Renewable energy technologies for irrigation water pumping in India: A preliminary attempt towards potential estimation

ARTICLE IN PRESS

0360-5442/$ - se

doi:10.1016/j.en

�CorrespondE-mail addr

(A. Kumar).

Energy 32 (2007) 861–870

www.elsevier.com/locate/energy

Renewable energy technologies for irrigation water pumping in India:A preliminary attempt towards potential estimation

Atul Kumara,�, Tara C. Kandpalb

aPolicy Analysis Division, The Energy and Resources Institute, Darbari Seth Block, IHC Complex, Lodhi Road, New Delhi 110003, IndiabCentre for Energy Studies, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016, India

Received 27 September 2004

Abstract

Simple frameworks have been developed for estimating the utilization potential of: (a) solar photovoltaic (SPV) pumps; (b) windmill

pumps; (c) producer gas based dual fuel engine pumps; and (d) biogas based dual fuel engine pumps for irrigation water pumping in

India. The approach takes into account factors such as: solar radiation intensity, wind speed, availability of bovine dung and agri-

residues, and their alternative uses, ground water requirements for irrigation and its availability, affordability, and propensity of the

users to invest in renewable energy devices, etc. SPV pumps are estimated to have the maximum utilization potential in India, followed by

windmill pumps.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Potential estimation; Renewable energy technologies; Irrigation water pumping

1. Introduction

In view of the increasing global climate change concerns,interest in the development and dissemination of re-newable energy technologies (RETs) have again beenrenewed. In India, a fully-fledged Ministry (Ministry ofNon-Conventional Energy Sources) is dedicated to theacceleration of RETs development. However, the reportedcumulative numbers of dissemination of most of the RETsare very low, particularly for agricultural applications.Table 1 presents the time trend of the reported growth inthe cumulative numbers of solar photovoltaic (SPV) pumpsand wind pumps installed in the country as against thecorresponding values for electric pumps [1]. Fig. 1 depictsthis data in graphical form. It may be noted that, thoughthe numbers of renewable energy based pumps arenegligibly small as compared to the number of electricpumps, the rate of growth of their dissemination in thecountry is much higher than the growth rate in the numberof electric pumps. Though the measures taken by the

e front matter r 2006 Elsevier Ltd. All rights reserved.

ergy.2006.05.004

ing author. Tel.: +9111 24682100; fax: +91 11 24682144.

esses: [email protected], [email protected]

government of India for promoting renewable energybased pumps have primarily contributed to the apparenthigh rate of growth in their dissemination, their actualcontribution is relatively very small. To find out niche areasfor immediate intervention, a realistic assessment of theutilization potential of renewable energy sources inagriculture sector is required. An attempt to estimate theutilization potential of following four RET for irrigationwater pumping in India has been made: (i) SPV pumps;(ii) windmill pumps; (iii) producer gas based dual fuelengine pumps; and (iv) biogas based dual fuel enginepumps.

2. Factors affecting the utilization potential

The potential of RET based pumps for irrigationdepends on several factors such as resource availability,groundwater requirement and its availability, affordabilityof farmers, and farmers’ propensity to invest in a RETbased pump, etc. While the issues relating to the first factor(i.e, the resource availability) may vary considerably fromone technology option to another, the issues concerning theremaining three factors are expected to be more or less

ARTICLE IN PRESSA. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870862

common. A brief discussion on the common factors ispresented in the following sub sections.

2.1. Groundwater requirement and its availability

Since surface water irrigation is usually the cheapestoption for irrigation [2], farmers having access to sufficientsurface water for irrigation may not choose any otheroption(s). Therefore, the areas in the country with surfacewater availability (a fraction fs of the net sown area) [3–5]have not been included in the potential estimation. It isassumed that the surface water, wherever available, issufficient to meet the irrigation water requirement.Furthermore, only those areas with groundwater require-ment (as a fraction of the total areas requiring groundwater) have been considered where the ground water table[6] is within a pre-specific depth (which may varydepending upon the RET being used). For example, acentrifugal surface pump coupled with an SPV array is

Table 1

Time-trend of cumulative number of reported installations of SPV pumps,

wind pumps and electric pumps in India

Year SPV pumps Wind pumps Electric pumps

1993–94 54 56 10276000

1994–95 113 146 10721300

1995–96 463 244 11101100

1996–97 1816 449 11565300

1997–98 2481 511 11849400

1998–99 2787 637 12216700

1999–2000 3131 714 12514200

2000–01 3575 793 12823500

2001–02 4208 888 13043900

2002–03 5113 945 13792400

0

1000

2000

3000

4000

5000

6000

1992 1993 1994 1995 1996 1997 1

Yea

Cum

ulat

iven

umbe

r of

SP

V a

ndW

ind

pum

ps (

in n

umbe

r)

