Resources, Conservation and Recycling 50 (2007) 459–474

Rainwater conservation and recycling by optimal size on-farm reservoir Balram Panigrahi a , Sudhindra N Panda b,∗ , Bimal Chandra Mal b a

Water Management Project, R. R. T. T. S., Chiplima, Sambalpur, 768 025 Orissa, India b Department of Agricultural and Food Engineering, Indian Institute of Technology, Kharagpur, 721 302 West Bengal, India

Received 31 January 2006; received in revised form 4 August 2006; accepted 8 August 2006 Available online 1 November 2006

Abstract Hydrologic and economic analysis of the on-farm reservoir (OFR) was carried out in rainfed rice–mustard cropping systems in Eastern India followed by 2 years of field experiments in 1999 and 2000. The average contribution (average of 2 years) of direct rainfall and surface runoff from the diked crop fields contributed, respectively, about 79.5 and 20.5% to the total OFR inflow. The average contribution of evaporation loss, seepage and percolation loss and supplemental irrigation from the OFR contributed, respectively, about 10.0, 31.2 and 58.8% to the total OFR outflow. There was an average increase of rice yield of 44.0% over the rainfed rice because of application of 8.4 cm supplemental irrigation from the OFR. Thus, with an application of 4.5 cm supplemental irrigation from the OFR, 15.40% increase of mustard yield was recorded in 1999. Economic analysis indicated average net profit of Rs. 700 (US$ 1 = Rupees (Rs.) 44.75 in Indian currency) from a farm area of 800 m2 . Average values of benefit–cost ratio, internal rate of return and pay back period of the OFR irrigation system were evaluated as 1.17, 14.8%, and 16 years, respectively. The study reveals that the OFR irrigation in small landholders is economically feasible system for rainwater harvesting and providing supplemental irrigation in rainfed farming system. © 2006 Elsevier B.V. All rights reserved. Keywords: Economic analysis; On-farm reservoir; Rainfed farming; Rainwater harvesting; Supplemental irrigation

∗

Corresponding author. Tel.: +91 3222 283140; fax: +91 3222 282244. E-mail address: [email protected] (S.N. Panda).

0921-3449/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.resconrec.2006.08.002

460

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

1. Introduction Rainwater harvesting agriculture is a specilised form of rainfed farming that has a significant potential to increase food production. Rain is the cheapest and often only available source of water for agriculture purposes in many places. In many dry regions and other places in the world where irrigation water is a scarce commodity, there is no alternative but a better and more effective use of rain to increase food production. The area under rainfed rice in Eastern India is about 67% of the country’s total rainfed rice area. Yet, its contribution to the country’s rice production is only about 50% (Ghildyal, 1989). Eastern region of India continues to lag behind other regions in production and productivity of all crops including rice because most of its rainfed lands lack adequate water supply. Rainfed rice and drought problems usually go hand in hand for most of the areas in the region. Enormous losses of rice production occur every year from drought in different rainfed rice growing regions of Eastern India. On an average, two continuous dry spells of 12 days duration each occur during the rice growing periods in rainy season (June–September). Out of this, the second continuous dry spell coincides with the reproductive phase of rice that may be detrimental for obtaining higher yield (Panigrahi et al., 2002). The most persistent constraint in increasing productivity from the rainfed areas of Eastern India is the lack of assured irrigation supply. With the present population growth rate, the productivity of rainfed rice of the region has to be raised from the present level of 800–2000 kg ha−1 (Central Research Institute for Dryland Agriculture, 1995). Problem will be still worse for the region especially for the rainfed upland rice farmers since conventional irrigation facilities are unlikely to be developed in near future because of undulating topography, surface drainage problem and financial constraints. Upland rice farmers are among the poorest of the world’s rice farmers. Modern rice technology developed for irrigated ecosystems cannot be transferable in toto to the rainfed uplands. Hence, generating technology for improved rice productivity in sustainable rainfed upland farming systems is very critical. Farmers of Eastern India depend entirely on southwest monsoon for their crop cultivation. Most often, temporal and spatial distribution of rainfall and its unpredictability affect rice yields. Even with the high concentrations of rainfall during the effective monsoon/rainy season (June–September), rainfed rice in most of the years suffers from in-season drought and consequent loss of yield. Eastern India is bestowed with ample of rainfall resources with average annual rainfall of 1500 mm, 80% of, which is concentrated during the rainy season between June and September. During this period, about 50% of the annual rainfall comes from a few intense storms. Water received from such intense storms is subjected to high runoff losses that cause a lot of soil and nutrient erosion. The fate of millions of rice growers in the region can be greatly improved by the use of technological advances such as effective rainwater conservation and management. Use of the on-farm reservoir (OFR) is one such alternative to overcome drought and to increase land productivity in rainfed areas. Constructed on the farm, the OFR helps the farmers to harvest runoff produced in the crop field and in situ conservation of rainfall and store the same for subsequent use to crops at their needs (Zimmerman, 1966). In recent years, several researchers have pursed their interest in the use of the OFR system to alleviate drought in rainfed areas. Vittal et al. (1988) reported that for semi arid regions of South India, an OFR area of 300 m2 for a 1.0 ha catchment would be desirable

