25 kgDAP/ha

Contour trenches0.5x0.5x4m in size.Trenches at 10mhorizontal intervalplanted with treesat 1m interval

Catch pits betweentrenches.0.5x0.5x0.5m insize. Trees plantedin pits

Seed rest of areawith harmatafodder legumes etc

Apply 20 kg DAP/ha

Threshing redgram in Kocheruvu, Dhone

85

Options Specific details Potential trade-offs

Increasing useful · Select high value crops Although farmers may increaseoutput per unit · Crop varieties responsive the efficiency or productivity ofof water to irrigation irrigation water use per unit(more crop per drop) · Select good genetic material area this does not necessarily

· Good pest and weed control mean that they will reduce· Good nutrition their total irrigation water use.· Minimise post-harvest losses In some cases this might· Minimise water lost during happen but, for many

land preparation farmers, it is water that is· Minimise risks of soil limiting rather than land.

salination These farmers are likely to useany surplus water resulting fromefficiency gains to increase theirgross irrigated area. Hence,overall improvements in wateravailability or equity, resultingfrom increased water useproductivity or efficiency may benegligible or non-existent

More effective use of · Plant early in Kharifrainfall · Good irrigation scheduling

Better in-field · Land levelling in surfacedistribution of water irrigated areas

· Adopt sprinkler ordrip irrigation

Localised irrigation · Use drip, pitcher or subsurfacepipe irrigation or similarwhere appropriate

Reduce conveyance · More reliable electricitylosses supplies making it possible

to pump water direct to fieldsinstead of into storage tanks

· Line channels and storage tanks

Reduce need for · More reliable electricity“insurance” irrigation supplies

· Community-management ofgroundwater

Reduction in · Localised irrigationsoil evaporation · Select crops, crop varieties

and cropping systems thatshade the ground effectively

· Use mulches where appropriate

Establishment of · Sharing of wells and pumps Increased irrigation,community irrigation is quite common amongst for whatever purpose,schemes members of the same family may exacerbate groundwater

· A number of farmers using the depletion and competitionsame well but with their own for water between domesticpumps is also possible and agricultural water users

· Many different “share cropping”arrangements are also possible

Table 15. Irrigated land options

86

Table 16. Management of groundwater resources

6 A collector well is a shallow hand-dug wellof large diameter with horizontal boreholesdrilled radially from the base to a distance ofapproximately 30m, typically in four directions

Options Specific details Potential trade-offs

Community Findings and recommendations There is a risk that communitymanagement of the DFID-supported management of groundwater willof groundwater COMMAN Project (Community lead to exclusion of some social

Management of Groundwater groups.Resources in Rural India)can be found via:www.bgs.ac.uk/hydrogeology/comman/home.html

Improving groundwater · Ideally in areas in which it has Improving groundwater rechargerecharge highest social, economic and in one part of the watershed may

environmental value be at the expense of existing· Range of water harvesting users elsewhere

techniques can be used

Demand management · Different options are listedin Section 8.5

Promote use of shallow · Reduces energy required for Initially groups of farmers willwells for irrigation and pumping have to reduce groundwaterdeep wells for RWSS · Helps re-establish a groundwater extraction until the shallowparticularly in peri-urban buffer that can meet domestic aquifer is replenished. Once it isareas water needs during dry seasons replenished, they can return to

and periods of drought extracting groundwater at rates· Farmers who adopt measures that are equivalent to annual

to improve groundwater recharge recharge. The transition costs ofare more likely to see benefits as this approach would be high asrecharge of shallow aquifers tends in many areas farmers wouldto be relatively localised when have to switch from usingcompared to recharge of deep borewells to wide diameter wellsaquifers or collector wells6. Note that

this approach will only besuccessful if carried out inconjunction with some level ofdemand management

87

Options Specific details Potential trade-offs

Village-level · Needed to ensure that attention Local-level ownership of planningparticipatory is given to the many surface processes may be reduced.planning within water resource management Additional work for district-levela District Planning issues that are not addressed line department staffFramework during local-level participatory

planning. See Section 8.4

Re-establishing · Fix gates in drainage-line structures In most cases, this option willinflows to tanks and leave these open result in distinct winners and

except during early and losers. Increasing tank flowsmid-monsoon. significantly can only be achieved

· Instead of surface structures at the expense of current waterwith high crest heights, users in the tank catchmentlow check dams using loose areasboulders or subsurface“dyke” barriers

· Re-establish shallow aquifers andthereby improve base flows

· Renovation and clearing of existingfeeder and diversion channelsto improve runoff to the tank

· Remove some gully-control structuresor reduce their crest heights

· Fill in old brick pits

Establish or · Based on affinity groups that There is a risk that changing andre-establish represent the multiple uses of tanks introducing tank managementtank management · Groups should be linked to procedures will lead to exclusionsystems Panchayati Raj institutions of some social groups

Repair tank sluices · Should be carried out with Seepage losses may be anand bunds community involvement / important source of recharge

supervision downstream of the tank

Reduce evaporation · Reduces the surface area Deepening tanks may reduce thelosses by deepening to volume ratio fodder and forage value of thetanks or ponds · Increases the storage capacity areas on which grasses grow as

the tank water recedes. Thisoption may also impact on theenvironmental and biodiversityvalue of tanks and the areassurrounding tanks

Reduce siltation · In-field soil and water conservation Less silt available to farmersof tanks measures that traditionally use silt as a

· Gully control structures means of improving soil fertilityof their fields

Pisciculture · Increase number of water bodiesused for pisciculture

· Establish local management groups

Table 17. Management of surface water resources

88

Options Specific details Potential trade-offs

Use waste-water · Drain runoff of acceptable Reduced groundwaterfor productive quality to areas where it can be rechargepurposes used for fodder or horticulture

