Integrating Vegetation Health and Precipitation Anomalies for Satellite‐Based Agricultural Drought Monitoring Using Google Earth Engine

ABSTRACT In this paper, a drought monitoring system for Siping City, Jilin Province, is designed using multi‐source remote sensing data from 2010 to 2023, integrating vegetation health and precipitation anomalies. We used Google Earth Engine to process and analyze three satellite‐based datasets: the Moderate Resolution Imaging Spectroradiometer (MODIS) (MOD13Q1) to assess vegetation greenness. The MODIS land surface temperature (MOD11A2) is used to assess thermal stress, and the CHIRPS daily precipitation is used to assess rainfall. We constructed a modified Vegetation Health Index (VHI) composed of an index of 70% Vegetation Condition Index (VCI) and an index of 30% Temperature Condition Index (TCI), especially suited to the agro‐ecology of Northeast China (NDVI ranging between 0.1 and 0.9, and land surface temperature between 15°C and 35°C). The summer VHI for July–August was combined with precipitation deviations from May–September, using a 500‐mm baseline to develop a drought index. Mapping the index each year, classifying drought severity‐ Normal 0–30, Mild 30–50, Moderate 50–70, Severe 70–85, Extreme > 85 examined trends and calculated area‐based statistics. For 14 years, drought has been replaced by evident year‐by‐year fluctuations in Siping. Between 2010 and 2023, the average index varied across space, with stronger drought signals in the central and western farming regions. Three severe drought years, namely 2014, 2018, and 2022, exhibit DI values far above 70, matching the recorded agricultural droughts in Northeast China. In 2022, the drought was so severe that the entire area fell within the drought zone, and 42.7% of the study region experienced severe‐to‐extreme droughts (DI > 70). From the time series, the drought severity trends increase slightly but not significantly ( R 2 = 0.314, p > 0.05) during the entire period.