SPV pumps Wind pumps

Electric pumps

Fig. 1. Time-trend of cumulative number of reported installatio

usually considered for shallow well water pumping. Themaximum suction head for such a surface pump is reportedto be 7m [7]. SPV systems with a submersible pump arereportedly capable of deep well water pumping up to 70m[7]. However, such deep well irrigation pumping may notalways be economical [8]. As a consequence, SPV waterpumping for irrigation is preferably carried out in areaswith ground water table less than 10m. Similarly the wateroutput of commonly available windmill pumps is reportedto be rather low for a head of more than 20m. A surfacemounted dual fuel engine pump coupled with a biomassgasifier or biogas plant is usually recommended for areaswith ground water table up to 5m [9].Central ground water board (CGWB) monitors the

ground water table in India from a network of about 15 000stations. Measurements of ground water tables are taken atthese stations four times in a year. It is observed thatduring pre monsoon period (May month) with maximumirrigation water demand the ground water table is deepest[6]. Therefore, the ground water table for the month ofMay (pre-monsoon period) has been considered in thisstudy [6]. Table 2 presents the state wise net sown area, itsfraction of area under surface irrigation (fs) and the areawith different ground water table in India (as a fraction oftotal area requiring ground water) [3,4,6].

2.2. Affordability of the user

To make a preliminary assessment of the affordabilityof a farmer to invest in a RET based water pumping systema comparison of the total annual cost likely to be incurredby the farmer with his annual income from agriculturecan be made. The following assumptions are made toestimate the total annual cost of RET based waterpumping system.

998 1999 2000 2001 2002 2003

r

0.0

2.5

5.0

7.5

10.0

12.5

15.0

Cum

ulat

ive

num

ber

of e

lect

ric p

umps

(in m

illio

ns)

ns of SPV pumps, wind pumps and electric pumps in India.

ARTICLE IN PRESS

Table 2

State wise net sown area (As), its fraction irrigated by surface water (fs)

and area with different ground water table in India

State As

(thousand

hectare)

fs (fraction) Area with different

ground water table (as a

fraction of the total area

requiring ground water)

Up to 5m Up to 10m

Andhra Pradesh 14460 0.1710 0.3153 0.8054

Arunachal

Pradesh

350 —a 0.4444 1.0000

Assam 3205 0.1129 0.7424 0.9694

Biharb 10743 0.1107 0.4850 0.9621

Goa 67 0.1045 0.7456 1.0000

Gujarat 10292 0.0624 0.1958 0.5602

Haryana 3711 0.3697 0.3763 0.7025

Himachal

Pradesh

1010 0.0040 0.2817 0.6056

Karnataka 12321 0.0941 0.2153 0.6615

Kerala 1796 0.0863 0.3823 0.8242

Madhya

Pradeshc22111 0.0943 0.1367 0.6709

Maharashtra 20925 0.0433 0.2116 0.7376

Orissa 5296 0.2368 0.4901 0.9517

Punjab 4033 0.3362 0.3439 0.7549

Rajasthan 20971 0.0830 0.1577 0.3963

Tamil Nadu 7474 0.2015 0.4678 0.8377

Uttar Pradeshd 17986 0.1764 0.4387 0.8457

West Bengal 5656 0.1733 0.3473 0.8069

aNegligible area.bIncluding Jharkhand.cIncluding Chattisgarh.dIncluding Uttaranchal.

A. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870 863

A fraction fcs of the capital cost (C0) of the system maybe provided as capital subsidy to the user by thegovernment.

A fraction fis of the balance of the cost (after capitalsubsidy) may be provided as loan to the farmer for a periodof t years at an interest rate r (in fraction).

The remaining fraction of the capital cost is assumed tobe paid by the user out of his savings. This fraction is notconsidered for the estimation of annual expenses of thesystem.

The annual operation, repair and maintenance cost ofthe system can be expressed as a fraction m of its capitalcost.

Mathematically, the annual expenses (Ca) can beexpressed as

Ca ¼ C0 ð1� f csÞf is

rð1þ rÞt

ð1þ rÞt � 1

� �þm

� �, (1)

A study carried out by World Bank for the state1 ofHaryana has estimated the gross agricultural income andfarmers’ willingness to pay for irrigation [2]. Since data forall states of India is not available, the per hectare value for

1India is divided into following main administrative units arranged

hierarchically: state, division, district, block, and village.

Haryana is assumed to be applicable for other states ofIndia as well in the example case(s) presented in the paper.The framework presented in this study can be used withbetter realistic value as and when available. On the basis offarm size, the farmers in India are classified in fivecategories (marginal, small, semi-medium, medium, andlarge) [4]. The willingness to pay for irrigation of farmers ofrespective categories have been estimated on the bases ofan all India average land holding size of different categoriesof farmers [4].

2.3. Propensity of the user to invest in a renewable energy

technology based pump

The propensity of the user to invest in a RET basedpump will depend on the awareness and knowledge of theuser about the technology for water pumping and also onthe availability, reliability, and economics of conventionaloptions to him. A factor (varying from 0 to 1) namedpropensity factor can be used to express the same. In thisstudy the value of propensity factor for marginal, small,and semi-medium farmers has been assumed at 0.2 while itis assumed to be 0.8 for medium and large farmers.As mentioned earlier, a large variation is expected in the

case of resource availability for the technology options forwater pumping considered in this study. The followingsection presents the resource availability related issues forall the four RET options for irrigation water pumping.

