Development of Minimum Soil and Plant Data Set for DSSAT Crop Simulation Model for Pigeonpea Cultivars under Varied Dates of Sowing

摘要

Traditional agronomic experiments conducted at particular points in time and space are site, season specific and time consuming. To overcome this, many computer based software have been developed called as Crop Simulation Models. Among the numerous crop growth models, the most widely used is the Decision Support for Agro Technology Transfer (DSSAT) model, which is designed to simulate growth, development, and yield of a crop along with changes in soil water, carbon and nitrogen under the system overtime. An attempt was made to developminimum data set for two pigeonpea cultivars (BRG-1 and BRG-2) from the observations recorded by conducting an experiment at UAS, GKVK, Bengaluru during 2017-18 and 2018-19. The results of 2018 revealed that the model underestimated the yields under thecrop sown during 30thMay (-108.4%) and overestimated the yields under the crop sown during 10 and 24th August (54.2% and 15.1%, respectively). Among the two varieties, model overestimated the yields for BRG-1 (3.7%) compared to BRG-2 (-29.8%). During 2018-19, model underestimated the yield for 10* May and 1 s,June sown crop (-41.0% and -74.9%, respectively) as compared crop sown on 12,h July (250.7%). Among the two varieties, BRG-2 recorded lower error (1.6%) as compared to BRG-1 (88.3%). Statistical evaluation of experimental yield using mean error (ME), root mean square error (RMSE), coefficient of residual mass (CRM) and modelling efficiency (EF) revealed that, simulation of BRG-1 grain yield was in good agreement with the observed values with comparatively low ME (3.6 and 46.0 during 2017-18 and 2018-19, respectively) indicating the variety is calibrated well among the two varieties. Minimum average RMSE (3511.2) values were recorded indicating less deviation of simulated values from observed values. Positive CRM (0.048) values were recorded indicating underestimation of yields by the model, requiring some more calibration by field experimentation.