Meta-learning-driven intelligent ensemble approach for robust drought evaluation across China

Abstract

Under global warming, droughts across China exhibit pronounced spatiotemporal heterogeneity, posing severe threats to agricultural production, water security, and ecosystem stability. To address the limitations of conventional drought monitoring methods, which include sparse station coverage, poor spatial continuity, and inadequate representation of nonlinear interactions, this study develops a Comprehensive Drought Monitoring Model based on a Meta-learning Ensemble Algorithm (CDMM_MLEA). The model integrates multi-source remote sensing data, including canopy temperature, vegetation indices, soil moisture, and canopy water content, with meteorological and auxiliary geospatial data. Using an ensemble learning framework, CDMM_MLEA significantly improves the accuracy and robustness of drought monitoring across China from 2001 to 2023. The findings suggest that (1) CDMM_MLEA outperforms five benchmark machine learning models across China, achieving the highest correlation with the Standardized Precipitation Evapotranspiration Index (SPEI) and lowest error, particularly in the Songliao River Basin (SLRB) and Yellow River Basin (YRB); (2) Compared to SPEI, it more effectively captures the spatial propagation of drought, including its boundary shifts, intensity gradients, and temporal persistence, even maintaining errors below 10 % in regions with sparse station data; (3) Analysis of drought evolution from 2001 to 2023 reveals a phased pattern: high-frequency, high-intensity, and long-duration droughts dominated during 2001–2010, followed by a mitigation phase after 2011, with intensity decreasing by 25 % and duration shortening by 0.5–1.3 months per event. Autumn droughts were most severe, affecting 62.3 % of the Loess Plateau and western SLRB. CDMM_MLEA provides a reliable tool for high-resolution spatiotemporal drought assessment, supporting operational early warning systems, optimized water resource allocation (e.g., South-to-North Water Diversion Project), and region-specific drought adaptation strategies (e.g., northwest water-saving irrigation) across China.