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

461

to provide enough stored water for supplemental irrigation to various dry land cropping systems and would be helpful in increasing and stabilizing the crop production. Oweis et al. (1999) have reported that in Western Asia and North Africa, there would be acute shortage of irrigation water in near future. If agricultural production is to be sustained, even at current levels, increasing the productivity of water in agriculture in dry areas becomes crucial. In such context, water harvesting and supplemental irrigation plays a vital rule. Water harvesting systems for crop production has a large potential in semi-arid regions of Eastern and Southern Africa: a potential that to a large extent is untapped. Although some farmers use water harvesting systems to supplement a crop with water during stress periods, there is much less efforts to design and manage the systems on a catchment scale and so research has to be conducted on-farm addressing bio-physical and socio-economic issues (Rockstrom, 2000). In Central Luzon, Philippines, farm reservoirs are very popular among the farmers to supplement rainfall for wet season rice production and to irrigate a dry season rice crop and are cost effective. Farmers without reservoirs cannot grow a dry season crop (Guerra et al., 1990). In Bangladesh, the OFRs have been utilised successful in alleviating drought and have helped to increase and stabilize the yield through supplemental irrigation. The system is cost effective and most farmers can pay back their OFR investment cost in 5–8 years (Islam et al., 1998). Similar findings on cost effectiveness of the OFR are also reported by Pal et al. (1994) for India and Syamsiah et al. (1994) for Indonesia. But in spite of these advances, there is a lack of understanding of the physical, hydrological and socio-economic factors that affect the successful operation and performance of the OFRs. The purpose of this study is to develop a better understanding of the hydrological characteristics of the OFRs, which is necessary for assessing the potential viability of these OFRs in other areas. Another objective of this paper is to evaluate the economic feasibility of the OFRs for drought alleviation in rainfed farming systems of Eastern India.

2. Materials and methods 2.1. Study area The site selected for the study was an agricultural farm of the Indian Institute of Technology, Kharagpur that lies in the West Bengal State of Eastern India. It is located at a latitude of 22◦ 19 N, longitude of 87◦ 19 E with an altitude of 48 m above mean sea level. It receives average annual rainfall of 1500 mm. The mean minimum and maximum temperatures are 12 and 40 ◦ C, occurring in the month of January and May, respectively. The mean relative humidity ranges from 35.5 to 90.5%. The dominant soil group is sandy loam (light textured), acid lateritic with pH ranging from 4.8 to 5.6, and poor in organic matter. Permanent wilting point, field capacity, and saturation moisture content of the soil are 9.3, 26.7, and 37.7 cm/m, respectively. The soils have very low water holding capacity and dries up quickly after cessation of rainfall. Hence, cultivation of crop on residual moisture is difficult. Rice is the major crop grown in the region in monsoon season covering about 90% of total cultivated area. In winter, after cultivation of rice, some farmers sow short duration and low water requiring crops like mustard, if the residual soil moisture is sufficient for seed germination.

462

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

2.2. Experimental technique and data collection The experiments were conducted during 1999 and 2000. Average farm sizes in Eastern India vary from 500 to 1000 m2 and farmers mostly prefer to have an independent source of water supply through the OFR by constructing it at one corner of their fields. Hence, in the present study, for hydrologic and economic analysis, we have taken four plots with farm area of 800 m2 each (40 m × 20 m). These four plots constituted two treatments with two replications each. The treatments were rainfed and irrigated, respectively. Under rainfed condition, crops were raised based on rainfall only without any supplemental irrigation (SI). Crops under irrigated condition were provided with SI from the OFR. Two OFRs were constructed in two irrigated plots at one corner of the plot such that each plot had an OFR. All the four plots were perfectly leveled. The cropping systems followed in all the four plots were rice in wet (rainy) season followed by mustard in dry (winter) season. For the said cropping pattern and farm area of 800 m2 in the study area, simulation study had been carried out and reported that 12% of the farm area is optimum for construction of the low density polyethylene (LDPE) lined OFR that gives maximum net profit and benefit–cost ratio (Panigrahi, 2001). Hence, based on the aforesaid simulation, two squareshaped pyramidal dug out pond/OFR were constructed with dimensions of each OFR as shown in Fig. 1. The storage volume of each OFR below ground level was 61 m3 . Dike heights of 30 cm around all the four plots were strengthened to conserve entire in situ rainwater in the field and/or OFR and not allowing any drainage water to enter the field from surroundings. However, to allow surface runoff to enter the OFR from the field, an inlet pipe was placed in the embankment of the OFR at 5 cm above the ground level (Fig. 1). During 10 days after the germination of rice till last 10 days to its harvest, excess of 5 cm ponding in the field was drained to the OFR through this pipe. Provision was made to lower this pipe to ground level, so that any ponding above the ground level in field could be drained to the OFR during rest of the periods in rice field, periods from harvest of rice to sowing of mustard and entire mustard growing period. The soil of the experimental site is light textured having high seepage and percolation (SP) losses. So, to check these losses, the OFR was lined at both bottom and sides by 600-gauge LDPE sheets and was covered by 30-cm soil layer. Panigrahi (2001) reported that by covering the water surface of the OFR with cut pieces of 600-gauge LDPE sheets, evaporation losses could be reduced by 50% as compared to that from the OFR with open top surface. Since, irrigation water in the rainfed ecosystem is a scarce commodity, the OFR was covered by the cut pieces of 600-gauge LDPE sheets so that it would not affect in allowing the in situ rainwater to fall in the OFR and at the same time it would reduce the evaporation losses in the OFR. Water saving irrigation technique was adopted in the study for supplying SI to rice. As per this technique, SI of 5 cm or the amount actually available in the OFR at that time to support irrigation (which ever is minimum) was applied to rice as flow irrigation during the reproductive stage when the management allowable deficit of soil moisture in the effective root zone of the crop was at 20% below saturation level moisture content. At other stages of growth of rice, crop was raised with rainfed condition without provision of any SI. For the 45 cm effective root zone depth of rice, the saturation level moisture content for the experiment was worked out to be 170 mm. Reproductive stage of rice (Variety MW 10 of 101 days duration taken in the study) started from 45 days after germination of rice

