Maize (Zea mays L.) productivity as influenced by sowing date and nitrogen fertiliser rate at Melkassa, Ethiopia: parameterisation and evaluation of APSIM-Maize
Mesfin, T and Moeller, C and Parsons, D and Meinke, H, Maize (Zea mays L.) productivity as influenced by sowing date and nitrogen fertiliser rate at Melkassa, Ethiopia: parameterisation and evaluation of APSIM-Maize, Proceedings of the17th Australian Society of Agronomy Conference, 20-24 September 2015, Hobart, Australia, pp. 1-4. (2015) [Refereed Conference Paper]
Crop modelling can assist in exploring the production risks and the yield uncertainty associated with rainfall variability but requires empirical data suitable for model testing within a specific system and environment. To parameterise the crop simulation model APSIM-Maize, a field experiment examining a medium maturing maize cultivar sown on two dates and grown at two nitrogen (N) fertiliser rates (0 and 100 kg N ha-1) was conducted at Melkassa, Ethiopia. Model performance was evaluated against six independent datasets from the same site. APSIM-Maize simulated crop phenology, leaf area index, and biomass well for both sowing dates. The model showed acceptable performance in simulating grain yield in most cases. However, at the late sowing date and high N supply, the model over-estimated yield by 37%. The model realistically captured variation in soil water dynamics as indicated by a RMSE of 0.039 mm mm-1. Evaluation of the parametrised model against independent data showed that it was able to reasonably simulate various crop responses including date of silking (RMSE=1 d), date of physiological maturity (RMSE=1.50 d), grain yield (RMSE=0.39 t ha-1), and biomass production (RMSE=0.48 t ha-1) for maize grown at different sowing dates (between June and July) and N application rates (up to100 kg N ha-1).The results showed that APSIM-Maize is credible and can be used for scenario analyses of maize systems in semiarid environments of Ethiopia.
Refereed Conference Paper
crop simulation modelling, Ethiopia, maize, model evaluation, model parameterisation