Evaluation of CropSyst model for clusterbean under hot arid condition

The study on “Evaluation of Cropsyst model for yield and water productivity of clusterbean” was conducted on farmers field during kharif 2012 at village Mainawali in Hanumangarh district of Rajasthan. The soils of the area are alluvial and calcareous in nature formed under arid and semi arid climate. The soils of site are brown to greyish brown and dark grey in colour, besides being calcareous and slightly alkaline in reaction having 67.7, 11.1 and 21.0 % of sand, clay and silt, respectively in 0-15 cm soil depth with pH 8.09 and low soil organic matter content. The simulate yield of clusterbean were closer to the observed clusterbean yield. Simulations of early clusterbean above ground biomass development matched the field data reasonably well. Final above ground biomass, however, was over estimated by the model. The total water applied in clusterbean was 405.8 mm out of this 326.7 mm consumed in ET. Thus, ET constituted 81% of total water applied and deep drainage constituted 13% and rest 6% stored as residual soil moisture.


INTRODUCTION
Water productivity, a concept expressing the value or benefit derived from the use of water, includes various aspects of water management and is very relevant for arid and semi-arid regions. It can be expressed in terms of grain (or seed) yield per amount of water used in different processes such as transpiration, evapotranspiration and percolation and provides a proper diagnosis of where and when water could be saved. Increasing water productivity is particularly appropriate where water is scarce compared with other resources involved in production.
Rajasthan is predominantly a rainfed state and precipitation being major source of annual renewable water supply. The total water resources of state account for 45.09 BCM, consisting 33.94 BCM share by surface water resources and 11.15 BCM by groundwater resources. The overall utilization of water resources is 81 % being 71 % for surface water and 104 % of groundwater resources. With the fast increasing population the water availability in the state is decreasing at an alarming rate and water scarcity is growing rapidly. According to an estimate, in the year 2001, the annual per capita water availability was 840 m 3 and expected to be as low as 439 m 3 by 2050 (Vision 2004a, 2004b. The situation of groundwater resources is very critical in the state. Out of total 237 groundwater blocks of the state, the number of safe blocks reduced to 162 to only 32 from 1984 to 2004, whereas in the same period the numbers of dark blocks has increased from 22 to 140. At present ~ 80.4 % of groundwater blocks of state fall under category of dark and critical. Water scarcity threatens food security for millions of people particularly in the arid and semi-arid regions. A major constraint to increase the food grain production in arid Rajasthan is limited surface water availability. Furthermore, the current irrigation systems in Rajasthan State are causing environmental problems of rising and declining groundwater levels, water logging and salinization. The Hanumangarh district, located in the north western part of Rajasthan State, represents the typical example of canal water misuse leading to rising groundwater levels, water logging and secondary salinization. These water management issues are very complex, and must be addressed by better planning and management. In order to improve water management and its productivity it needs to reveal the cause-effect relationships between hydrological variables such as evaporation, transpiration, percolation and biophysical variables such as dry matter and grain yields under different eco-hydrological conditions (Singh et al. 2006). Measurements of the required hydrological variables under field conditions are difficult, and need sophisticated instrumentation. Moreover, field experiments yielding site-specific information are very expensive, laborious and time consuming. However, suitable models like the CropSyst in combination with field experiments offer the opportunity to gain detailed insights into the system behaviour in space and time. Simulation models are an important tool to understand plant-soil interactions on water balance components and their effects on crop growth. They can assist field experimentation because direct measurement of all elements of the water balance (evaporation, transpiration, drainage, run off and profile water content change) is often not possible. Cropsyst has been applied to perform risk and economic analyses of scenarios involving different cropping systems, management options and soil and climatic conditions. Cropsyst (Stockle and Nelson, 1999) is a process-based model to simulate crop growth and water dynamics in the soil-plant atmosphere continuum. It has been widely used for cereals and other cropping systems (Stockle et al., 1994). The accuracy of these predictive models depends upon the proper identification of input parameters. As the information pertaining to water productivity of clusterbean and use of simulation models are non-existent for Indira Gandhi Nahar Pariyojana stage-I command area. Drawing on these insights, the study was planned to evaluate yield and water productivity of clusterbean at scheme level with objectives to quantify water balance and to calculate water productivity and economics of clusterbean.