· Minimise health risks· Establish user groups

Increased · Direct good quality runoff from There is a risk that pollutedgroundwater roads, open areas, rock water will also enter the soakrecharge outcrops, roofs etc. into soak pits resulting in pollution of

pits aquifers

Harvesting of · Good quality runoff from Reduced groundwater rechargewater into cisterns roads, open areas, rock

outcrops, roofs etc. can be pipedinto private or community-ownedcisterns. This is a good option inareas with polluted groundwateror rapid groundwater recession

· Water can be used for domestic orproductive purposes

Minimising risk of · This includes safe handling of Additional costs may beground and surface agro-chemicals and industrial incurred by small and largewater pollution effluents industries

· Treatment and disposal of sewage

Establishment of · Using harvested or piped-watercommunity or nutrition supplies, allotment-typegardens gardens can provide landless

families with a source of vegetablesfor home consumption or sale

O&M of environmental · Action by community groups Less silt available tosanitation systems or the panchayats farmers that traditionally use

· Improves the urban and silt as a means of improvingperi-urban environment soil fertility of their fieldsand reduces health risks

Flood control · Maintenance of drains androutine clearance of rubbish fromstreams flowing through urbanareas

· Construction of bunds to protecthousing at high risk of flooding

Table 18. Management of water resources in urban and peri-urban areas

89

Options Specific details Potential trade-offs

Improve quality · Identification of waterof drinking water point with levels of F,

TDS etc. above permissiblelimits

· Identification of alternativesources of water

· Household water treatment

Source protection · Increasing recharge near Reduced watermeasures to domestic groundwater availability for

sources and controls on agricultureextraction for other uses

· Minimising pollution riskto water sources

· Establishment of ground orsurface water reserves(or buffers) that are sufficientto meet domestic water needsduring dry seasons and periodsof drought

Community management of · Collective responsibilityground and surface water and management of waterresources resources developed ideally in

conjunction with the panchayats· Establishment and management

of a basic human needs reserve· Enforcement of legislation

relating to spacing betweendomestic and agricultural wells

District Planning Frameworks · Regional planning and regulation Less water availableused to ensure that surface water upstreamresources of any given villageor town are not captured byupstream users

Table 19. Protection of domestic water supplies

90

8.3 Recommendation 2 –Targeting of options

In general, watershed development-type programmes do not try to targetactivities to appropriate physical, socialand institutional settings nor do theyattempt to regulate the number or scaleof interventions (e.g. the number ofcheck dams along a particular drainageline). The current tendency is to use a“one-size fits all” approach and to workunder the assumption that waterharvesting technologies are totally benign.The underlying logic is that evenintensive use of water harvestingtechnologies will not result in unintendednegative impacts. In contrast, resultspresented in this report show that themajority of water-related interventionspromoted by watershed development-type programmes have potential negativetrade-offs associated with them. Theresults also show that the scale of thesenegative trade-offs is, in many cases,directly related to the density of theinterventions and the scale of theadditional storage capacity that is created.

Better targeting of interventions canbe achieved by using decision trees thatutilise a combination of physical, social,economic and institutional data in thedecision-making process. This approachcan be used by NGO staff as part of aparticipatory planning process or it canbe used by line department staff at thedistrict level as part of a process thatinvolves stakeholder representation.Although potentially time consuming,using decision trees can become simpleand rapid, once data are readily available(e.g. once a water audit has beencompleted and a GIS database created).Also the approach can easily beincorporated into a more general MISsystem. Figure 51 is an example of adecision tree that uses numerical,physical, social and institutional datacollected by the Water Audit to identifyvillages in Kalyandurg in which “peri-urban” irrigation with waste waters ispotentially a good option.

91

Figure 51. Example of decision tree that uses physical, social and institutional data

Use of waste water for community basedperi-urban horticulture, fodder production etc.

Earmark funds and proceed with detailedvillage-level participatory planning

Yes

Yes

Yes

Yes

No

No

No

No

Sources of waste water in sufficient volume andof an acceptable quality to permit waste water irrigation ?

Land available (either public or private) to which wastewaters can be safely channeled that is suitable for horticulture or

fodder production ?

High QPA score for Panchayat responsiveness towater issues OR for community action to solve water

related problems ?

High QPA score for participation by thepoor in community decision making ?

Waste waterirrigation is notan option in this

village

Capacity building(second batch

attention)

Gated check dam under construction near Battuvani Palli, Kalyandurg

92

ID code Village NamePanchayat

response to WSS Problems

Community participation in solving RWSS

problems

Participation by poor in community

decision making

Participation by women in

community decision making

Strong panchayats and/or community

action

Strong participation by

the poorBoth tests

1 Hulikal 25 75 100 100 0 1 02 Vitlampalli 15 70 75 75 0 0 03 Mallikarjunapall 60 70 90 90 0 0 04 Chapiri 80 0 80 80 0 0 05 Madhireddipalli 25 80 100 100 0 1 06 Chapirithanda 0 100 100 100 1 1 17 Mudigal 80 0 85 75 0 0 08 Borampalli 100 80 90 50 1 0 09 Golla 100 0 100 75 1 1 110 Seebai 100 100 100 35 1 1 111 Pathacheruvu 100 0 60 0 0 0 012 Manirevu 0 75 75 0 0 0 013 Obulapuram 80 50 100 0 0 1 014 Nusikattala 50 100 100 0 1 1 115 Kondapuram 0 0 75 0 0 0 016 Kondapurampalli 15 75 80 75 0 0 017 Nusikottalathand 80 0 100 75 0 1 018 Thimmasamudram 0 0 100 0 0 1 019 Mangalakunta 50 0 100 0 0 1 020 Kadadarkunta 50 0 0 0 0 0 021 Balavenkatapuram 20 0 0 0 0 0 022 Pinjirikottala 50 90 100 50 0 1 023 Muddinayanampall 50 0 100 0 0 1 024 Venkatampalli 90 25 100 0 0 1 025 PTRPalli 20 15 65 0 0 0 026 Mouthikapuram 95 90 100 50 0 1 027 Kaparalapalli 0 65 100 35 0 1 028 PTRDiguvaThanda 20 75 80 0 0 0 029 PTREguvaThanda 60 100 100 0 1 1 130 Varli 90 100 100 25 1 1 131 Kodipalli 0 80 100 0 0 1 032 Mallipalli 25 100 100 75 1 1 133 Thimmaganipalli 100 85 75 35 0 0 034 Battuvanipalli 25 100 75 0 1 0 035 Ontimidi 25 50 75 35 1 0 036 Kurakulathata 100 85 100 30 0 1 037 Devadulakonda 50 100 100 0 1 1 138 Dodagatta 100 25 25 25 0 0 039 GubanaPalli 90 30 65 10 0 0 040 Garudapuram 40 50 100 75 1 1 1

