강수량 기반 가뭄지수의 구조적 차이에 대한 비선형 동역학 및 시차 분석: 지역 민감도 분류를 중심으로

초록

Global climate change is increasing the frequency, intensity, and duration of droughts, thus posing a significant and dynamic threat to both natural and socioeconomic systems. This study applied dynamic system indicators to the Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) in order to analyze their structural features, time lags, event responses, and regional sensitivity. Interrelationship analysis of the SPI-SPEI, focusing on dynamic time lags and nonlinear dependencies, revealed complex temperature-influenced drought patterns. The performed integrated sensitivity analysis identified four spatial clusters from regional temperature-related sensitivity differences. These clusters exhibited characteristics such as being structurally complex and sensitive to short-term event changes; possessing a relatively stable structure yet showing high responsiveness to event intensity; and displaying complexity in time-lag patterns along with strong non-linear dynamic characteristics. Consequently, the SPEI showed a significantly higher nonlinear sensitivity and complexity than the SPI. For 3-month period, the SPEI exhibited statistically significant higher values than the SPI for key metrics. Its maximum Lyapunov exponent averaged 0.0217 greater (p-value ≈ 1.66 × 10-33), and its average mutual information was 0.0184 greater (p-value ≈ 4.32 × 10-12). The SPI-SPEI relationship also showed asymmetric and dynamically varying time-lag structures. These findings highlight the need for the complementary use of SPI and SPEI for comprehensive drought understanding and developing customized water resource management considering regional sensitivity and dynamic time-lag variability.

키워드