Integrating APSIM model with machine learning to predict wheat yield spatial distribution

Abstract

Traditional simulation models are often point based, thus more research is needed to emphasize the spatial simulation, providing decision makers with fast recommendations. Combining machine learning algorithms with spatial process‐based models could be considered an appropriate solution. We created a spatial model in R (APSIMx_R) to generate fine resolution data from coarse resolution data, which is typically available at the regional level. The APSIM crop model outputs were then deployed to train and test the Artificial Neural Network (ANN), creating a hybrid modelling approach for robust spatial simulations. The APSIMx_R package facilitates preparing the required model inputs, executes the prediction, processes, and analyzes the APSIM crop model outputs. This note demonstrates the use of a new approach for creating reproducible crop modelling workflows with the spatial APSIM next generation model and machine learning algorithms. The tool was deployed for spatial and temporal simulation of potential wheat yield under different nitrogen rates, and various wheat cultivars. The spatial APSIMx_R was validated by comparing the simulated yield at 100 kg Nha−1 to the analogues' actual yield at the same grid points, which showed good agreement (d = 0.89), between spatially predicted and actual yield. The hybrid approach increased such precision, resulting in higher agreement (d = 0.95) with actual yield. When the interaction between cultivars and nitrogen levels was considered, it was found that the novel cultivar Sakha95 is nitrogen voracious, exhibiting a larger drop in yield (65%) under minimal nitrogen treatment (0 kg N ha−1) relative to the potential yield.This article is protected by copyright. All rights reserved

Developing automated machine learning approach for fast and robust crop yield prediction using a fusion of remote sensing, soil, and weather dataset

“…Integrating both approaches can leverage their complementary strengths to enhance yield predictions and better understand the impacts of environmental variability and climate change on agriculture. Crop models and machine learning can be integrated [83] by employing crop model inputs and outputs as predictors in ML, together with other external variables that crop models cannot include, such as salinity, pests, diseases, and terrain, to predict yield at spatial explicit. Therefore, future directions could be done by coupling the developed approach (H 2 OautoML) with crop models to not only predict the current spatial yield, but also to explore future impacts, adaptation, and mitigation to climate change.…”

The Fusion Impact of Compost, Biochar, and Polymer on Sandy Soil Properties and Bean Productivity

“…Furthermore, the effect of these materials on soil microbial activity and greenhouse gas emissions should be considered as a future research direction. On the other hand, including these resources into decision support tools such as crop models [48] will enable them to be used at scale in a cost-effective manner.…”