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

463

Fig. 1. The OFR constructed at the experimental site for field verification. (a) Cross sectional view of OFR; (b) plan of OFR; l and b = length and width of OFR at bottom, respectively = 3.4 m; Do = depth of OFR below ground level = 2.0 m; L and B = length and width of OFR at ground level, respectively = 7.4 m; Ls and Bs = length and width of OFR at top, respectively = 9.8 m; Sl = side slope of OFR = 1:1 (horizontal:vertical); Te = top width of embankment = 30 cm; He = height of embankment = 30 cm; Be = bottom width of embankment = 90 cm; Se = side slope of embankment = 1:1 (horizontal:vertical); Bw = berm width = 30 cm.

and continued for 30 days. Stored water of the OFR was used as pre-sowing irrigation to mustard crop grown in winter (dry season) if the residual soil moisture was not sufficient for germination of seeds. If in a year, the pre-sowing irrigation to mustard was not required, then the stored water of the OFR was used as SI to mustard at other periods when the actual evapotranspiration fell below the potential value. In mustard, the amount of SI from the OFR was used till the soil moisture content of the effective root zone of mustard (100 cm found by the authors from root sampling) came to field capacity. Mustard crop was irrigated by sprinkler method of irrigation. Daily soil moisture sampling by a calibrated neutron probe moisture meter as well as gravimetric method helped to decide the timing of irrigation. Rice was direct sown on 11th and 10th June in 1999 and 2000, respectively. The crop was harvested on 22nd and 21st September in 1999 and 2000, respectively. After the harvest of rice, mustard (variety B-54 of 70 days duration) was sown on 8th October and harvested

464

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

Fig. 2. Nomograph used for computation of various OFR water balance parameters.

on 19th December in 1999. In irrigated mustard, pre-sowing irrigation was not required during sowing time for germination of seeds in 1999 and so the stored water in the OFR was used as SI during the pod formation stage when the actual evpotranspiration was found to be lower than the potential value. The year 2000 was a drought year. The residual soil moisture during the sowing time of mustard was not sufficient for germination of the seeds. Moreover, there was not sufficient water in the OFR to support the pre-sowing irrigation demand of mustard. Because of the afore-mentioned reasons, mustard could not be grown either in the rainfed or in the irrigated condition in 2000. A staff gauge was installed at the center of each OFR to indicate the water level reading (depth of water) in the OFR. From the known depth of water level in the OFR and dimensions of the OFR, capacity/volume of the OFR storage, top water surface area and wetted surface area of the OFR were computed (Panigrahi, 2001). A nomograph (Fig. 2) was prepared for different sizes of the OFR (all OFRs were square sized trapezoidal type) ranging from 6 to 20% of farm area (800 m2 ). For these different sizes of the OFR, all dimensions were same as mentioned in Fig. 1. The nomograph was used to estimate the seepage and percolation and evaporation losses from the OFR and determine its storage at any known depth of water on any day. Daily water balance of the OFR was carried out during the rice–mustard-cropping season by the water balance model as presented below. Using installed staff gauges, water depth in each OFR was recorded daily at a fixed time and immediately before and after irrigation. Changes in storage and irrigation amounts used were calculated on a daily basis by using the depth–volume graphs (Fig. 2). Pan evaporation rates in U.S. Class A pans and rainfall depth was recorded daily at the meteorological center located at close proximity to the experimental site. Moreover, yield of rice grains and mustard seeds including their by-products were

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

465

also recorded at harvest time for both rainfed and irrigated treatments. Finally, the economics of irrigation system was worked out to justify the investment in construction of the OFR. 2.3. Water balance model of the OFR Water balance model of the OFR was run by considering all the inflow and outflow components of the OFR. The inflows are the direct rainfall in the OFR and surface runoff from the field coming to the OFR and the outflows are evaporation, seepage and percolation and supplemental irrigation supplied to crops in the field. The various components of the OFR water balance model are: Si − Si−1 = Si = QSRi + QPi − QEi − QSIi − QSi