Figure 52. Example of decision-support Excel work sheet.

Figure 53. Example villages targeted using the decision tree presented in Figure 51

93

Box 19. Issues not normally addressedin village-level participatory planning

These include:. Inter-village equity and/or upstream-

downstream equity;

. Rural-urban equity and anticipatedincreases in demand for water inurban and peri-urban areas and byindustrial users;

. Inter-generational equity andconsequences arising fromdemographic change and increaseddemand for water per household;

. Protection of biodiversity and rarehabitats in a given micro-watershed;

. Protection of surface and groundwaterfrom pollution whether this bedomestic, agricultural or industrial;

. Flood protection. Poorly-rehabilitatedtanks can pose a threat tocommunities living downstream.

The method that is recommended is modelledon the approach to water resource planning andmanagement that is currently being pioneered inSouth Africa. This recommendation may seem tomany like a retrograde return to top-down masterplanning; however, it is clear from this study thatwater resource planning and management isneeded urgently at levels above the village if manylarge scale equity and sustainability challenges areto be addressed.

The example decision tree presented in Figure51 has four tests. The first two tests requirephysical data relating to water quality, volumesof waste water, availability of land and thetechnical feasibility of channelling water to thisland. The third test requires data relating to thefunctionality of institutions in the village andwhether or not these institutions have a historyof being effective in tackling water-relatedchallenges. The fourth final test requiresinformation on whether poor social groups areinvolved in decision-making. The assumptionbehind this final test is that irrigation ofhorticulture or fodder using waste water is anactivity that could provide landless groups withaccess to irrigated land.

This type of decision tree is only feasiblebecause of the ordinal scoring system used duringthe Water Audit’s participatory assessment in allthe villages in Dhone and Kalyandurg. Or putanother way, such decision trees when developedas Excel-driven methodologies, willnot fit well with the qualitative informationthat is normally collected during participatoryassessments. Figure 52 is an example of anExcel work sheet which shows the villages fromKalyandurg that passed all the tests. In theory, asthis selection process has not involved village-levelparticipation, the next steps for the APRLPshould be to earmark funds and proceed withdetailed village-level participatory discussionsand, if there is strong local interestin this option, proceed with planning andimplementation. Figure 53 presents these“selected” Kalyandurg villages as a GIS layout.

Additional prototype decision trees for bettertargeting of interventions have been developedand it is strongly recommended that these bedeveloped further. It is also recommended thatthis approach be field tested by the APRLP and,if appropriate, combined into the MIS that theAPRLP is also developing and piloting.

8.4 Recommendation 3 –Village-level participatoryplanning within a wider DistrictPlanning Framework (DPF)

The aim of proposed District PlanningFrameworks is to ensure that water-relatedplanning and regulation issues not covered invillage-level planning (see Box 19) receiveattention when decisions are being made on theallocation of funds at the district and state levels.

Figure 54 is the proposed process by which aDPF would be developed. It is recommendedthat, at least initially, outputs from the DPFwould be consistent with norms, estimates ofstage of development and limits of acceptablechange that, in theory, are currently used fordecision-making at the district level. Once a limitof acceptable change has been set for, say, surfacewater resource development in an area, this figurecan be used in a computerised system to map outareas in which certain activities (e.g. additionalconstruction of check dams) would bepermissible as part of government-fundedprogrammes. It is envisaged also that DPFs couldbe used pro-actively in identifying and managingreserves of ground and surface water resourcesthat would meet basic human needs during

94

95

Figure 54. Proposed process for developing a waterrelated District Planning Framework

periods of drought. Finally, it is recommendedthat DPFs be used to identify and setmanagement guidelines for protecting rare andimportant aquatic eco-systems.

8.5 Recommendation 4 –Shift from supply to demandmanagement of water resources

Demand management seeks to maximise theservices provided by a given volume of watermainly by curbing non-essential or low value usesthrough price or non-price measures. Althoughdemand management actions are clearly not to bepreferred to supply-side actions in every case, it isapparent that they need to be given moreattention than is currently the case. The results

of the Water Audit show clearly thatthe emphasis of the APRLP and similarprogrammes should be on resource managementas opposed to resource development oraugmentation. Although demand management ofwater resources has a vitally important role toplay in the future in semi-arid areas of AndhraPradesh, it is not going to be a panacea, nor willit provide an instant solution to current andfuture challenges. If demand management is to bepolitically acceptable and receive public support,its introduction needs to be handled sensitivelyand be associated with appropriate publicawareness raising campaigns. Additionallydemand management has many potentialunintended consequences some of which couldimpact severely on poorer social groupings.

OBJECTIVE SETTINGSpecific planning and

management objectivesIDENTIFYING UNITSDelineating appropriate

physical and administrativemanagement units

M&EIdentify indicators andassociated locations,

determine limits ofacceptable change andmanagement responses

ASSESSMENTof current status of

water courses, aquifersand tanks; uses of water,

nature or problems

IMPLEMENTATIONof measures to protect drinking

water and to allocate/controlwater use for other purposes.