3. Resource availability for renewable energy technology

options

3.1. SPV pumps

It is recommended that for installing SPV pumps, theaverage daily solar radiation in the least sunny monthshould be greater than 3.5 kW/m2 on a horizontal surface[10]. Ideally, detailed solar radiation data for each locationshould be used in evaluating the potential of using an SPVpump. However, to make an initial macrolevel assessment,broad solar radiation availability characteristics readilyavailable in the literature [11] are used. On a macrolevel,seven north-eastern sates and the northern states of Jammuand Kashmir, Himachal Pradesh and Uttranchal in Indiacan be given low priority in the process of identification ofniche areas for installation of SPV pumps.

3.2. Windmill pumps

The water output of a windmill pump is very sensitive toany variation in wind speed. Wind speed at any locationdepends upon a variety of site specific factors and largevariations in annual monthly mean wind speeds have beenreported at different locations in a state [12–15]. Therefore,analysis for the estimation of potential of windmill pumpsfor irrigation water pumping in India has been carried outusing district wise data on resource availability, net sown

ARTICLE IN PRESSA. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870864

area, surface water irrigation, and ground water table, etc.[5,6,12–15].

The wind speed at any location increases with anincrease in height from the ground level. In most of theprevailing windmill pump designs rotor is fixed on a towerat about 10 meters height from the ground. From theavailable data on measured values of annual monthly meanwind speed at different locations [12–15], the annualmonthly mean wind speeds at the 10m height from theground level (v10) have been estimated by using thefollowing relation [15,16]:

v10 ¼ vh

10

h

� �a

, (2)

where vh represents the annual monthly mean wind speedmeasured at a reference height of h meters from the groundlevel and a the ground surface friction coefficient. Thoughthe value of a varies from 0.1 to 0.5 depending upon theterrain of the location, a value of 0.2 as reported for cropsand shrubs type terrain has been used in the study [16].

As the reported water output of commercially availablewindmill designs at wind speeds below 10 km/h is very low[17], the areas with annual monthly mean wind speedsgreater than 10 km/h (as a fraction, fw, of the net sown areain the district) only have been considered in this study forwindmill pump based irrigation.

3.3. Producer gas based dual fuel engine pumps

In view of decreasing availability of fuel wood fromforests in India [18], it is assumed that the same is not

Table 3

State wise sown area and yield of selected crops in India (for the year 2000–0

State Arhar Cotton Jute

Sown area

(thousand

hectares)

Yield

(kg/ha)

Sown area

(thousand

hectares)

Yield

(kg/ha)

Sow

(thou

hecta

Andhra Pradesh 510 426 1020 277 80

Assam —a — 70

Biharb 60 1280 — 170

Gujarat 320 337 1620 122 —

Haryana — 560 424 —

Himachal Pradesh — — —

Jammu and Kashmir — — —

Karnataka 580 441 560 298 —

Madhya Pradeshc 350 665 510 80 —

Maharashtra 1100 602 3080 100 30

Orissa 150 503 — 30

Punjab — 470 430 —

Rajasthan — 510 268 —

Tamil Nadu 80 710 190 285 —

Uttar Pradeshd 400 1254 — —

West Bengal — — 620

aVery small.bIncluding Jharkhand.cIncluding Chattisgarh.dIncluding Uttaranchal.

available as feedstock for large-scale dissemination ofproducer gas based dual fuel engine pumps. Agro-proces-sing residues such as rice husk, groundnut shells andbagasse, etc. also may not be available for the use asfeedstocks for agricultural water pumping systems, as thesemay be used by the respective agro-processing units formeeting their own thermal/electrical energy requirement.Some crop residues may find use as fodder and/or fertilizer.Hence, the following non-fodder, non-fertilizer cropresidues have been considered as potential feedstocks: (a)arhar stalks; (b) cotton stalks; (c) jute sticks; (d) maize cobsand stalks; and (e) mustard stalks. These residues areavailable in substantial amounts in the country and havealso been found suitable as feedstocks in biomass gasifiers[4,19].The annual amount of crop residue produced depends

on the annual crop production and the correspondingresidue to crop ratio (Rj) [20]. A fraction (fl) of the totalamount of crop residues produced may be lost duringcollection. A fraction (fou) of the collected crop residuesmay be used in other applications [21].The amount (Qr,j) of jth crop residue available per

hectare of sown area can be estimated as

Qr;j ¼ Y jRjð1� f lÞð1� f ouÞ, (3)

where Yj represents the yield of jth crop in the state and Rj

the residue to crop ratio for jth crop.It is assumed that producer gas based dual fuel engine

pumps for irrigation can only be used in those states wherethe crop(s) considered in this study are produced in

1)

Maize Mustard

n area

sand

res)

Yield

(kg/ha)

Sown area

(thousand

hectares)

Yield

(kg/ha)

Sown area

(thousand

hectares)

Yield (kg/

ha)

1586 520 2727 —

1674 — 270 514

1410 730 2159 110 933

400 1550 190 1236

— 400 1371

300 2293 —

330 1592 —

670 3157 —

820 1468 450 804

266 260 850 —

908 — —

160 2793 50 1222

970 1047 1430 918

120 1619 —

910 1632 910 987

2167 40 2501 440 956

ARTICLE IN PRESSA. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870 865

substantial amounts. In fact, in this study, only those stateand crop combinations are considered, where the cropyield is enough to provide the amount of residue that isrequired as feedstock for irrigating at least one crop in ayear. The sown areas of different crops considered in thisstudy along with their respective yields for various statesare presented in Table 3 [4]. To avoid ambiguity it isassumed that a farmer with surplus availability of the cropresidue is not trading the surplus amount with any otherfarmer.

