Simulating potential yield of rainfed soybean in northeast Nigeria.

We used the CROPGRO-Soybean model to simulate the production potential of rainfed soybean in northeast Nigeria. Data from ten soybean experiments conducted under optimal conditions in 2016-2018 at Kano and Dambatta in the Sudan savanna (SS) agroecological zone were used to determine the cultivar coefficients and calibrate the model for the varieties TGX 1448-2E and TGX1951-3 F. The model was evaluated with data from four phosphorous response trials conducted at Zaria and Doguwa in the northern Guinea savanna (GS) of Nigeria between 2016 and 2018. Results show that the CROPGRO-Soybean model was able to accurately simulate soybean growth and grain yield with low RMSE and high d-index values. Consequently, the model was used to investigate the rainfed yield potential of the two varieties in 24 sites in northeast Nigeria under different sowing windows using 30-year (1985-2014) weather data. The result shows that soybean can be grown in northeast Nigeria, but yield performance is dependent on location, variety and sowing window. The simulated yield was higher in the SS than in the GS agro-ecozone despite the longer growing period in the later. Low yield was simulated for TGX 1448-2E for most of the sites. The yield of TGX1951-3 F was above a threshold of 1500 kg ha