Management of surfacewater, groundwater,

infrastructure etc

DECISION MAKINGby authorities at

appropriate levels, withappropriate stakeholder

representation andconsultation

VISIONINGof options and

scenarios to addressproblems

D I S T R I C TP L A N N I N G

F R A M E W O R K

It is fundamentally important that policies anda legislative environment are created that provideincentives for more productive use of water byindividual users and disincentives for practicesthat are wasteful or lead to environmentaldegradation. Recent and current state-levelpolicies (e.g. grants for well construction,subsidised electricity for pumping irrigationwater, support prices for paddy) have theunintended consequence of encouraginginefficient and inequitable use of water.Watershed development propaganda, in the formof wall paintings, street plays and exhortationsfrom NGO staff and many newspaper articleshave also created the mindset that waterharvesting as part of watershed development canresult in almost limitless augmentation of waterresources in semi-arid areas. Changing thismindset will be not be an easy task.

Demand managementobjectives

. Mixture of short and longterm priorities

. Equitable developmentand sustainability

. Meets targets

. equity issueseg. upstream/downstream

. improved wateravailabilty andenvironmentalsanitation

. improved localeconomy

. collective benefit

. reduce conflict

. individual benefit,use of water when needed

. multiple goods andservices

Demand managementactions

. Legislation, institutional andpolicy development (includingmacro economic instruments)

. Programme development/targets

. M & E

. Development/updating districtplanning frameworks

. Within programme targeting ofinvestments

. Authorisation eg. borewells

. M & E

. Public awareness andexhortation

. M & E including communitypolicing

. Better management of CPRs

. Tradable water rights and watermarkets

. Watershed/affinity groupsmanaging tanks/groundwater

. Decision making including levelsof abstraction/cropping patterns

. M & E

. Social pressure

. Agreeing to and following rules

. More efficient and productivewater use

. Re-use of waste waters

Figure 55. Demand managementoptions and objectives

S TA T E

D I S T R I C T / M A N D A L

V I L L A G E

G R O U P

I N D I V I D U A L

96

8.7 Recommendation 6 –Reassessment of “official”water-related statistics

The survey work carried out primarily bygovernment departments to groundtruthsecondary data has shown large discrepanciesbetween “official” water-related statistics and thereality observed in the study mandals. The“official” statistics that showed the largest andmost important discrepancies included: wellnumbers, land areas under irrigation and numbersof domestic water points not meeting RajivGandhi National Drinking Water Mission norms.Since well number and irrigated area statistics areused for estimating stage of groundwaterdevelopment, this is currently grosslyunderestimated. The discrepancies relating todomestic water supplies, if they are representativeof the region, put in question the state levelstatistics. More recent information gathered onfluoride levels by the WHiRL project casts somedoubt over official statistics relating to theincidence of fluoride in domestic water supplies.Many of the statistics listed in this report are usedto underpin policies and inform decisions thatinvolve disbursement of huge amounts ofexpenditure, for example, decisions relating to:watershed development-type work, groundwaterdevelopment, NABARD loans, power sectorreform, protection of rural water supplies, riverbasin management, inter-basin transfers, tankrehabilitation and irrigation development.Findings from the Audit have also shown thatsome of the beliefs that also underpin the manyprogrammes (e.g. annual runoff is 30-40% ofannual rainfall) need to be re-evaluated. Hence itis strongly recommended that further checks onrelevant official statistics be carried out.

In discussions with line department and NGOstaff, the accessibility of information wasfrequently cited as being a problem. Although, inmany cases, relevant staff know that the water-related information they need is being collected orheld by a particular department, it is not alwaysreadily accessible in a format that makes itimmediately useful. It is therefore recommendedthat consideration be given to creating systemsthat improve both the quality and accessibility ofwater-related information. Given the importanceof water resource management in AndhraPradesh, it is also recommended thatconsideration be given to setting up a unit thathas specific responsibility for the management ofwater-related information.

To date, much of the debate in AndhraPradesh has centred on only two demandmanagement options, namely, introduction ofirrigated-dryland crops (i.e. crops with anassumed higher water use productivity thanpaddy rice and sugar cane) and increasedelectricity tarifs for pumping groundwater forirrigation. However, as Figure 55 shows there aremany other demand management options thatcould be considered. Figure 55 also illustratesthat demand management actions and objectivesvary with scale.

8.6 Recommendation 5 –Use of GIS-linked participatoryassessments as part of the M&Eof rural water supplyprogrammes

Participatory assessments carried out as part ofthe Water Audit revealed a major disparitybetween official statistics and the users’ view ofthe status of domestic water supplies. One reasonfor this disparity is the strong emphasis of currentrural water supply M&E on the functioning ofinfrastructure and the meeting of supply norms(i.e. 40 lpcd). Many of the other potentialproblems faced by users are ignored by existingM&E systems. However, these additionalproblems can be considered and appropriate stepscould be taken if routine M&E and responsesystems take users’ views into account.

With regard to routine M&E, theparticipatory assessment methodology usedduring the Water Audit does not have many ofthe constraints of participatory rural appraisal-type methodologies. In particular, it is rapid(a small team takes one day per village) and itproduces numerical data that can be stored andanalysed with relative ease and presented intables, GIS layouts or other formats that are easyto assimilate and act upon by decision makers.

97

Discussions before, during and after the final workshopof the APRLP Water Audit in September 2002 led to theidentification of a number of follow-up activities.These included (not in order of importance):

. “Demonstration” watersheds and villages. In the lightof the findings and recommendations of the WaterAudit, it is suggested that clusters of micro-watershedsor villages in Dhone and Kalyandurg be developedinto “model” watersheds that can be used primarily fordemonstration and training purposes. By July 2003,Water Audit recommendations were being piloted andrefined in four “pilot” villages in Kalyandurg as part ofthe WHiRL Project.