MATERIALS AND METHODS
An experiment on farmers field was conducted during kharif 2012 at village Mainawali in Hanumangarh district of Rajasthan (074 o 20'34"E to 074 o 20'60" longitude and 28 o 37'62" N to 29 o 21'39" N latitude and 235 m above mean sea level). Soil physical (texture and bulk density) and chemical (pH, EC, CEC, ammonical-nitrogen and nitrate nitrogen) properties of experimental field were determined up to 1.0 m depth following the standard procedures. The sand, silt and clay contents were determined with Hydrometer method (Bouyoucos, 1962), bulk density with core method (Blake and Hartge, 1986), EC was measured with conductivity meter and pH with pH meter (Richards, 1954), OC by Wet digestion method (Walkley and Black, 1934). Ammonical nitrogen was determined by Nessler's method (Peech et. al., 1947) and nitrate nitrogen was determined by Phenoldisulphonic acid method (Harper, 1924 andPrince, 1945). The field capacity was determined in the field by covering the fully saturated soil surface with a polythene sheet and measuring the moisture content after 24-72 hours depending on soil type. In order to ascertain the physico-  (Stockle et al. 2003) was used to simulate yield and water productivity for clusterbean. The Cropsyst model was calibrated on yield of clusterbean using the observed phenological parameters (emergence, flowering, grain filling and physiological maturity) and harvest index of clusterbean from the experiment. The other parameters for the crop file were taken as default with slight adjustments. These adjustments were made within the range from the reported elsewhere (Jalota et al., 2006) so that the periodic crop growth like phenological stages, periodic biomass and final grain yield were matched with the experimentally observed values. The crop parameters used in the model are given in Table 2. During the first step simulated phenological stages (germination, flowering and physiological maturity) were matched with the observed by adjusting the degree days. The degree days were 165 for beginning of flowering, 200 for grain filling and 500 for physiological maturity, respectively.
Cropsyst is a multi-year, multi-crop, daily time step cropping systems simulation model developed to serve as an analytical tool to study the effect of climate, soils, and management on cropping systems productivity and the environment. Cropsyst simulates the soil water budget, crop phenology, canopy and root growth, biomass production, crop yield, residue production and decomposition, soil erosion by water, and salinity. These processes are affected by weather, soil characteristics, crop characteristics, and cropping system management options including crop rotation, cultivar selection, irrigation, nitrogen fertilization, soil and irrigation water salinity, tillage operations, and residue management. The development of CropSyst started in the early 1990s. The motivation for its development was based on the observation that there was a niche in the demand for cropping systems models, particularly those featuring crop rotation capabilities, which was not properly served. Efficient cooperation among researchers from several world locations, a free distribution policy, active cooperation of model developers and users in specific projects, and careful attention to software design from the onset allowed for rapid and cost-effective progress. Another important factor was the advantage of learning from a rich history of crop modelling efforts. Attention to a balance between the incorporation of sound science in the models and the utilization of adequate software design practices has been a trait of CropSyst since the beginning of its development. In this regard, it shares somewhat common objectives with APSIM (McCown et al., 1996, Keating et al., 2003, a modelling approach that has evolved to place substantial resources in the development of quality software engineering practices. CropSyst model will be applied to carry out the research study. The model has been developed to serve as an analytic tool to study the effect of cropping systems management on productivity and the environment.

RESULTS AND DISCUSSION
The various physical and chemical characteristics of the soil of the experimental site are given in Table 3. Model calibration was conducted following the procedure outlined by Hu et al., (2006). For calibration of clusterbean, data of the green area index (GAI), seed yield, above ground biomass (AGB) and N-uptake were used to determine the best crop model parameters. The simulated GAI, seed yield, above ground biomass and N-uptake were closer to the observed values of clusterbean during the season. The simulated GAI agreed well with field measurements from 20 DAS to maturity as shown in Table 4. The maximum GAI of 2.66 was observed at 60 DAS which was lower than simulated value (3.2). The observed and simulated GAI matched well with a RMSE of 0.55, correlation coefficient of 0.94 and Index of agreement of 0.95 observed for GAI of clusterbean. The seed yield of clusterbean was simulated with CropSyst model by inputting the observed data on duration of different phenol-phases during the experiment under field conditions. The simulate yield (1532 kg/ha) of clusterbean were closer to the observed yield of 1530 kg/ha as it is evident from the 7.8 % RRMSE (Table 5 and Fig 1). Simulations of early    (Fig 2). The drop in aboveground biomass of the clusterbean around late August was not properly captured by the model. As it was set for optimal conditions, CropSyst could not properly simulate the late season plant stress that impaired growth on these sites. Although clusterbean yield were simulated well and it did not respond to variation with correlation close to one. The reason for the moderate variation in yield was a very low annual variation in measured clusterbean yield. The simulated N-uptake (75 kg/ha) was closer to observed Nuptake (74 kg/ha) with 8.0 % RMSE (Fig 3). Correlation coefficient of 0.79 and Index of agreement of 0.81 observed for N-uptake of clusterbean. Increased uptake of N seems to be due to the fact that uptake of nutrient is a product of biomass accumulated by particular part and its nutrient content (Singh et al., 2011).
The total water applied in clusterbean was 405.8 mm out of this 326.7 mm may consume in ET. Thus, ET constituted 80.5 % of total water applied and deep drainage constituted 13.1 % and rest 6.4 % stored as residual soil moisture. Results showed that 1/5 th of total water applied   (Table 6) with water productivity of 0.38 kg m -3 . The seasonal water loss (Soil water evaporation + transpiration + drainage below root zone) matched reasonably well the measured values (Irrigation + rainfall) for clusterbean. Measured water loss ranged from 800 to 1000 mm for cotton (Aujla et al., 1991) and 400 to 450 mm for wheat (Arora et al., 1997). A close relationship between simulated and measured water loss values under different crops suggest that the simulation of water balance components were realistic with the model and can be used for assessing water loss components in cropping systems including the intervening bare period. It is significant to note that there was net depletion of soil water storage in long duration crops like cotton and wheat. These results show trends and magnitudes of soil water depletion similar to field observations (Jalota et al., 1985).