Spatio-temporal dynamics of migrating wet and dry extremes: structural coherence and land-use associations across India

Abstract Global hydroclimatic extremes increasingly occur as cascading wet–dry sequences, challenging conventional single-hazard management. However, their spatiotemporal migration and associations with land-use patterns remain poorly understood. We present an integrated framework combining the standardized precipitation evapotranspiration index, a spatio-temporal density-based spatial clustering of applications with noise algorithm, and phase-wise land-use/land-cover (LULC) exposure analysis to characterize the migration, structural coherence, and persistence of major wet and dry extremes across India. Both dry and wet events exhibit a range of migration directions rather than a single dominant pathway, spanning northward, southward, and eastward trajectories. Although several recent events are among the least coherent, a linear-trend test does not establish a temporal shift towards fragmentation ( p = 0.74 for dry and p = 0.16 for wet events). A pronounced asymmetry between rapid onset and a prolonged recovery phase (2–3 months) highlights the persistence of hydroclimatic extremes. Phase-wise LULC analysis indicates that agricultural landscapes show the greatest absolute exposure, largely reflecting their extensive national footprint, while urban areas are frequently associated with recovery-stage wet events. The proposed framework provides a transferable approach for improving early-warning systems and supporting land-use-informed adaptation to compound hydroclimatic hazards.