. Critical review of AP’s water harvesting policies andpractices. A key conclusion of the Water Audit is thatintensive water harvesting along drainage lines iscausing water shortages in many downstream areasparticularly during years with low rainfall and runoff.Hence, it is suggested that concerned governmentofficers undertake a critical review of current policiesand programmes.

. Water auditing manual and interactive CD. Asshortcomings have been identified in important water-related statistics, it is strongly recommended thatfurther water auditing takes place. It is suggested that awater auditing manual and an interactive CD would beof value in any scaling up of the work already taken.

. Awareness raising. Current watershed developmentpublicity is often highly misleading in that it suggeststhat there are quick fixes to water-related problems insemi-arid areas (e.g. check dam construction, contourbunding and tree planting). It is suggested that morerealistic awareness raising material be developed.

. IWRM capacity building. Capacity building is required98at all levels if Andhra Pradesh is to move rapidlytowards adopting the principles and practices ofintegrated water resource management and water-related project-cycle management. It is suggestedtherefore that an appropriate capacity buildingprogramme be developed using materials prouced bythe GWP, the EU (e.g. EC, 1998) and others as astarting point.

Proposed follow upactivities9

98

Mention should also be made of the vital contribution ofthe numerous support staff that contributed in allphases of the work. These include staff from CSWCRTI(Bellary) who assisted with the processing of data.Dr VAK Sarma (PS retd) from NBSSLUP (Bangalore)provided valuable comments and suggestions on thefinal draft of this report.

Particular thanks are also due to Mrs Lena Padukone,Mrs Rajyashree Dutt, Mrs Indira Bharadwaj and thestaff of Write Arm for their excellent work on thelayout and publication of this report.

Last but not least the Water Audit would not have beenpossible without the active participation and assistanceof many people living in the rural areas of Dhone andKalyandurg.

Acknowledgements

Funding for the APRLP Water Audit came primarilyfrom the UK’s Department for InternationalDevelopment; however, the editors are responsible forthe interpretations and conclusions presented here.The authors would also like to acknowledge the support,encouragement and assistance from the large numberof people that were involved. The editors gratefullyacknowledge the valuable advice and encouragementof Mr S Ray (Special Chief Secretary, Panchayati Raj andRural Development), Mr SP Tucker, (Director, APRLPPSU), Mr Anil Punetha (Commissioner, RuralDevelopment), Mr Somesh Kumar (Collector, Anantapur),Mr Sai Prasad (Collector, Kurnool) and Mr Ashok Jain(Conservator of Forests, Kurnool).

Many DFID staff made significant contributionsto the design and conduct of the Water Audit.These included: Mr Simon Croxton, Dr Peter Reid,Ms Sally Taylor, Ms Moutushi Sengupta, Mr Gopi Menon,Ms Aditi Rajyalaxmi and Mr Ajay Kumar. The editorswould also like to thank Mr Joe Hill for providingresults and data from his MSc dissertation.

The following made significant contributionsto the work in Kalyandurg. Kalyandurg Well Survey:Mr A Ramesh Reddy (Assistant Hydrologist),Mr K Thippeswamy (Assistant Hydrologist),Mr Jayaram Reddy (Assistant Hydrologist).Kalyandurg tank survey and secondary datacollection: Mr M Obula Reddy (MDT-Kalyandurg),Mr Rajendra Kumar (Computer Operator).Hired diploma holders: Mr Kishore, Mr Suresh,Mr Venkatachalam, Mr Arunachalam,Mr Venkataramana, Mr Sreenivasulu and Mr Valli.Kalyandurg QPA supervisors: Mr P Naganna andMr G Bheemappa. QPA team members: B Thomas Reddy,G Prabhakar, Giridhar, Vasantha, M Srinivasulu,B Rajanna, K Jayasimha, GH Rayudu, Prabhakar,Santhappa, V Bhaskar Reddy, Thimma Rajulu, Ishak,L Sivasankar, P Ramachandraiah, B Ramanjineyulu,M Ramanjaneyulu, A Sreerami Reddy, B Adimurthy,M Sivasankar, Baba Buden, S Suguna, G Bhaskar Babu,A Subba Reddy, Vasundhara, Pothi Reddy,BC Yerriswamy, Sadasiva, Thatheppa. S Manmohanand laboratory staff from CSWCRTI provided valuablesupport.

The following made significant contributionsto the work in Dhone. Spatial data team:Ramalinga Reddy, D Vijayakumar, S Ravikumar,B Ravindra Goud, SVN Rajamouli, Sekar Goud,Sreenivasa Reddy and Mastanvali. Non-spatial datateam: SK Nabisaheb, G Thirupathireddy,Ramakrishnudu, Sashikumar, Sudhakar Reddy,Prasada Reddy, Deepa Priyadarshini, RajyaLaxmi,Suvarna, Chandrani and Venkateswarlu. GroundwaterSurvey: C Raja, P Purushotham Reddy, S Mastanvali,A Balaramudu, J Shankaraiah, D Vijayavardhan Rao,Dr KL Narasimha Murthy, G Sreeramulu, M Venkat Raoand B Bramhananda Reddy.

99

IWMI. 2002. Water accounting for integrated water resourcesmanagement. www.cgiar.org/iwmi/tools/PDF/accounting.pdf

Moench, M., Caspari, E., Dixit, A. (Eds). 1999. Rethinking theMosaic: Investigations into local water management.Nepal Water Conservation Foundation, Kathmandu.

Molden, D. 1997. Accounting for water use and productivity.SWIM Paper No.1. IWMI, Colombo, Sri Lanka.

Molden, D., Sakthivadivel, R. and Habib, Z. 2001. Basin-leveluse and productivity of water: Examples from South Asia. IWMI Research Report 49, IWMI, Colombo, Sri Lanka.

