Simulation of yield and water productivity of Cowpea cultivars under deficit irrigation conditions using the DSSAT model

Introduction
Deficit irrigation is a solution for less water use with the aim of maximum use of the unit volume of water use and water storage and saving for the development of agriculture or the development of other consumption sections. Although the direct result of the lack of irrigation is the reduction of yield at the unit level, the reduction of production costs and the optimization of net profit will compensate for the reduction of yield. In order to adapt and deal with the limitation of water, it is necessary to use mechanisms to increase the efficiency of water use and water resources. Deficit irrigation is used as a technical and economic method in irrigation to improve the relationship between drinking water and the functioning of agricultural plants, and also as one of the productivity solutions in conditions of water shortage. This method requires consistent, accurate, and efficient management, which is different from traditional irrigation management. Another to solve this problem is the use of crop models that simulate crop yield under different soil and climate conditions. The great advantage of these models is the low cost and the short time spent to obtain results, besides helping better agricultural planning and management towards higher profits. The DSSAT model is a computational system that includes a database management system, crop models, and application programs. Plant simulation models can be useful for predicting crop yield and investigating the effect of drought stress on plant growth and development. The DSSAT model is able to evaluate the impact of environmental factors such as weather, soil properties, and field management decisions. The data and information needed to run the model include spatial location (longitude and latitude, altitude above sea level, average annual temperature) and meteorological information (daily minimum and maximum temperature, solar radiation, and radiation), soil science (such as soil texture, soil structure and depth of each layer, apparent specific weight, nutrients, wilting point, and electrical conductivity), and agricultural operations (type of number and Its type is planting date, planting depth, line spacing, planting density, irrigation dates, amount of irrigation water). In this research, the DSSAT model was used to simulate seed yield, pod, biomass, soil water balance components, and water consumption efficiency in cowpea cultivars.
 
Materials and Methods
This experiment was conducted for two consecutive cropping seasons in 2018 and 2019 in Guilan province and in Astaneh-Ashrafiyeh city with an average height of -5 meters at sea level. The amount of rainfall during the growing season of cowpea in the first and second years was 93 and 94 mm, respectively. This research was done in the form of split plots and in the form of a randomized complete block design with three replications, and the main factor in it includes irrigation at three levels: 100% of water requirement (I₁), 75% of water requirement (I₂) and 50% of water requirement (I₃), and the secondary treatment included three cultivars of cowpea, Kamran cultivar (C₁), Khuzestan local variety (C₂) and Dehsari local cultivar (C₃). The amount of water supplied to each experimental unit was measured by a counter. The amount of water used during the plant growth period included the total amount of irrigation water and the amount of rainfall.
 
 
Results and Discussion
The results showed that the average relative error between the observed and simulated values in 2018 and 2019 for biomass yields were -0.88 and -0.89%, respectively. With full irrigation and providing 100% of water needs, the amount of biomass, seed, and pod yield in cowpea cultivars increased and the model was able to simulate the process of yield changes in different water needs. The relative error (MRE) between the observed and simulated values in biomass yields in 2017 were between -1.14 and -0.55 percent and in 2018 between -1.19 and It was -0.55 percent. The root mean square error in estimating the productivity of water use based on biomass yield based on water use, for Kamran, Khuzestan, and Dehsari cultivars in 2017 respectively 0.0106, 0.01078, and 0.01087 kg.m^-3 and per the year 2018 were estimated as 0.01044, 0.01079, and 0.01091 kg.m^-3 respectively. Examining the values of simulation and observation on biomass, seed, and pod yield showed that root means square error and average relative error rate and other statistical tests were within the acceptable range and the DSSAT model was able to reproduce the yield of cowpea cultivars in simulate well the conditions of deficit irrigation. The average relative error between observed and simulated values in water productivity based on water use in biomass, seed, and pod yields in 2018 were -0.95, 0.29, and -0.47 % respectively, and in 2019 it was -0.67, 0.001, and -0.40 % respectively. The mentioned values in water productivity based on transpiration and also based on evaporation and transpiration in the yield of biomass, seeds, and pods were -0.88, 0.08, and -0.45% respectively in 2018 and in 2019, were -0.89, 0.07, and -0.45%, which indicates the appropriate evaluation of the model in simulation and observation conditions.
 
Conclusion
In the field experiment, with the increase in the amount of irrigation water, the seed and biomass yields in cowpea cultivars increased and the DSSAT model was able to change these traits at different levels of irrigation, in accordance with the results observed in the simulation field. However, with the increase of drought stress, the percentage of simulated relative error was higher in stressed conditions than in irrigated conditions. The values of RMSE and RMSE_n statistics for each function and water productivity parameters showed that the DSSAT model has an acceptable error. The simulation of the yields of seed, pod, and biomass under the influence of deficit irrigation was acceptable with the mentioned model and it is able to be an efficient tool to support decisions and improve research in management. Water use in cowpea should be recommended for the study area.