Assessing the performance of multi-timescale drought indices for monitoring agricultural drought impacts on wheat yield

Crop yields are increasingly threatened by intensifying droughts in southern Europe, yet the long-term, spatially explicit quantification of yield response to agricultural drought remains limited. Remote sensing can address this gap by providing continuous spatiotemporal estimates of crop water stress. This study quantified the response of wheat yield to agricultural drought from 2003 to 2021 across four autonomous communities in Spain—La Rioja, Castilla y León, Castilla-La Mancha, and Andalucía—using three drought indicators, including a meteorological drought index, the Standardized Precipitation-Evapotranspiration Index (SPEI), and two remote sensing-based indices, the Standardized Precipitation-Actual Evapotranspiration Index (SPET) and the Standardized Evapotranspiration Deficit Index (SEDI), derived from a physical model that estimates actual crop evapotranspiration (ET c act ). Drought indices were aggregated at timescales from 1 to 12 months to identify the accumulation of timescales most relevant to wheat yield variability in each region. Results indicated that correlations varied spatially, with the strongest wheat yield–drought correlation in La Rioja (r = 0.79 for SPEI, 0.62 for SPET, and 0.81 for SEDI) and the weakest in Andalucía (r ≈ 0.33–0.35). Mediterranean regions (Andalucía and Castilla-La Mancha) showed the strongest correlation at short timescales (1–3 month) during late spring, while temperate continental regions (Castilla y León and La Rioja) responded to longer timescales (3–6 month) in early summer. Among indices, SEDI exhibited the strongest and most consistent correlation with wheat yield variability. These results highlight the value of integrating remotely sensed ET c act with ERA5 reanalysis for region-specific drought monitoring, offering significant potential for advancing operational agricultural water management strategies under increasing drought frequency and climate change. • Remote sensing-derived evapotranspiration improves agricultural drought monitoring. • Integrating actual and potential evapotranspiration enhances drought assessments. • Two decades of remote sensing-based drought indicators capture yield variability. • Region-specific drought monitoring frameworks enhance decision-making. • Remote sensing-driven drought indices can strengthen crop yield forecasting.