Four-decade spring droughts in Taiwan

Abstract

Study region: Taiwan Study focus: The spring drought in Taiwan not only impacts local socio-economic conditions but also has global implications due to the region's critical role in semiconductor manufacturing. Despite extensive research, local studies have yet fully integrated multiple drought indices for the comprehensive assessment of meteorological, hydrological, and agricultural droughts. This study addresses the gap by utilizing Standardized Drought Indices (SDI), the Standardized Temperature Index (STI), and the Normalized Difference Water Index (NDWI) to explore the characteristics and potential climate change links of spring droughts from 1982 to 2021 in Taiwan. It employed methodologies including remote sensing, statistical analysis, and machine learning, complemented by ground-validated remote sensing data assimilation. New hydrological insights for the region: This study reveals that Taiwan presents a unique case in drought dynamics, deviating from typical global patterns. In Taiwan, hydrological droughts respond more quickly to rainfall deficits than agricultural droughts, showing strong correlations (r > 0.8) for hydrological and moderate correlations (0.4 < r < 0.6) for agricultural droughts. Furthermore, although agricultural droughts have decreased, hydrometeorological droughts have shown an increase since the early 2000s, marked by an abrupt change point. Notably, the central and southern regions are the hotspots for droughts, with significant temperature-drought coherence occurring in 4–6-year resonant cycles. These findings suggest the need for enhancing drought management, particularly crucial for Taiwan's industrial economy.

Keywords

• Spring drought
• Climate change
• Remote sensing
• Wavelet coherency
• Standardized drought indices