(1)

where S is the OFR water storage, m3 ; QSRi the volume of surface runoff coming from the field to the OFR, m3 ; QPi the volume of direct rainfall in the OFR, m3 ; QEi the volume of water lost as evaporation from the OFR, m3 ; QSIi the volume of water used as supplemental irrigation in the cropped field, m3 ; QSi the volume of water lost as seepage and percolation from the OFR storage, m3 and i is the time index taken as 1 day in the study. Value of QSRi is given as: QSRi = SRi Afield

(2)

where SR is the surface runoff coming from field to the OFR and Afield is the field area given as: Afield = FA − AOFR

(3)

where, FA is the farm area and AOFR is the area of the OFR. The value of surface runoff coming from rice field to the OFR on any day is estimated by knowing the depth of water drained to the OFR. The depth of water is estimated by taking the difference of the depth of ponded water in the rice field on the day before opening the inlet pipe in the OFR and the predetermined ponded depth of water need to be kept in field mentioned as earlier. The ponded depth of water in rice field is computed by a water balance model (Panigrahi, 2001), which includes evapotranspiration and seepage losses in the field draining to the OFR. No ponding was allowed in the mustard field. The surface runoff contributed by the mustard field to the adjacent OFR was estimated by a two-layered soil water balance model as proposed by Panigrahi (2001). Value of QPi is given as: QPi = Pi AOFR

(4)

Value of QEi of the OFR covered by cut pieces of 600-gauge LDPE sheets is given by Panigrahi (2001) as: QEi = 0.37Epani Awsi

(5)

where Epani is the pan evaporation on ith day and Awsi is the top water surface area/water spread area of the OFR on ith day, which is a function of depth of water in the OFR.

466

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

Major losses of water in the OFR are due to seepage and percolation that can be checked by lining it with plastic films. Daily seepage loss (QSi ) is computed by the OFR water balance model (Eq. (1)) by knowing all other parameters of the model. Value of QSIi from the OFR is given as QSIi =

SIi Afield η

(6)

where η is the efficiency of irrigation system. The cumulative storage (S) in the OFR is: S=

n 

Si

(7)

i=1

where n is the number of days since observation is taken. 2.4. Economic analysis In any economic analysis, all cash flows must be evaluated at some reference time. A present worth analysis is used for the study since all cash flows over the life of the system can thus be converted, using appropriate factors to account for interest and inflation into an equivalent present value (Palmer et al., 1982). In the economic model used for the study, following items are considered: (a) initial investment, (b) the OFR maintenance cost, (c) irrigation cost, (d) land lease cost and (e) annual returns from irrigation. It is assumed that all inputs except irrigation are same for both irrigated and rainfed crops. Initial investment of the OFR irrigation system includes the following components: (a) cost of construction of the OFR, (b) material cost for lining the OFR and covering its top surface and (c) labor cost for lining. Initial investment is fixed for a particular size of the OFR for all years during its life span. The OFR maintenance cost, land lease cost and irrigation cost constitute the total yearly cost that is also called as the variable cost since they vary from year to year depending on the interest rate and inflation. Cost of construction of the OFR is computed by knowing the volume of the OFR that needs to be excavated below the ground level and the rate of excavation of the soil. The rate of excavation of the soil as per the 1999 government schedule rate is Rs. 14.85 m−3 (Rs. indicates the Indian currency rupees, and US$ 1 = Rs. 44.75) for a depth of excavation up to 2 m. For the present study, the volume of each OFR for which excavation work is done is estimated to be 61 m3 . The lining cost of the OFR by 600 gauge LDPE sheets is computed by knowing the area of LDPE sheets required for lining both the beds and sides and its present market price, which is Rs. 8.38 m−2 . Similarly, the cost of covering the top surface area of the OFR by 600-gauge LDPE sheets is computed by knowing the top surface area of the OFR and its present market price, which is same as mentioned above. The labor cost/wages for lining the OFR with LDPE sheets is four men-days. The government schedule rate of labor wage for the year 1999 is Rs. 50.00 per man-day. The OFR maintenance cost is a constant yearly cost assumed as 2% of the initial investment (Palmer et al., 1982; Mishra et al., 1998). Maintenance cost is used to desilt the OFR every year and maintain the dikes. For construction of the OFR, some land is utilised which

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

467

otherwise would have been available for cultivation. Since it is difficult to calculate the cost involved due to the utilization of this land, the cost may be assumed to be equal to the prevailing lease rate of upland. The lease rate of land in the study area in 1999 is Rs. 1200.00 ha−1 per year for rainfed farming system. Rice is irrigated by flow irrigation with a five-horse power diesel pump and mustard is irrigated by sprinkler irrigation. As per the hiring rate in the study area in 1999, irrigation cost comes to Rs. 225.00 for providing 5 cm SI to 1 ha of both the crops. Annual returns from irrigation are the values of the increased yields of the crops (both rice and mustard) over the rainfed ones due to SI. This is calculated by knowing the increased yield due to application of SI and the minimum support price of the products (both rice grains/mustard seeds and rice straw/mustard stover). Minimum support price of rice grains and mustard seeds as proposed by the government of India for 1999 are Rs. 490.00 and 1000.00 per 1000 kg, respectively. The existing prices of rice straw and mustard stover in the study site in 1999 are Rs. 30.00 and 15.00 per 100 kg, respectively. 2.5. Present worth analysis As proposed by Palmer et al. (1982), a present worth analysis is used to evaluate all the cash flows in order to account for the interest and inflation factor in investment. The present worth of total yearly cost A, i.e. PWA as: PWA =

