Kernel Ridge Regression Hybrid Method for Wheat Yield Prediction with Satellite-Derived Predictors

Wheat dominates the Australian grain production market and accounts for 10–15% of world’s 100 million tonnes annual global wheat trade. Accurate yield prediction is critical to satisfying local consumption increasing exports regionally globally meet human food security. This paper incorporates remote satellite-based information in a wheat-growing region South Australia estimate by integrating kernel ridge regression (KRR) method coupled with complete ensemble empirical mode decomposition adaptive noise (CEEMDAN) grey wolf optimisation (GWO). The hybrid model, ‘GWO-CEEMDAN-KRR,’ employing an initial pool 23 different predictors, seen outperform all benchmark models feature selection (ant colony, atom search, particle swarm optimisation) methods that are implemented using set carefully screened satellite variables or CEEMDAN approach. A suite statistical metrics infographics comparing predicted measured shows model error can be reduced ~20% proposed GWO-CEEMDAN-KRR model. With verifying accuracy simulations, we also show it possible optimise achieve agricultural profits quantifying including effects on potential yield. further improvements methodology, adopted simulation requires sensing data establish relationships between crop health, yield, other productivity features support precision agriculture.