Rama Mohan Rao, M.S., Adhikari, R.N., Chittaranjan, S.,Chandrappa, M. and Padmaiah, M. 1993. Influence ofconservation measures on water yield from catchments inthe semi-arid tract of South India. J. of Hydrol (NZ),Vol. 31, No 1, 23-38.

Rama Mohan Rao, M.S., Adhikari, R.N., Chittaranjan, S., andChandrappa, M. 1996. Influence of conservation measures ongroundwater regime in a semi-arid tract of South India. Ag.Water Manage., 30, 301-312.

RGNDWM. 2000. Guidelines for implementation of rural watersupply programme. Rajiv Gandhi National Drinking WaterMission, Ministry of Rural Development, India.

Samra, J.S., Sharda, V.N. and Sikka, A.K. 1996.Water Harvesting and Recycling: Indian experiences.CSWCRTI, Dehradun, India.

Seckler, D. 1996. The new era of water resources management:from “dry” to “wet” water savings. Research report 1, Colombo,Sri Lanka: International Irrigation Management Institute.

Singh, A.K., Rama Mohan Rao, M.S., Batchelor, C.H.and Mukherjee, K. 2003. Impact of watershed development ontraditional tank systems. Conference on “WatershedDevelopment and Sustainable Livelihoods: Past Lessons andFuture Strategies”, January 16-17 2003, Bangalore.

Singh, G., Venkataramanan, C., Sastry, G. and Joshi, B.P. 1990.Manual of soil and water conservation practices. Oxford/IBHPublishers, New Delhi.

References

Anon. 2001. A Fine Balance: Managing Karnataka’s scarcewater resources. Karnataka Watershed Development Project:Water Resources Audit. KAWAD Report 17, KAWAD Society,No. 250 1st Main, Indiranagar, Bangalore 560 038.

APGWD. 1999. Report on groundwater developmentpossibilities in Kalyandurg Mandal, Anantapur District.Andhra Pradesh Groundwater Department.

APGWD. 2000. Status report on groundwater resources inDhone Mandal. AP Groundwater Department.

Anon. 2003. Water Vision: Volume Position Papers and DistrictReports. Mission Support Unit, Water Conservation Mission,Government of Andhra Pradesh.

APWELL. 2003a. Judicious management of groundwaterthrough participatory hydrological monitoring: a manual.APWELL Project, PR and RD Departments, Government ofAndhra Pradesh.

APWELL. 2003b. APWELL Project Final Report. APWELL Project,PR and RD Departments, Government of Andhra Pradesh.

Batchelor, C.H., Rama Mohan Rao, M.S. and James, A.J. 2000.Karnataka Watershed Development Project: Water ResourcesAudit. KAWAD Report 17, KAWAD Society, No. 250 1st Main,Indiranagar, Bangalore, India.

Batchelor, C.H., Rama Mohan Rao, M.S., Mukherjee, K. andJames, A.J. 2001. Implementing watershed development andlivelihood projects in semi-arid areas: experiences from theKAWAD Project. Paper presented at the DFID “Livelihoods andWater” Workshop, 31 October 2001, London.

Batchelor, C.H., Rama Mohan Rao, M.S., and Manohar Rao, S.2003. Watershed development: A solution to water shortagesin semi-arid India or part of the problem? Land Use & WaterResources Res., 3, 1-10.

EC. 1998. Towards sustainable water resources management:a strategic approach. European Commission, Brussels.

James, A.J. (2000). MPA: A Methodology for ParticipatoryAssessments’, Waterlines, October Issue.

James, A.J. 2002a. Quantified participatory assessment for thewater resources audit of the Andhra Pradesh Rural LivelihoodsProject: Kalyandurg Mandal, Anantapur District. DFID, Delhi.

James, A.J. 2002b. Economic analysis of water use options:Dhone Mandal, Kurnool District and Kalyandurg Mandal,Anantapur District, Andhra Pradesh, India. DFID, Delhi.

James, A.J. and Snehalatha, M. 2002. Quantified participatoryassessment for the water resources audit of the AndhraPradesh Rural Livelihoods Project: Dhone Mandal, KurnoolDistrict. DFID, Delhi.

Janakarajan, S. 2000. Issues relating to official database ongroundwater resource in Tamil Nadu. Rev. of DevelopmentChange, V, 2, 385-407.

GWP. 2000. Integrated Water Resources Management.Global Water Partnership TAC Background Paper No. 4.(http://www.gwpforum.org/gwp/library/IWRMataglance.pdf)

Hill, J. 2001. Watershed Development: Spatial and temporalrainfall analysis and comparative Micro-watershed assessmentin Anantapur, Andhra Pradesh. MSc Dissertation, School ofGeography, University of Leeds, England.

100

Glossary

Adangal: Village record.

Adaptive water resource management: The principle of adaptiveresource management is to start with identificationof specific problems or needs and to select solutions that aredirectly relevant to addressing these problems or needs. Adaptivemanagement is also a process by which management actions anddirections are continually adjusted in the light of newinformation on current and likely future conditions.

Antispoofing: The Global Positioning System can be used fora range of unintended military purposes. Hence, antispoofing isthe practice of deliberately degrading the utility of the systemduring times of political unrest.

Aquifer: A geological formation that has sufficient water-transmitting capacity to yield a useful water supply in wells andsprings. All aquifers have two fundamental characteristics: acapacity for groundwater storage and also for groundwater flow.

Available water: The amount of water available to a service oruse, which is equal to the inflow less the committed water.

Bajra: Pearl millet.

Base flow: That portion of flow in streams that originatesfrom springs or groundwater seepage.

Betta: Hill.

Bucket-type water balance model: A mathematical model thatdescribes a hydrological or hydraulic process as a network ofinterconnected buckets (or storages). The model works on theprinciple that only when a bucket (or storage) is full can watermove along a potential gradient to the next bucket in thenetwork.