n 

A(1 + f )n−1 (1 + r)−n

(8)

i=1

where A is the total yearly cost that is summation of the OFR maintenance, land lease and irrigation cost; f the inflation rate; r the interest rate and n is the life of the OFR irrigation system. The total cost of the irrigation system is summation of initial investment and the present worth value of total yearly cost (Palmer et al., 1982; Islam et al., 1998). In the similar way of total yearly cost, present worth value of total annual returns from irrigation (PWRI ) is also computed. In the economic analysis, 6% inflation rate, 12% interest rate (the rate of agricultural loan from bank for the year 1999) and 25 years economic life of the OFR with LDPE sheets lining are assumed (Mishra et al., 1998). 2.6. Economic indicators In the study, we have used four economic indicators, i.e. net profit (NP), benefit–cost ratio (BCR), pay back period (PBP) and internal rate of return (IRR) for economic analysis of the OFR irrigation system. The net profit is calculated by subtracting the total cost of irrigation system from the present worth value of total annual returns. Benefit–cost ratio is calculated as the ratio of present worth value of total annual returns to the total cost of irrigation system. Internal rate of return is computed as that interest rate at which the BCR value is just 1.0. It is estimated as suggested by Islam et al. (1998). Pay back period is estimated as suggested by Mishra et al. (1998).

468

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

Fig. 3. Daily variation of OFR storage, rainfall and supplemental irrigation in 1999 during the rice–mustard cropping season.

3. Results and discussion 3.1. The OFR water balance Daily variation of the OFR storage during the rice–mustard cropping season in 1999 and 2000 are presented in Figs. 3 and 4, respectively. These figures indicate that the OFR storage ranges from 0 to 58.77 m3 and 0 to 43.19 m3 in 1999 and 2000, respectively. The components of the OFR inflows and outflows for the years 1999 and 2000 are presented in Tables 1 and 2, respectively. From these tables, it is observed that the total inflow to the OFR during the entire experimental period of rice–mustard cropping season is higher than the OFR storage capacity. This reflects the dynamic process of the water collection and use that

Fig. 4. Daily variation of OFR storage, rainfall and supplemental irrigation in 2000 during the rice–mustard cropping season.

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

469

Table 1 Water balance parameters of OFR during 1999 Parameters (m3 )

Rice growing season (1)

Initial storage 2.22 Rainfall contribution 94.62 (67.1) Surface runoff contribution 46.39 (32.9) Total inflow 141.01 Evaporation loss 9.51 (8.2) Seepage and percolation loss 35.74 (30.9) Supplemental irrigation 70.40 (60.9) Total outflow 115.65 Final storage 27.58

Land preparation period for sowing of mustard (2)

Mustard growing season (3)

Rice–mustard-growing seasona (4) = (1) + (2) + (3)

27.58 8.11 (100.0) 0 (0.0) 8.11 1.32 (20.0) 5.27 (80.0) 0 (0.0) 6.59 29.10

29.10 12.05 (89.1) 1.48 (10.9) 13.53 3.18 (7.5) 7.77 (18.2) 31.68 (74.3) 42.63 0

2.22 114.78 (70.6) 47.87 (29.4) 162.65 14.01 (8.5) 48.78 (29.6) 102.08 (61.9) 164.87 0

Note: values in parenthesis represent percentage of total inflow/outflow. a Rice–mustard growing season refers to the time period from the sowing of rice to harvest of mustard including the land preparation period for sowing of mustard. Summation of column (1), (2), and (3) is not applicable in case of initial and final storage.

occurs in the OFR. The storage efficiency index that is expressed as the ratio of total inflow to the OFR storage capacity (Guerra et al., 1994) is found to be 2.62 in 1999 and 1.22 in 2000. A low value of storage efficiency index in 2000 is obtained due to low rainfall and surface runoff generated from the field. Islam et al. (1994) have also obtained a low value of storage efficiency index (1.01) of the OFR for rice field receiving an annual rainfall of 81.4 cm. 3.2. OFR inflows Rainfall constitutes the major components to total inflow. This is 67.1 and 88.3% of total water inflows during rice growing season in 1999 and 2000, respectively. The direct rainfall Table 2 Water balance parameters of OFR during 2000 Parameters (m3 )

Rice growing season (1)

Land preparation period for sowing of mustard (2)

Mustard growing season (3)

Rice–mustardgrowing seasona (4) = (1) + (2) + (3)

Initial storage Rainfall contribution Surface runoff contribution Total inflow Evaporation loss Seepage and percolation loss Supplemental irrigation Total outflow Final storage

10.83 65.30 (88.3) 8.66 (11.7) 73.96 7.82 (10.1) 22.25 (28.7) 47.38 (61.2) 77.45 7.34

7.34 0.28 (100.0) 0 (0.0) 0.28 0.73 (24.2) 2.29 (75.8) 0 (0.0) 3.02 4.60

4.60 0 (0.0) 0 (0.0) 0 1.12 (24.3) 3.48 (75.6) 0 (0.0) 4.60 0

10.83 65.58 (88.3) 8.66 (11.7) 74.24 9.67 (11.4) 28.02 (32.9) 47.38 (55.7) 85.07 0

Note: values in parenthesis represent percentage of total inflow/outflow. a Rice–mustard growing season refers to the time period from the sowing of rice to harvest of mustard including the land preparation period for sowing of mustard. Summation of column (1), (2), and (3) is not applicable in case of initial and final storage.

