Improve State-Level Wheat Yield Forecasts in Kazakhstan on GEOGLAM's EO Data by Leveraging A Simple Spatial-Aware Technique

TL;DR

This paper introduces a simple spatial-aware technique called state-wise additive bias to improve wheat yield forecasts in Kazakhstan, significantly reducing prediction errors by addressing regional heterogeneity in remote sensing data.

Contribution

The study proposes and validates a novel, simple bias correction method that enhances machine learning yield predictions by explicitly accounting for spatial heterogeneity.

Findings

Reduces overall RMSE by 8.9%

Decreases maximum state-wise RMSE by 28.37%

Highlights importance of spatial heterogeneity in geospatial forecasting

Abstract

Accurate yield forecasting is essential for making informed policies and long-term decisions for food security. Earth Observation (EO) data and machine learning algorithms play a key role in providing a comprehensive and timely view of crop conditions from field to national scales. However, machine learning algorithms' prediction accuracy is often harmed by spatial heterogeneity caused by exogenous factors not reflected in remote sensing data, such as differences in crop management strategies. In this paper, we propose and investigate a simple technique called state-wise additive bias to explicitly address the cross-region yield heterogeneity in Kazakhstan. Compared to baseline machine learning models (Random Forest, CatBoost, XGBoost), our method reduces the overall RMSE by 8.9\% and the highest state-wise RMSE by 28.37\%. The effectiveness of state-wise additive bias indicates machine learning's performance can be significantly improved by explicitly addressing the spatial heterogeneity, motivating future work on spatial-aware machine learning algorithms for yield forecasts as well as for general geospatial forecasting problems.