3.4. Biogas based dual fuel engine pumps

It has been reported that households having less thanfour bovines cannot normally install family size biogasplants [22]. Therefore, in this study it is assumed thathouseholds owning 1–3 bovines may contribute dung foroperation of community type biogas plants for supplyingbiogas to a dual fuel engine pump for irrigationwater pumping (besides other end uses). Based on the1991 census data on the livestock holding and itsdistribution in 16 states of India (Table 4), number ofhouseholds owning 1–3 bovines have been estimated [23].An average value of 10.88 kg of dung availability perbovine is assumed in this study [24]. It is also assumed thatonly 75% of the available dung is finally collected. It is alsoassumed that 25 kg of raw dung has to be fed daily tothe biogas plant per m3 of rated capacity of the biogasplant [24].

Table 4

Pattern of bovine ownership in different states of India (for the year 1991)

State Number of rural

households

(thousand)

Rural households owning n bovines

as a % of all rural households

n ¼ 1 n ¼ 2 n ¼ 3

Andhra Pradesh 10327 12 11 7

Assam 3364 12 11 7

Bihara 12175 17 17 8

Gujarat 4804 14 13 12

Haryana 1882 14 13 12

Himachal

Pradesh

861 8 8 11

Karnataka 5552 12 11 9

Kerala 4102 14 14 5

Madhya

Pradeshb8945 10 9 8

Maharashtra 9259 10 9 6

Orissa 5168 14 14 7

Punjab 2355 14 13 12

Rajasthan 5574 10 10 —c

Tamil Nadu 8435 12 11 5

Uttar Pradeshd 18024 20 19 13

West Bengal 8910 13 13 7

aIncluding Jharkhand.bIncluding Chattisgarh.cVery small (negligible).dIncluding Uttaranchal.

4. Framework used for potential estimation

4.1. SPV pumps

On the basis of assumptions made in earlier section, netsown area in a state with requirement for ground water andalso its availability (Ag) can be estimated as

Ag ¼ Asð1� f sÞf g10, (4)

where As represents the net sown area in the state, fs theareas in the state with surface water availability (as afraction of the net sown area in the state), and fg10 the areawith ground water table up to 10m (as a fraction of thetotal area requiring ground water in the state).As discussed earlier, the factors pertaining to afford-

ability of the users and their propensity to invest in an SPVpump are quite different for farmers of different categories.Therefore, a category wise break-up of the above estimatedarea has been used for further analysis. The net sown areawith requirement and availability of ground water for ithcategory of farmers (Ag,i) can be expressed as

Asg;i ¼ Asð1� f sÞf g10f lh;i, (5)

where flh,i represents the net sown area operated by ithcategory of farmers (on the basis of land holding size) as afraction of net sown area in the state (i ¼ 1; 2; 3; 4, and 5correspond to marginal, small, semi-medium, medium, andlarge, respectively) [4].Using average size of land holding of different categories

of farmers in the state [4], the number of farmers of ithcategory (Nsg,i) that own the above estimated area (Eq. (5))can be determined as

Nsg;i ¼Asð1� f sÞf g10f lh;i

Si

, (6)

where Si represents the average size of land holding of ithcategory of farmers in the state.Finally, using correction factors for the affordability of

the ith category of farmers (fa,i) and their propensity toinvest in SPV pumps (fp,i) the potential number of SPVpump users (Nspv) in the state can be estimated as

Nspv ¼X5i¼1

ðNg;if a;if p;iÞ. (7)

Two sizes of SPV pumps (0.9 and 1.8 kWp) arecommercially available in India. The financial incentivesprovided by government are available for purchasing onlyone pump. Therefore, in the initial phase of disseminationit is assumed that the farmer will install either of the twopumps depending upon his need and affordability. In thisstudy it is assumed that the farmers with large landholdings will continue to use other conventional pumps forsatisfying their balance water requirement for irrigation.

ARTICLE IN PRESSA. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870866

4.2. Windmill pumps

On the basis of assumptions made in Section 3.2, netsown area (Awg) with requirement for ground water andalso its availability in a district where annual monthlymean wind speed is greater than 10 km/h can be estimatedas

Awg ¼ As;df wð1� f s;d Þf g20, (8)

where As,d, fs,d and fg20, respectively, represent the net sownarea, fraction of sown area under surface irrigation and thefraction of total area requiring ground water with groundwater table up to 20m in the district [5,6,12–15].

