Soil moisture sensor location-allocation using spatial association of surface moisture data

Balancing cost and performance is typically required when deploying a soil moisture sensor array. The sensor array's performance is essentially dependent on the appropriate placement of the sensors, which is fundamentally a location-allocation problem. In this study, a novel approach based on spatial association of surface soil moisture (SASM) is presented. It proposes selecting a sub-sample of sensor locations that best represent the spatial distribution of soil moisture while maximizing the variance in soil moisture with the minimum number of sample sites. This approach was tested at two sites with maize cultivated fields in Colorado. Neutron probe readings were collected at 15 cm depth across 41 and 31 locations throughout the entire crop growing season in two maize fields in Colorado. The number of soil sensors were optimized in a range of 17–19 with optimum site configuration for all different data acquisition dates. A global measure of spatial association (GMSA) analysis indicated consistency in spatial pattern between reduced number of sub-samples and original samples. Strategic sensor placement, driven by insights into soil-water dynamics patterns and intrinsic field properties, is essential for informed decision-making in water management within an irrigated maize field.