470

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

contribution to OFR is 70.6 and 88.3% of total water inflows during the entire rice–mustardcropping season in 1999 and 2000, respectively (Tables 1 and 2). Total rainfall during the rice and rice–mustard growing season in 1999 is obtained as 98.6 and 119.6 cm, respectively. Similarly, 68 and 68.3 cm rainfall are recorded during the rice and rice–mustard growing season in 2000, respectively. In the study, no mustard crop is grown in fields in 2000 but the OFR water balance is carried until a period that would have taken for the harvest of the crop had it been sown that is 70 days after sowing of the crop. The OFR water balance till the harvest day of mustard in 2000 is carried on for the sake of comparison of the data in both 1999 and 2000. During the rice-growing season in 1999 and 2000, surface runoff generated from the upland rainfed rice field having light textured soil contributes 32.9 and 11.7%, respectively, of the total OFR inflow volume. This also accounts for 29.4 and 11.7% of the total OFR inflow volume during the entire rice–mustard-cropping season in 1999 and 2000, respectively (Tables 1 and 2). A higher value of surface runoff contribution to the OFR in 1999 is because of adequate rainfall that generates higher values of surface runoff (6.59 and 1.73 cm during rice and mustard growing season, respectively) from the cropped field. On the other hand there is inadequate rainfall in 2000 that produces only 1.23 cm surface runoff from the rice field (no runoff from the mustard field as there is no rainfall during mustard growing season) and thereby contributing a lower percentage to total water inflows. 3.3. OFR outflows Evaporation loss accounts for 8.2 and 8.5% of the total water outflow from the OFR in 1999 during the rice-growing season and rice–mustard growing season, respectively. The loss due to evaporation is observed to be more for 2000 than that for 1999 accounting for 10.1 and 11.4% of the total water outflow during rice and rice–mustard growing season, respectively (Tables 1 and 2). The reason for higher values of evaporation loss in 2000 may be due to less number of rainy days as compared to 1999. This is obvious from the fact that there are three continuous dry spells in 2000 as against one in 1999. Evaporation loss from the OFR is reported to vary from 10 to 25% of the total losses by several researchers (Guerra et al., 1994; Syamsiah et al., 1994; Pal et al., 1994). Seepage and percolation loss in the OFR accounts for 30.9 and 28.7% of the total water outflow during the rice-growing season in 1999 and 2000, respectively. However, it accounts for 29.6 and 32.9% of the total losses during the whole rice–mustard-growing season, respectively, for these years (Tables 1 and 2). Losses of SP in unlined OFR are reported to vary from 45 to 67% of total outflow volume (Guerra et al., 1994; Pal et al., 1994). In the present field experiment, the SP losses are observed to be less because of the LDPE lining. On an average the SP loss during the experimental period (mean of 2 years) is found to be 6.56 l m−2 day−1 (6.56 mm day−1 ) whereas the average SP loss of the unlined OFR in the study site is earlier reported to be 135 l m−2 day−1 (135 mm day−1 ) (Pant, 1988). Values of SP losses in 600-gauge LDPE lined OFR with 20 cm thick soil cover in both sides and bottoms of the OFR are observed to be 7 l m−2 day−1 by Sharma et al. (1992) and 11.76 l m−2 day−1 by Pant (1988). Verma and Sarma (1990) have reported that with LDPE lining of 800-gauge thickness in the bottom and 75 mm thick brick cement lining in the sides of OFRs, SP losses of the unlined OFR (520 l m−2 day−1 ) is reduced to 12.71 l m−2 day−1 .

