The anatomy of drought in Italy: statistical signatures, spatiotemporal persistence, and forecasting potential

Drought is a multifaceted hazard with profound socio-environmental consequences in the Mediterranean, where Italy exemplifies a climate vulnerability hotspot shaped by pronounced spatial heterogeneity and intensifying climatic pressures. This study advances drought research by conducting a comprehensive analysis of six-month Standardized Precipitation–Evapotranspiration Index (SPEI-6) time series across Italy, integrating higher-order statistical descriptors, persistence diagnostics based on the Hurst exponent (H) and Detrended Fluctuation Analysis (DFA), advanced clustering algorithms, and deep learning forecasting. Distinct from conventional mean–variance assessments, the analysis emphasizes skewness and other higher-order moments to capture asymmetries in drought intensity and frequency, and employs scaling metrics to quantify long-range dependence and memory in hydroclimatic signals. A comparative suite of clustering approaches, including K-means, Agglomerative Hierarchical, Gaussian Mixture Models, and Spectral Clustering, delineates a coherent tripartite drought structure: a persistent southern and insular regime with strong temporal memory and prolonged droughts, an intermediate northeastern corridor with moderate persistence, and a volatile northwestern Alpine domain characterized by weak persistence, high variability, and abrupt transitions. Forecasting experiments employing Kolmogorov–Arnold Fourier (KAF) networks, benchmarked against Long Short-Term Memory (LSTM) architectures, reveal substantial skill at one-month lead, particularly in persistent southern and insular regions, while performance declines at seasonal horizons and in highly variable northern areas. These findings highlight the necessity of regionally tailored monitoring and adaptive management strategies. The methodological framework presented here, modular and transferable, provides a rigorous and replicable template for drought diagnosis and early warning in Mediterranean and other drought-prone regions facing escalating climate variability.