How we used APSIM to simulate conservation agriculture practices in the rice-wheat system of the Eastern Gangetic Plains

Share this to :

Examples of how to simulate performance of conservation agriculture (CA) and conventional tillage (CT) practices using cropping systems models are rare in the literature, and from the Eastern Gangetic Plains (EGP). Here we report a comprehensive evaluation of the capacity of APSIM for simulating the performance of CA and CT cropping practices under a diverse range of tillage (CT vs zero tillage (ZT)), crop establishment options (puddled transplanted rice vs unpuddled transplanted rice), residue, N rates, and irrigation practices from two sites in the EGP that differed in soil type, water table dynamics, and agro-climatic conditions. We followed a robust procedure of model parameterisation, calibration, and validation, then undertook statistical analyses to evaluate model performance. We have demonstrated that when different values for key model input parameters are employed (i.e. change in soil properties (Ks, BD)), crop rooting parameters (xf- root hospitality, kl- root extraction efficiency) and soil microorganism activity (Fbiom- fraction of soil organic matter present as microbial biomass and Finert- the inert fraction of soil organic matter), the model performed well in simulating the different performances of CA and CT management practices across the environments in the EGP. Model performance was markedly better in the full-N than in zero-N, but both are still considered acceptable. In addition to well-watered and fertilised treatments, the model was able to capture an observed crop failure in rainfed unpuddled transplanted rice accurately, illustrating an ability to capture crop response under a wide range of water stress environments. As demonstrated by robust statistical criteria, APSIM was able to capture the effect of cropping system, irrigation, tillage, residue, and N-application rate within the bounds of experimental uncertainty, hence is now deemed a suitable tool for scenario analyses around the relevant practices.

Share this to :