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

471

Water used for SI to rice and mustard crops accounts for the highest fraction of total water outflow in both the years. It constitutes 60.9 and 61.2% of the total water outflow volume during the rice-growing season in 1999 and 2000, respectively. Considering the entire rice–mustard cropping season, SI uses from the OFR are observed to be 61.9 and 55.7% of the total outflow volume in the above-mentioned 2 years in sequence (Tables 1 and 2). In total, 70.40 m3 of water from the OFR is used in 1999 that provides two SI to rice amounting to a total of 10 cm to an area of 704 m2 . In addition, 42.63 m3 (Table 1) of stored water is available in the OFR when one SI of 4.5 cm is applied to the mustard field of 704 m2 areas during the pod forming stage in 1999. Volume of water used for SI to rice in 2000 is 47.38 m3 that provides two SI of 5 and 1.73 cm each. Since there is inadequate rainfall and more number of dry spells in 2000, storage of water in the OFR is observed to be less and so volume of water used for SI is less. Moreover, after the termination of SI to rice, little rainfall till sowing of mustard (14.02 cm) along with no surface runoff from the field is recorded. As a result only 4.60 m3 of water is left over in the OFR in 2000 at the time of sowing of mustard (Table 2). This limited stored water was not sufficient to provide pre-sowing irrigation to mustard and so the crop (mustard) could not be grown in 2000. 3.4. Yield response Yield of rice grain under rainfed condition is found to be 3050 and 1511 kg ha−1 in 1999 and 2000, respectively with a mean value of 2280.5 kg ha−1 . Yield of rice grain under irrigated condition in 1999 and 2000 are recorded as 3951 and 2394 kg ha−1 , respectively, with mean value of 3172.5 kg ha−1 . Because of application of total 10 and 6.73 cm SI in 1999 and 2000, respectively, the yields of rice grains under irrigated treatment are found to increase by 29.5 and 58.4% in 1999 and 2000 over that under the rainfed treatment. Thus, the average increase of rice grain under the irrigated treatment is 44.0% because of average application of 8.37 cm of SI. The yield of rice grain in 1999 is high for both the treatments because of high (98.6 cm during rice growing season) and well distribution of rainfall. On the other hand, in 2000, there is only 68.0 cm of rainfall and further there are three continuous dry spells varying from 12 to 15 days. Out of these three continuous dry spells, the second one is observed to coincide with the last 13 days of reproductive stage of rice and it adversely affects the yield to a low level of 1511 kg ha−1 under rainfed condition. However, application of 6.73 cm SI during the reproductive stage in 2000 improves the yield of rice grain by 58.4% over the yield under rainfed condition. Application of 4.5 cm SI from the OFR to mustard during the pod formation stage in 1999 increases the yield by 15.4% (723.6 kg ha−1 under rainfed condition to 835 kg ha−1 under irrigated condition). In a field study, with application of 5 cm SI to rice and mustard each, Singh (1983) has observed 71.6 and 23.6% increase of rice and mustard yield, respectively. 3.5. Economic analysis Initial investment in the present study is computed as Rs. 2743.00, which is same for both the years 1999 and 2000. The OFR maintenance cost as well as land lease cost for the

472

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

Table 3 Economic viability study of the OFR irrigation system for a farm area of 800 m2 Parameters

Initial investment (Rs.) Total annual cost (Rs.) Present worth value of total annual cost (Rs.) Total cost (Rs.) Total return (Rs.) Present worth value of total return (Rs.) Net profit (Rs.) Benefit–cost ratio value Internal rate of return (%) Pay back period (years)

Year

Mean

1999

2000

2743 112 1399 4142 420 5234 1092 1.26 16.1 13

2743 88 1093 3836 333 4143 307 1.08 13.4 19

2743 100 1246 3989 377 4689 700 1.17 14.8 16

construction of the OFR are computed as Rs. 55.00 and 11.00, respectively, and are assumed to remain same for both the years. Irrigation costs are estimated as Rs. 46.00 and 22.00 in 1999 and 2000, respectively. Thus, the total annual cost of the OFR irrigation system is determined as Rs. 112.00 and 88.00 in 1999 and 2000, respectively (Table 3). Present worth value of the total annual costs are calculated as Rs. 1399.00 and 1093.00 in 1999 and 2000, respectively. Total costs of OFR irrigation system are determined as Rs. 4142.00 and 3836.00 in 1999 and 2000, respectively (Table 3). Total returns due to SI from the increased yields of rice grains including its byproduct and that of mustard seeds including byproducts in 1999 and 2000 are determined as Rs. 420.00 and 333.00, respectively. Present worth value of the total return are Rs. 5234.00 and 4143.00 in 1999 and 2000, respectively, with a mean value of Rs. 4689.00 (Table 3). 3.6. Economic viability study The economic viability of the OFR irrigation system is evaluated in terms of NP, BCR, IRR and PBP and is presented in Table 3. Values of NP, BCR, IRR and PBP in 1999 for a farm area of 800 m2 are computed as Rs. 1092.00, 1.26, 16.1% and 13 years, respectively. For the year 2000, values of NP, BCR, IRR and PBP for the same farm area are found to be Rs. 307.00, 1.08, 13.4% and 19 years, respectively. The economic viability factors in 2000 are found to be less as compared to those in 1999 because of the low yield of rice and for only one crop (rice) grown in the year. However, the BCR in this year is found to be more than 1.0 indicating that the OFR irrigation system is economically viable. The average values (average over 1999 and 2000) of NP, BCR, IRR and PBP are calculated as Rs. 700.00, 1.17, 14.8% and 16 years, respectively. Since average value of BCR is obtained to be more than 1.0, the investment in OFR irrigation system is justified. From the data of Table 3, it can be observed that the investment in OFR irrigation system can be paid back in about 16 years and during the rest of the periods of the life span of the OFR, profit from the system can be obtained. Thus, the investigation reveals that rainwater harvesting and recycling through the OFR in the eastern India has tremendous scope and it is viable both technically and economically.

