Simulation of productivity and soil moisture under Marandu palisade grass using the CSM-CROPGRO-Perennial Forage model

Abstract

Crop models are important tools for assisting farmers and crop consultants to make decisions about fertilisation, irrigation and harvest management, because they allow users to understand productivity from the view of integrated sensitivities of basic plant physiological processes. The first objective of this study was to evaluate the performance of the CSM-CROPGRO-Perennial Forage model (PFM) to simulate regrowth of Urochloa brizantha (Hochst. ex A. Rich.) R.D. Webster cv. Marandu under varying irrigation and nitrogen levels. The second objective was to evaluate the water-balance module of the model under soil and climatic conditions in the Cerrado biome of central-eastern Brazil. The experimental data for model evaluation were obtained from a field experiment conducted during 2015, 2016 and 2017, and included herbage production, plant-part composition and plant nitrogen (N) concentration. The results suggest that the model can be used to simulate growth of Marandu palisade grass adequately under different managements of irrigation and N fertilisation. The findings indicate also that the agreement between simulations and field-observed soil moisture shows good performance of the water-balance module of CSM-CROPGRO-PFM. The most important parameterisation required by the model was the determination and calibration of inputs such as the stable soil carbon pool(SOM3) for N mineralisation, which affected the N response, and the soil water-holding characteristics, which affected the irrigation response. The default parameterisation ( species, ecotype, cultivar) of cv. Marandu inCSM-CROPGRO-PFM was sufficient for adequate performance of the model for this new environment and new crop management. However, minor modifications of species parameters were helpful to account for winter-kill of foliage.