A Machine Learning Approach to Forecasting Remotely Sensed Vegetation Health

TL;DR

This paper presents an open-source machine learning tool that forecasts vegetation health at high resolution using MODIS satellite data, aiding drought management worldwide.

Contribution

It introduces a novel automated pipeline combining MODIS data processing with gradient-boosted models for high-resolution vegetation forecasting.

Findings

The tool outperforms baseline models in predictive accuracy.

It is effective across diverse ecological regions.

High-resolution forecasts are achievable with global satellite data.

Abstract

Drought threatens food and water security around the world, and this threat is likely to become more severe under climate change. High resolution predictive information can help farmers, water managers, and others to manage the effects of drought. We have created an open source tool to produce short-term forecasts of vegetation health at high spatial resolution, using data that are global in coverage. The tool automates downloading and processing Moderate Resolution Imaging Spectroradiometer (MODIS) datasets, and training gradient-boosted machine models on hundreds of millions of observations to predict future values of the Enhanced Vegetation Index. We compared the predictive power of different sets of variables (raw spectral MODIS data and Level-3 MODIS products) in two regions with distinct agro-ecological systems, climates, and cloud coverage: Sri Lanka and California. Our tool provides considerably greater predictive power on held-out datasets than simpler baseline models.