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

473

4. Conclusions The average (average of 2 years) contributions of direct rainfall and surface runoff to the OFR during the study period are observed to be 79.5 and 20.5% of the total inflows, respectively. The average values of evaporation, seepage and percolation, and supplemental irrigation of the OFR during the study period in rice–mustard growing season are obtained to be 10.0, 31.2, and 58.8%, respectively, of the total OFR water outflows. The OFR supplies 10 and 6.73 cm of supplemental irrigation to rice in 1999 and 2000, respectively, and 4.5 cm supplemental irrigation to mustard in 1999. Yields of rice grain under rainfed condition are recorded to be 3050 and 1511 kg ha−1 in 1999 and 2000, respectively, whereas that under the irrigated condition are recorded to be 3951 and 2394 kg ha−1 in 1999 and 2000, respectively. The yield of mustard seed under rainfed condition in 1999 is measured as 723.6 kg ha−1 , whereas that under irrigated condition is recorded as 835.0 kg ha−1 . The economic analysis of OFR irrigation system indicates that the average values of net profit from 800 m2 farm areas with rice–mustard cropping pattern will be Rs. 700.00 with average benefit–cost ratio of 1.17. The mean value of internal rate of return and pay back period of the system are estimated as 14.8% and 16 years, respectively. Since, the value of benefit–cost ratio is more than 1.0, the experimental study reveals that OFR irrigation system is economically viable in the study region. References Central Research Institute for Dryland Agriculture. CRIDA towards 2020. In: Prospective plan. India: Indian Council of Agricultural Research; 1995. Ghildyal BP. Rice production in Eastern India-issues in rice research. J Agric Issues 1989;1:79–91. Guerra LC, Watson PG, Bhuiyan SI. Hydrological analysis of farm reservoirs in rainfed areas. Agric Water Manage 1990;17(4):351–66. Guerra LC, Watson PG, Bhuiyan SI. Hydrological characteristics of on-farm reservoir in rainfed rice-growing areas. In: Bhuiyan SI, editor. On-farm reservoir systems for rainfed ricelands. Manila, Philippines: International Rice Research Institute; 1994. p. 7–20. Islam MT, Saleh AFM, Bhuiyan SI. Agro-hydrologic and economic analyses of on-farm reservoirs for drought alleviation in the rainfed ricelands of northwest Bangladesh. Rural Environ Eng 1998;35:15–26. Islam MdT, Siddiqui MR, Hassan MdN, Islam MdN, Musa AM, Kar NK. On-farm reservoirs for drought alleviation in the rainfed ricelands of the Barind area of Bangladesh. In: Bhuiyan SI, editor. On-farm reservoir systems for rainfed ricelands. Manila, Philippines: International Rice Research Institute; 1994. p. 153–64. Mishra PK, Rama Rao CA, Siva Prasad S. Economic evaluation of farm pond in a micro-watershed in semi-arid alfisol deccan plateau. Ind J Soil Conserv 1998;26(1):59–60. Oweis, T, Hachum, A, Kijne, J, 1999. Water harvesting and supplemental irrigation for improved water use efficiency in dry areas. SWIM Paper No. 7. Colombo, Sri Lanka: International Water Management Institute; p. 41. Pal AR, Rathore AL, Pandey VK. On-farm rainwater storage systems for improving ricelands productivity in eastern India: opportunities and challenges. In: Bhuiyan SI, editor. On-farm reservoir systems for rainfed ricelands. Manila, Philippines: International Rice Research Institute; 1994. p. 105–25. Palmer WL, Barfield BJ, Haan CT. Sizing farm reservoirs for supplemental irrigation of corn. Part II. Economic analysis. Trans ASAE 1982;15:377–80, 387. Panigrahi, B, 2001. Water balance simulation for optimum design of on-farm reservoir in rainfed farming system. Ph.D. Thesis. Kharagpur, India: Indian Institute of Technology. Panigrahi B, Panda SN, Mull R. Prediction of hydrological events for planning rainfed rice. Hydrol Sci J 2002;47(3):435–48.

474

B. Panigrahi et al. / Resources, Conservation and Recycling 50 (2007) 459–474

Pant, N, 1988. Theoretical and experimental determination of seepage from lined and unlined farm ponds. M. Tech. Thesis. Kharagpur, India: Indian Institute of Technology. Rockstrom J. Water resources management in smallholder farms in Eastern and Southern Africa: an overview. Phys Chem Earth (B) 2000;25(3):275–83. Sharma A, Singh J, Singh R. Technical Journal of 7th National Convention of Agricultural Engineers. Water resources utilization and management on small watersheds in Rajasthan 1992:95–105. Singh, RP, 1983. Farm pond. Project Bulletin No. 6, All India Co-ordinated Research Project on Dryland Agriculture, Hyderabad, India. Syamsiah I, Fagi SAM, Bhuiyan SI. Collecting and conserving rainwater to alleviate drought in rainfed ricelands of Indonesia. In: Bhuiyan SI, editor. On-farm reservoir systems for rainfed ricelands. Manila, Philippines: International Rice Research Institute; 1994. p. 142–52. Verma HN, Sarma PBS. Design of storage tanks for water harvesting in rainfed areas. Agric Water Manage 1990;18:195–207. Vittal KPR, Vijayalakshmi K, Rao UMB. Interception and storage of surface runoff in ponds in small agricultural watersheds, Vol. 9. Andhra Pradesh, India: Irrigation Science; 1988. p. 69–75. Zimmerman JD. Irrigation. New York, NY: John Wiley and Sons; 1966. p. 203–53.