As discussed earlier the factors pertaining to afford-ability of the users and their propensity to invest in awindmill pump is different for farmers of differentcategories. Therefore, the category wise numbers of farm-ers owning the area as estimated by Eq. (8) need to beobtained. Since district level data on sizes of land holdingsand land holdings distribution among different categoriesof farmers are not available, it is assumed that the valuespertaining to a state [4] are applicable for each district ofthat state. Thus, the number of farmers of ith category(Nwg,i) that own the above estimated area (Eq. (8)) can beestimated as

Nwg;i ¼As;d f wð1� f s;dÞf g20f lh;i

Si

. (9)

As in the case of SPV pumps, by introducing correctionfactors for affordability of the farmers and their propensityto invest in windmill pumps the potential number ofwindmill pump users (Nwp) in the district can be estimatedas

Nwp ¼X5i¼1

ðNwg;if a;if p;iÞ. (10)

4.3. Producer gas based dual fuel engine pumps

The required hours of operation (trq) of a producer gasbased dual fuel engine pump for irrigating one hectare landcan be estimated as

trq ¼rwgHV w

3:6� 106ðPpgZdpÞ, (11)

where rw represents the density of water, g the accelerationdue to gravity, H the total head, Vw the volume of waterrequired for irrigating one hectare land during a singlecropping season, Ppg the size of producer gas based dualfuel engine pump (in kW) and, Zdp the efficiency of thepump used with the dual fuel engine.

The rate of consumption (Xcr,j) of jth crop residue in aproducer gas based dual fuel engine pump of ratedcapacity, Ppg, would also depend upon its calorific value(CVcr;j), efficiency of gasifier (Zgas), efficiency of dual fuelengine (Zdfe), and diesel replacement factor (rd,pg) of the

producer gas based dual fuel engine pump. Xcr,j can beestimated as

X cr;j ¼3:6ðPpgrd ;pgÞ

CVcr;jZgasZdfe

, (12)

where j ¼ 1; 2; 3; 4; 5 correspond to arhar stalk, cottonstalk, jute sticks, maize cobs and stalk, and mustard stalkrespectively.Therefore, the annual hours of operation (tav;j) of a

producer gas based dual fuel engine pump using the cropresidue available from one hectare sown area of jth cropcan be estimated as

tav;j ¼Qr;j

X cr;j. (13)

For each of the crop considered in this study, it isassumed that the prevailing situation pertaining to thesurface water availability, land holding sizes, land holdingdistribution among farmers, and the ground water table isthe same over the entire state. Thus, the sown area of acrop with ground water requirement and its availability(Ag,j) can be estimated as

Ag;j ¼ As;jð1� f sÞf g5, (14)

where As,j represents the sown area of jth crop in the state,fg5 the area with 5m ground water table (as a fraction ofarea requiring ground water). This exercise has beenundertaken only for those state and crop combinationsfor which the value of tav,j is greater than the requiredhours of operation (trq) as estimated from Eq. (11).Similarly, as in pervious cases, using distribution of land

holdings among different categories of farmers and theiraverage size of land holdings, the number of ith category offarmers (Npg;j;i) growing the jth crop and owning the areaas estimated by Eq. (14) can be estimated as

Npg;j;i ¼As;jð1� f sÞf g5f lh;i

Si

. (15)

Once again, using correction factors for the affordabilityof farmers and their propensity to invest in a producer gasbased dual fuel engine pump the potential number ofproducer gas based dual fuel engine pump users (Npg) inthe state can be estimated as

Npg ¼X5i¼1

X5j¼1

ðNpg;j;iÞf a;if p;i. (16)

4.4. Biogas based dual fuel engine pumps

A 20m3 daily biogas production capacity communitybiogas plant has been chosen as it can provide sufficientvolume of biogas for operating a 7.46 kW (10 hp) dual fuelengine pump with reasonable capacity utilization factor. Inthe case of community biogas plants (being considered inthis study for providing biogas to the dual fuel enginepump) the affordability criteria can be somewhat relaxed.

ARTICLE IN PRESS

Table 5

State wise average sizes of land holdings of different categories of farmers

(for the year 1990–91)

State Marginal

farmers

(ha)

Small

farmers

(ha)

Semi-

medium

farmers

(ha)

Medium

farmers

(ha)

Large

farmers

(ha)

Andhra Pradesh 0.45 1.43 2.71 5.86 15.66

Arunachal

Pradesh

0.63 1.53 2.80 5.65 16.40

Assam 0.40 1.40 2.68 5.18 80.80

Bihara 0.35 1.36 2.73 5.65 16.41

Goa 0.33 1.54 2.25 4.50 19.00

Gujarat 0.53 1.47 2.83 5.99 16.45

Haryana 0.47 1.52 2.81 5.86 15.41

Himachal

Pradesh

0.40 1.42 2.74 5.69 16.17

Karnataka 0.47 1.46 2.75 5.93 15.28

Kerala 0.17 1.37 2.60 5.43 59.33

Madhya

Pradeshb0.45 1.45 2.78 6.04 16.44

Maharashtra 0.49 1.46 2.77 5.86 15.13

Orissa 0.49 1.38 2.63 5.40 16.80

Punjab 0.55 1.61 2.91 6.21 16.07

Rajasthan 0.48 1.44 2.85 6.23 19.11

Tamil Nadu 0.36 1.41 2.73 5.71 18.52

Uttar Pradeshc 0.38 1.41 2.73 5.55 15.42

West Bengal 0.44 1.53 2.78 5.39 203.00

All India

Average

0.39 1.43 2.76 5.90 17.33

aIncluding Jharkhand.bIncluding Chattisgarh.cIncluding Uttaranchal.

A. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870 867

Moreover, due to economy of scale, the unit cost ofcommunity biogas plant (per m3 biogas productioncapacity) would be much lower than that for family sizebiogas plants. Thus, it is possible for the participating usersto share equity funds for the installation of the biogas plantand other equipment for the system. However, an index ofcommunity participation (l) has been introduced to takeinto account the practical problems faced in such aparticipatory programme. In fact, a low value (0.10) hasbeen assigned to the index of participation in this study toobtain rather conservative estimates. It may be noted thatthe index of community participation would also take intoaccount the propensity of the individuals to invest in acommunity biogas plant based water pumping system.Further, the value of l can only be estimated throughdetailed field survey based studies. In the absence of suchstudies, in the present work, the value is assumed on thebasis of discussion with the experts in the relevant areas.

Considering all the above mentioned factors thepotential number of community biogas plants (Nbg) thatcan be used for irrigation water pumping in a state can beestimated by the following relation:

Nbg ¼1

20

lðNh1 þ 2Nh2 þ 3Nh3Þ 10:88� 0:75ð Þ

25

� �

�ð1� f sÞf g5, ð17Þ

where Nh1, Nh2 and Nh3 represent the number of house-holds owning 1, 2 and 3 bovines, respectively, in a state. Asin the cases of other three technological options theproduct ‘(1�fs)fg5’ is used to consider the area with groundwater requirement and its availability.

5. Results and discussion

The use of the framework presented in the earlier sectionrequires detailed data on a variety of input parameters.However, due to unavailability of required data an initialexercise towards estimation of potential numbers andidentification of niche areas has been undertaken on thebasis of the proposed framework. Table 5 presents the statewise average sizes of land holding of different categories offarmers in India [4]. The fractional distribution of net sownarea among different categories of farmers in various statesof India is presented in Table 6 [4]. Other input parametersused in calculations are presented in Table 7 [9]. Using thebase values of various input parameters, as presentedearlier some exemplifying calculations have been made andthe results obtained are briefly presented in this section.

On the basis of assumptions made in Sections 2, 3 and 4,and the values of input parameters as listed in Table 7, thecategories of farmers who can afford to invest in SPVpump, windmill pump and producer gas based dual fuelengine pump are indicated in Table 8. It may be noted thatonly for these combinations of category of farmers andRET based pumps, the affordability factor (fa,i) is assigneda value of one (1) while for others it is assumed to be zero.

Table 9 presents the estimated potential of four REToptions considered for irrigation water pumping in thisstudy. Potential number of SPV pumps are estimated at6.03 million (out of which 4.29 million are of capacityof 1.8 kWp and the balance 1.74 million of capacity of0.9 kWp). With 1.06 million SPV pumps the stateof Maharashtra in India has highest utilization potentialclosely followed by Madhya Pradesh (0.99 million). Sincethe percentage share of marginal and small farmers is veryhigh in the states of Kerala and West Bengal, in spite oflarge area with ground water table of less than 10m inthese states, the estimated potential of SPV pumps is verylow as compared to the total number of farmers. More-over, due to larger number of smaller land holdings inKerala and West Bengal, the potential number of 0.9 kWp

SPV pumps is much higher than that of 1.8 kWp SPVpumps. In other states it is just the reverse as the landholding are larger.The potential number of windmill pumps for irrigation is

estimated at 2.39 million. In view of long coastal area, largenumber of farmers with small or higher land holdings, andsubstantial area under 20m ground water table the state ofMaharashtra has highest utilization potential of windmillpumps (0.64 million), followed by another coastal state

ARTICLE IN PRESS

Table 6

State wise fractional distribution of net sown area among different

categories of farmers in India (for the year 1990–91)

State Marginal

farmers

Small

farmers

Semi-

medium

farmers

Medium

farmers

Large

farmers

Andhra Pradesh 0.1638 0.1955 0.2517 0.2612 0.1278

Arunachal

Pradesh

0.0286 0.0743 0.2400 0.4200 0.2343

Assam 0.1894 0.2446 0.2864 0.1535 0.1261

Bihara 0.3343 0.1819 0.2398 0.1845 0.0596

Goa 0.2836 0.1642 0.1343 0.1343 0.2836

Gujarat 0.0475 0.1305 0.2444 0.3891 0.1886

Haryana 0.0795 0.1248 0.2544 0.3503 0.1911

Himachal

Pradesh

0.2129 0.2327 0.2554 0.2030 0.0960

Karnataka 0.0870 0.1873 0.2597 0.3061 0.1600

Kerala 0.4816 0.2133 0.1420 0.0635 0.0991

Madhya

Pradeshb0.0637 0.1259 0.2188 0.3515 0.2401

Maharashtra 0.0773 0.1903 0.2810 0.3276 0.1237

Orissa 0.1973 0.2693 0.2948 0.1911 0.0476

Punjab 0.0407 0.0813 0.2088 0.4022 0.2670

Rajasthan 0.0346 0.0700 0.1441 0.3020 0.4493

Tamil Nadu 0.2834 0.2400 0.2257 0.1741 0.0768

Uttar Pradeshc 0.3143 0.2441 0.2338 0.1691 0.0386

West Bengal 0.3649 0.2995 0.2244 0.0753 0.0359

Lakshdweep 1.0000 0.0000 0.0000 0.0000 0.0000

A& N island 0.0370 0.1111 0.2963 0.3333 0.1852

aIncluding Jharkhand.bIncluding Chattisgarh.cIncluding Uttaranchal.