Bund: A ridge of earth used to control runoff and soil erosion.Sometimes used also to demarcate a field or plot boundary.

Catch crop: A fast maturing crop that is often planted as anadditional crop to make use of late rains or residual soil moisture.

Catchment: See “watershed”.

Check dam: A structure placed across a water-course primarily toreduce or check the velocity of water flow. Check dams can beconstructed using a wide range of designs and materials.Increasingly, check dams have been designed with the addedpurpose of impounding water so as to increase localisedgroundwater recharge.

Chickoo: Sapota – a horticultural crop.

Closed basin: A basin where utilisable outflows are fullycommitted.

Collector well: A collector well is a shallow hand-dug well of largediameter with horizontal boreholes drilled radially from the baseto a distance of approximately 30 m, typically in four directions.This drilling technique has been used successfully by the DFID-supported Western Indian Rainfed Farming Project.

Command Area: Area of irrigated land downstream of a tank thatreceives (or used to receive) irrigation water from thetank via a system of canals.

Committed water: The part of outflow that is reserved for otheruses (note this really applies to surface water).

Compartmental bunding: Water harvesting technique usedon relatively flat land that involves bunds in the form ofcompartments.

Contour: An imaginary line joining points of equal elevation ona land surface.

Contour bunds: Earthen ridges or embankments that areconstructed along the contours.

Contour trenching: Water harvesting technique that involvesdigging trenches along the contours of sloping land.

Cost recovery: Fee structures that cover the cost of providing theservice or investment.

Crore: One crore = 10,000,000.

Dead furrows: These are the furrows that are created in rainfedarable areas between crop rows generally 30-45 days aftersowing. The aim is to conserve moisture and dispose of excesswater.

Decision tree: An aid to systematic decision making thatsimplifies a complex decision-making process into a series of“yes/no” questions.

Demand management: The use of price, quantitativerestrictions, and other devices to limit the demand for water.

Depleted fraction: The fraction of inflow or available water thatis depleted by process and non-process use.

Dharna: Non-violent “sit-in” demonstration.

Dibbe: Mound.

Domain: the area of interest bounded in time and space wherewater auditing or accounting is to be carried out.

Drainage lines (or drainage network): Network of stream andrivers that drain a watershed.

Drought proofing: A series of technical, social or institutionalactions that reduce the vulnerability of a habitation or region tothe shock of drought.

Ephemeral stream: Streams in which water flows for only partof the year.

Etti: Assistant to the Revenue Department Village AdministrativeOfficer who lives in the relevant village or associated hamlets.

Evaporation: Process in which water passes from the liquid stateto the vapour state.

Evapotranspiration: Total evaporation from a natural surface(i.e. sum of soil evaporation, evaporation of water transpired byvegetation and evaporation of rain or overhead-irrigation waterintercepted by foliage and other vegetative material).

Externality: The unintended real (generally non-monetary) sideeffect of one party’s actions on another party that is ignored indecisions made by the party causing the effects.

Fully committed basin: a water basin that has beendeveloped to the extent that all water has been allocated or,in other words, all outflows are committed.

GEC Method: Groundwater Estimation Committee’s methodologyfor estimating groundwater resources.Used throughout India as the standard method.

Geographical Information System(GIS): A computer system forstorage, analysis and retrieval of information, in which all thedata are spatially referenced by geographic coordinates.

Global Positioning System: A system of 24 satellites whichcircle the earth twice a day in a very precise orbit and transmitinformation to earth. If GPS handsets are able to receive signalsfrom three or more satellites, they are able to calculate thelatitude and longitude of their current location.

101

Non-process depletion: depletion of water by uses other thanthe process for which the diversion was intended.

Open basin: a basin where uncommitted utilisable outflowsexist.

Panchayat: elected council.

Panchayati Raj: Local government.

Parameter: A property that is measured or observed.

Process depletion: the amount of water diverted and depletedto produce an intended good.

Rabi: The cropping season that follows the kharif.

Ragi: Finger millet.

Ratchakattas: platform in the village where meetings takeplace.

Runoff (or surface runoff): The portion of rainfall that flowsover the land surface. Runoff can concentrate in depressions orbehind impounding structures or it can continue to flow overthe land surface into water-courses.

Sarpanch: Elected head of the gram panchayat.

Seteria: Fox-tail millet.

Shramadan: The voluntary labour component of an organisedcommunity-based activity.

Stakeholders: In the context of this study, stakeholders areconsidered to be institutions and individuals that are concernedwith or have an interest in water resources and that would beaffected by decisions relating to water resource management.Stakeholders include people who may have little knowledge ofsuch affects and lack the means to participate.

Surface water drought: A period during which surface waterresources become severely depleted. In general, surface waterdroughts are caused by a combination of meteorological droughtand unsustainable use of surface-water and groundwater.

Sustainable rural livelihood: A livelihood is sustainable whenit can cope with and recover from stresses and shocks andmaintain its capabilities and assets both now and in the future,while not undermining the natural resource base.

Taluk: Sub-district.

Tank: major reservoir or large water body.

Tank spillage: Volume of water that passes over the wasteweir of a tank.

Thalarry: (same as etti)

Uncommitted outflow: Outflow from the domain that is inexcess of requirements for downstream uses.

Watershed: An area drained by a river system; also referredto as a catchment area.

Waste weir: The part of a tank bund over which surplus tankwater will flow once the tank has filled to its design capacity.

Zilla Parishad: District council.

Zingg terrace: Sometimes called a conservation bench terrace.A water harvesting practice that consists of a contributing areawith natural or slightly altered slope and a receiving area withno slope in any direction. Rainfall runs off the contributing areaand concentrates in the receiving area.

Gradonis: Bench terraces of small width formed on contoursby disturbing soil in areas having gentle to steep slopes.

Gram Panchayat: Elected village council.

Gram sabha: Meeting that involves the whole village.

