Rajeev RanjanNegi, Ankita2019-08-102019-08-102019-07http://krishikosh.egranth.ac.in/handle/1/5810120805The present investigation was carried out at E4 plot of Norman E. Borlaug Crop Research Centre of G.B. Pant University of Agriculture and Technology Pantnagar, US Nagar district of Uttarakhand. The main aim of the study was to develop best management strategies under varied climatic conditions using CROPGRO simulation model for soybean variety PS 1347. The experiment was conducted in two factor randomized complete block design with two treatments and three replications under three dates of sowing i.e. June 29, July 9 and July 19 during kharif season of 2018. The plant growth and development parameters were noted throughout the season. The change in soil moisture percent and leaf area index was measured at every ten days interval starting from sowing to harvesting. The plant biomass was calculated at twenty days of interval after sowing of crop. The phonological stages of soybean were recorded for all the three dates of sowing. The calibration and validation of CROPGRO model was done after developing genotypic coefficient for soybean variety PS 1347. The different dates of sowing, tillage, irrigation and fertilizer treatments was simulated using CROPGRO simulation model and compared with observed dataset to select the best management practices for soybean variety PS 1347. The study of effect of weather parameters on soybean productivity revealed that bright sunshine hour of July first week was the most important parameter to affect soybean productivity followed by minimum temperature of October fourth week and maximum temperature of November first week. A total of three models were developed using 10 years (2007-2016) average weekly weather parameters with district level yield using SPSS software. The first model (CD=0.71) only used average bright sunshine hours of 1st week of July. The observed yield ranged between 0.92 q ha-1 and 1.93 q ha-1 and yield predicted by model-3 varied between 0.92 q ha-1 and 1.92 q ha-1 with RMSE value of 2.6%. It was found that bright sunshine hour of July 1st week shows significant positive relationship with soybean yield. Minimum temperature of 4th week of October has more significant effects (R2=0.66) than maximum temperature of 1st week of November (R2=0.14). Observed phenophases of soybean variety PS 1347 was compared with the model simulated value. The observed yield ranged from 5124 kg/ha (D3F2) to 6543 kg/ha (D1F1) and yield predicted by CROPGRO model ranged from 5035 kg/ha (D3F2) to 6564 kg/ha (D1F1), respectively for the year 2018. The observed grain yield was found close to the CROPGRO simulated yield for both the years 2018 (RMSE=7.2%) and 2017 (RMSE=10.9%). The effect of climate change on soybean yield was analyzed using the MarkSim weather generator for the year 2020, 2030, 2050 and 2080. Soybean yield showed decreasing yield trend from 2018 to 2080 with increase in temperature. The simulation results of model confirmed that soybean will produce highest yield when crop is sown on 20th June after two plowing followed by two harrowing under 90 mm irrigation and fertilizer dose (N:P:K:S) @ 25:60:40:20.ennullThesis