Table 7

List of input parameters used in calculations

Parameter Symbol Unit Value

Capital cost

of the

system

SPV pump (0.9 kWp) C0 Rs 170 000

SPV pump (1.8 kWp) Rs 328 000

Windmill pump Rs 104166

Producer gas based

dual fuel engine pump

(5 hp)

Rs 75000

Annual

operation,

repair and

maintenance

cost (as a

fraction of

capital cost)

SPV pump (0.9 kWp) m Fraction 0.01

SPV pump (1.8 kWp) Fraction 0.01

Windmill pump Fraction 0.02

Producer gas based

dual fuel engine pump

(5 hp)

Fraction 0.05

Capital subsidy (as a fraction of the

capital cost)

fcs Fraction 0.50

Amount of low interest loan (as a

fraction of the balance of the cost)

fis Fraction 0.80

Interest rate on loan r Fraction 0.10

Loan repayment period t Years 10

Calorific

value of

agricultural

residues

Arhar stalks CVar,j MJ/kg 14.88

Cotton stalks MJ/kg 17.88

Jute and sticks MJ/kg 19.04

Maize cobs and stalks MJ/kg 14.53

Mustard stalks MJ/kg 18.84

Residue to

crop ratio

Arhar stalks Rj Fraction 1.32

Cotton stalks Fraction 3.00

Jute and sticks Fraction 2.30

Maize cobs and stalks Fraction 1.86

Mustard stalks Fraction 1.85

Fraction of the total amount of crop

residues produced that may be lost

during collection.

fl Fraction 0.10

Fraction of the collected crop residues

that may be used in other applications

fou Fraction 0.20

Total

pumping

head

H m 5

Volume of water required for

irrigating one hectare land during a

single cropping season

Vw m3 6000

Efficiency of gasifier Zgas Fraction 0.80

Efficiency of dual fuel engine Zdfe Fraction 0.23

Efficiency of the pump used with the

dual fuel engine

Zdp Fraction 0.40

Diesel replacement factor of the

producer gas based dual fuel engine

pump

rd,pg Fraction 0.80

A. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870868

Karnataka. Once again due to dominance of marginal landholdings (unable to satisfy affordability criteria) theestimated potential of windmill pump utilization for thestate of Kerala is rather low.

Owing to a variation in the calorific value of differentcrop residues and the corresponding values of residue tocrop ratios, yields of the respective crops in the state, etc.,the number of feasible hours of operation of a 3.73 kW(5 hp) producer gas based dual fuel engine pumps wouldvary considerably for different combinations of residuesand states. The potential of producer gas based dual fuelengine pumps for irrigation pumping in India is estimatedat 0.46 million with highest potential in the state ofMaharashtra.

It is estimated that around 36 thousand communitybiogas plants of 20m3 daily biogas production capacity canbe installed for irrigation pumping in India. Since a dualfuel engine pump can be used with a community typebiogas plant, the potential number of biogas based dualfuel engine pumps is essentially equal to the potentialnumber of community biogas plants. Uttar Pradesh has thehighest potential (approximately 10,000 plants) of commu-nity biogas plants for irrigation pumping.

It may be noted that the number of SPV pumps users(Nspv) or windmill pumps users (Nwp) also correspond to

the area of land that one SPV pump or windmill pump caneffectively irrigate. However, this study assumes that onefarmer can not own more than on SPV pump or otherrenewable energy based pump. Though the size(s) of thesame may vary.

6. Concluding remarks

Though the compounded annual rate of growth in thenumber of electric pumps in the country is merely 3%

ARTICLE IN PRESS

Table 9

Potential of different renewable energy technologies for irrigation

pumping in India (thousands)

State SPV pumps Windmill

pumps

Producer gas

based dual

fuel engine

pumps

Biogas

based dual

fuel engine

pumps

0.9 kWp 1.8 kWp

Andhra Pradesh 179 407 274 49 2.42

Assam 59 68 —a 18 1.99

Biharb 161 267 — 34 6.43

Gujarat 93 331 442 45 0.91

Haryana 30 95 — 21 0.55

Himachal

Pradesh

— — — 7 0.22

Karnataka 139 367 549 33 1.08

Kerala 15 14 38 — 1.33

Madhya Pradeshc 211 783 81 18 0.94

Maharashtra 300 757 643 88 1.41

Orissa 86 118 90 6 1.99

Punjab 29 132 — 14 0.67

Rajasthan 77 439 123 33 0.39

Tamil Nadu 83 139 132 12 2.52

Uttar Pradeshd 215 330 3 63 10.31

West Bengal 61 43 — 21 2.51

Goa 1 2 9 — —

Andmand and

Nicobar Island

— — 6 — —

All India 6031 2390 462 35.67

aNegligible or not estimated.bIncluding Jharkhand.cIncluding Chattisgarh.dIncluding Uttaranchal.