Groundwater drought: A period during which aquifers becomeseverely depleted such that wells run dry and demand for water isnot met. In general, groundwater droughts are caused by acombination of meteorological drought and unsustainableextraction of groundwater for irrigation and other purposes.

Habitation: Village or hamlet having a population in excessof 250 people.

Halla: Gully lands.

Inbore well: Borewell drilled into the base of an openwell.

Indicator: A parameter or a value derived from parameters, whichpoints to, provides information about, describes thestate of a phenomenon/environment/area, with a significanceextending beyond that directly associated with a parameter value.

Information management: Process of gathering, storingand analysing information needed for a specific purpose,such as planning or making management decisions.

Integrated water resource management: Is a process whichpromotes the co-ordinated development and management of water,land and related resources, in order to maximise the resultanteconomic and social welfare in an equitable manner withoutcompromising the sustainability of vital ecosystems.

Inter-basin transfer: Practice of pumping or channellingwater from one river basin to another.

Intercropping: Growing two or more crops in the same fieldat the same time.

Jowar: Sorghum.

Kharif: June to October cropping season.

Lakh: One lakh = 100,000.

Limit of acceptable change: Indicators only have meaning in thelight of specific targets and threshold values. These targets orthreshold values can be used as triggers for management responses,as warning signals or as a means of evaluatingproject performance. Depending on the context LACs mayalso be referred to as safe minimum standards, critical loads,or critical thresholds. LACs may be determined locally,nationally or as part of international conventions.

Livelihood: A livelihood compromises the capabilities, assets(including both material and social resources) and activitiesrequired for a means of living.

Management: The decision-making process whereby a plan ora course of action is implemented. Planning forms part of thisprocess as does the allocation of resources and the resolutionof conflicts of interest. Effective management is only possible if managers have access to reliable information.

Mandal: A sub-district.

Meteorological drought: A period during which rainfall is lowand/or insignificant. Short periods of meteorological drought leadto depletion of soil moisture and damage to plants. In general, longperiods of meteorological drought lead to surface-water droughtand subsequently to groundwater drought.

Nala: A stream or dry water-course.

Nala bund: An earthen water-harvesting structure builtacross a nala.

102

ICRISAT International Centre for Research in theSemi-Arid tropics

IMD Indian Meteorological Department

IRS Indian Remote Sensing Satellite

IWMI International Water Management Institute

KAR Knowledge and Research

KAWAD Karnataka Watershed Development Project

lpcd litres per capita per day

M&E Monitoring and evaluation

MIS Management information system

MRO Mandal Records Officer

NABARD National Bank for Agriculture and Rural Development

NBSSLUP National Bureau of Soil survey and Land Use Planning

NC not covered

NGO Non-Government Organisation

NRIS Natural Resources Information System

NRSA National Remote Sensing Agency

NRSP Natural Resources Systems Programme

NSS No safe source (refers to villages in the RWSSprogramme)

O&M Operation & Maintenance

PC partially covered

PC Personal computer

PD Project Director

ppm parts per million

PR & RD Panchayati Raj and Rural Development

PRP Policy Research Programme

PSU Project Support Unit

RDT Rural Development Trust

RGNDWM Rajiv Gandhi National Drinking Water Mission

ROR Record of Right

R/S Remotely sensed

RWS Rural Water Supply

RWSS Rural Water Supply and Sanitation

SC Scheduled caste

SCS Soil conservation service

SHG Self-help group

SS safe source

ST Scheduled tribe

TDF Total dissolved solids

QC Quality control

QPA Quantitative participatory assessment

VAO Village Administrative Officer

VWSCS Village Water and Sanitation Committee

WH Water harvesting

WHiRL Water, Households and Rural Livelihood Project

WHO World Health Organisation

WL Water level

WRA Water resource audit

WSS water supply and sanitation

Abbreviations andAcronyms

AF Accion Fraterna (an NGO)

AIS & LUS All India Soil and Land Use Survey

AP Andhra Pradesh

APARD Andhra Pradesh Academy of RuralDevelopment

APGWD Andhra Pradesh GroundwaterDepartment

APRLP Andhra Pradesh Rural LivelihoodsProgramme

APSRAC Andhra Pradesh State Remote SensingApplications Centre

APWELL Netherlands-assisted Andhra PradeshGroundwater Borewell IrrigationSchemes Project

ARCVIEW Geographical Information Systemsoftware package (ESRI Inc.)

AWRA APRLP Water Resource Audit

COMMAN Project Community Management ofGroundwater Resources in Rural India

CPR Common pool resources

CRIDA Central Research Institute for DrylandAgriculture

CSWCRTI Central Soil and Water ConservationResearch and Training Institute

CWC Central Water Commission

DA Daily allowance

DCBC District Capacity Building Centre

DDP Desert Development Programme

DFID Department for InternationalDevelopment (UK Government)

DPAP Drought Prone Area Programme

DPF District Planning Framework

DRD Department of Rural Development

DWCRA Development of Women and Children inRural Areas Programme

EC European Commission

ETp Potential evaporation

EU European Union

F fluoride

FAO United Nations Food and AgricultureOrganisation

FC fully covered

GIS Geographical Information System

GoAP Government of Andhra Pradesh

GoI Government of India

GPS Global Positioning System

GWD Groundwater Department

GWP Global Water Partnership

Ha hectare

Ham hectare metre

HP Horse power

ICAR Indian Council for Agricultural Research

103

Andhra Pradesh Rural Livelihoods Programme(Government of Andhra Pradesh)

A. Madhava Reddy Academy of Rural Development (Formerly APARD)Rajendranagar, Hyderabad - 500 030, India

Telephone: 91-40-400 1953, 400 1954, 550 2896 Telefax: +91-40-401 8773E-mail: psu@aprlp.org; aprlp@in-biz.net

www.aplivelihoods.org