Persistence patterns in subfield crop water use to guide variable rate management

Agricultural water management and optimal resource allocation are critical for sustainable and efficient food production. Modern farming practices, underpinned by precision agriculture, leverage advanced technologies for optimal resource allocation. While precision agriculture has substantively advanced through technological interventions, it remains a challenge to identify spatial patterns that persist over time within field zones. In this study, we used historic 30-m gridded seasonal consumptive irrigation requirement maps (i.e., actual evapotranspiration /crop water uptake) developed by the University of Idaho and the Idaho Department of Water Resources over 16 years to identify fields that demonstrate persistent crop water use (CWU) patterns. Statistical analysis of the CWU, relative to the field average, reveals persistent patterns across irrigated areas of the Magic Valley region in Idaho, USA. Variable importance analysis using the Boruta feature selection algorithm showed that factors such as the dominant crop type across multiple seasons, within-field topographic differences, soil physical properties, and water source used in irrigation are all factors in the observed patterns. Persistent patterns in reduced CWU near field edges may not manifest in fields surrounded by other irrigated fields due to the oasis effects originating from the neighboring fields. This study advances knowledge of persistent CWU patterns and contributes a cost-effective and efficient methodology for evaluating field variability. Mapping persistent patterns enables identifying and targeting specific areas within the field that could benefit from variable rate management. The resulting persistence maps offer valuable insights for growers, agronomists, and water managers to allocate water resources and optimize crop yields.