Linking drought to agricultural water use across contrasting cropping systems in an arid environment
Year: 2026
Type: article
Abstract: Understanding agricultural water use responses to climatic drought is essential for improving water management in drylands. This study investigated multi-timescale drought variability using the Standardized Precipitation–Evapotranspiration Index (SPEI) and its relationship with WaPOR-derived actual evapotranspiration and interception (AETI), representing consumptive water use, across the Pishkooh Plain, an arid region in Iran, for 2010–2024. AETI was validated against soil moisture depletion measurements in 2025, showing strong agreement (r = 0.97; mean bias 0.92, MAE 1.47, RMSE 1.67 mm dekad⁻¹) with no significant differences. SPEI–AETI Pearson correlations were calculated for 18 agricultural zones and grouped using K-means and Ward’s hierarchical clustering. K-means, as the better-performing approach, identified four spatial clusters and three temporal regimes, representing statistical groupings that do not necessarily reflect agricultural processes. The zonal clusters differed in geographic distribution, proximity to irrigation sources, and cropping/land-use patterns, ranging from a compact cluster with persistently negative correlations (mean r = −0.56) to a widely distributed cluster with positive summer correlations. The temporal regimes, interpreted post hoc as irrigation-dominated, mixed-control, and climate-influenced, showed distinct seasonality. The irrigation-dominated regime, mainly in spring, showed consistently negative relationships (mean r = −0.42), consistent with irrigation buffering climatic drought effects on evapotranspiration. The climate-influenced regime (June–September) showed predominantly positive relationships, which may reflect peak summer evaporative demand and heat stress, alongside water availability. The mixed-control regime, mainly in winter, showed weak, scale-dependent relationships (mean r = −0.16). Irrigation-related interpretations are inferred from correlation patterns, as no direct irrigation data were used. Together, these spatial and temporal patterns offer an exploratory multi-timescale framework for identifying where and when drought is most closely associated with agricultural evapotranspiration, supporting spatially and seasonally differentiated water management in the Pishkooh Plain.
Source: Agricultural Water Management
Institution: Agricultural Research & Education Organization
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,070
OpenAlex ID: W7233584545
