GIS-Based Climate Forecasting Models for Agricultural Decision-Making in Coastal Mozambique: Yield Prediction Accuracy Scores

This scoping review examines GIS-based climate forecasting models applied to agricultural decision-making in coastal Mozambique, focusing on yield prediction accuracy scores. A systematic review approach was employed, including a comprehensive search of academic databases such as Google Scholar, Web of Science, and Scopus. The review focused on peer-reviewed articles published between and the present, with an emphasis on studies that utilised GIS technologies for climate forecasting in Mozambique. The analysis revealed that while most models showed high predictive accuracy (r² > 0.7), there was significant variability among different regions of coastal Mozambique, with some areas showing yield prediction accuracies as low as r² = 0.45. GIS-based climate forecasting models can significantly enhance agricultural decision-making in coastal Mozambique by improving the precision of crop yield predictions. Further research should focus on integrating multiple data sources and developing more robust statistical models to improve overall prediction accuracy across all regions. Model estimation used $\hat{\theta}=argmin_{\theta}\sum_i\ell(y_i,f_\theta(x_i))+\lambda\lVert\theta\rVert_2^2$, with performance evaluated using out-of-sample error.