Table 8

Categories of farmers who can afford to invest in SPV pump, windmill

pump, and producer gas based dual fuel engine pump

Renewable energy

option for water

pumping

Size Categories of farmers

who can afford to

invest in the

renewable energy

option

SPV pump 0.9 kWp Semi medium

1.8 kWp Medium, large

Windmill pump — Small, semi medium,

medium, large

Producer gas based

dual fuel engine

pump

3.73 kW (5 hp) Small, semi medium,

medium, large

A. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870 869

(primarily due to increasing uncertainty in the availabilityof electricity) the incremental annual number of newelectric pumps is far more than the reported cumulativenumber of renewable energy based pumps disseminated sofar. It is therefore, extremely important to developappropriate technologies for this purpose beside formulat-ing and implementing suitable dissemination strategies forrenewable energy based options for water pumping.

Even with rather conservative values of relevant inputparameters the estimates indicate that, there is a vastpotential of renewable energy utilization for irrigationwater pumping in India. Out of four renewable energytechnologies considered in this study, the estimatedutilization potential of SPV pumps is highest followed bywindmill pumps. The potential is estimated considering theprevailing conditions such as cost and technical specifica-tion of renewable energy based devices, income of farmers,etc. The future potential of renewable energy use mayincrease due to cost reductions and/or technology advance-ment, as well as increase in income of farmer.

References

[1] The Energy and Resources Institute. TERI energy data directory and

yearbook 2003/04. New Delhi, India: TERI Press; 2004.

[2] World Bank. Power supply to agriculture, vol. 1. Report no. 22171-

IN. Washington, DC: Energy Sector Unit, South Asia Regional

Office World Bank, 2001.

[3] The Fertilizer Association of India. Fertilizer statistics 1999–2000. 10,

New Delhi, India: FAI, Shaheed Jit Singh Marg; 2000.

[4] Ministry of Agriculture. Agricultural statistics at a glance. New

Delhi, India: MOA, Government of India, Krishi Bhavan, Dr.

Rajendra Prasad Road; 2002.

[5] Ministry of Agriculture. Indian agricultural statistics 1992–1993. vols.

I, II, all India, state-wise and district-wise. New Delhi, India: MOA,

Government of India, Krishi Bhavan, Dr. Rajendra Prasad Road;

1993.

[6] Central Ground Water Board. Ground water year book. Reports of

various states in India. New Delhi, India: CGWB, Government of

India, Jamnagar House, Mansingh Road, 2001.

[7] Central Electronics Limited. Solar powered water pumping system.

Product brochure. Sahibabad (UP), India: CEL, 4, Industrial Area;

2003.

[8] Bhattacharya T. Terrestrial solar photovoltaics. New Delhi, India:

Narosa Publishing House; 1998.

[9] Kumar A. CO2 emissions mitigation by using renewable energy

technologies for irrigation water pumping and crop drying in India,

Doctoral thesis. New Delhi, India: Indian Institute of Technology

Delhi; 2004.

[10] United Nations Economic and Social Commission for Asia and

Pacific. Solar powered water pumping in Asia and the Pacific.

Bangkok, Thailand: UNESCAP, The United Nations Building,

Rajadamnern Nok Avenue; 1991.

[11] Mani A, Rangarajan S. Solar radiation over India. New Delhi, India:

Allied Publishers Private Limited; 1982.

[12] Mani A, Mooley DA. Wind energy data for India. New Delhi, India:

Allied Publishers Private Limited; 1983.

[13] Mani A. Wind energy resources survey for India—I. New Delhi,

India: Allied Publishers Private Limited; 1990.

[14] Mani A. Wind energy resources survey for India—II. New Delhi,

India: Allied Publishers Private Limited; 1992.

[15] Mani A. Wind energy resources survey for India—III. New Delhi,

India: Allied Publishers Private Limited; 1994.

[16] Patel MR. Wind and solar power systems. New York: CRC Press;

1999.

[17] Ministry of Non-Conventional Energy Sources. Global windpump

evaluation programme, preparatory phase: country study on India.

New Delhi, India: MNES, Government of India, 14 CGO Complex.

Lodhi Road, 1987.

[18] Rai SN, Chkrabarty SK. Demand and supply of fuelwood and timber

in India. Ind Forest 2001;3:263–79.

[19] Grover PD. Thermochemical characterization of biomass residues for

gasification. New Delhi, India: Biomass Research Laboratory,

ARTICLE IN PRESSA. Kumar, T.C. Kandpal / Energy 32 (2007) 861–870870

Chemical Engineering Department, Indian Institute of Technology

Delhi; 1989.

[20] National Productivity Council. A report on improvement of

agricultural residues and agro-by-products utilisation. New Delhi, India:

NPC, Utpadakta Bhavan, 5-6 Institutional Area, Lodhi Road; 1987.

[21] Tripathi AK, Iyer PVR, Kandpal TC, Singh KK. Assessment of

availability and costs of some agricultural residues used as feedstocks

for biomass gasification and briquetting in India. Energy Conver

Manag 1998;39(15):1611–8.

[22] Purohit P, Kumar A, Rana S, Kandpal TC. Using renewable energy

technologies for domestic cooking in India: a methodology for

potential estimation. Renew Energy 2002;26(2):235–46.

[23] Centre for Monitoring Indian Economy. Basic statistics relating to

the Indian economy. Bombay, India: Centre for Monitoring Indian

Economy Private Ltd.; 1993.

[24] Rubab S, Kandpal TC. A methodology for financial evaluation of

biogas technology in India using cost functions. Biomass Bioenergy

1996;10(1):11–23.