DROUGHT SEVERITY AND MAJOR CROP YIELDS IN SINDH, PAKISTAN: EVIDENCE FROM A MULTI-INDEX PANEL DATA ANALYSIS

Abstract

Drought increasingly constrains agricultural productivity in arid and semi-arid regions, particularly in Sindh, Pakistan, where crop production relies heavily on irrigation and variable climatic conditions. This research investigates the impact of drought on major cash crops in Sindh using a district-level multi-index panel data approach. Annual data on cotton, sugarcane, and rice production and cultivated area are combined with climatic and land-surface information to construct crop-yield measures and multiple indicators of drought severity. Meteorological drought is quantified using the Standardized Precipitation Index (SPI), climatic water-balance drought is assessed through the Standardized Precipitation Evapotranspiration Index (SPEI), and agricultural drought is measured using a standardized soil-moisture index. Crop-specific drought exposure is determined over agronomically relevant growing periods. Two-way fixed-effects panel regression models estimate within-district effects of drought while controlling for time-invariant district characteristics and common annual shocks. Additional model specifications evaluate nonlinear drought severity, crop-specific responses, lagged effects, and robustness across alternative drought time scales and error structures. The multi-index framework facilitates comparison of precipitation deficits, temperature-driven atmospheric water stress, and soil-moisture shortages, offering a comprehensive assessment of drought vulnerability and informing evidence-based irrigation, crop management, and climate adaptation strategies for Sindh’s agricultural sector.