Simulation-Guided Subsurface Drainage Detection

Subsurface drainage is an agricultural water management practice widely implemented in low-relief landscapes to regulate soil moisture, particularly in the root zone, and enable earlier planting. The benefits for agricultural production are countered by potential negative impacts on the environment and water quality, highlighting the importance of quantifying their effects and adopting best practices to minimize those negative impacts. However, comprehensive assessment has been challenging due to a lack of information on subsurface drainage spacing and layout. Further, existing methods to detect subsurface layouts are not applicable for large areas.

 

Researchers at Arizona State University has developed a novel method which combines imaging, computational modeling and AI to detect, analyze and map subsurface drainage systems in agricultural landscapes. It employs a soil moisture model to simulate moisture dynamics along with neural networks to recognize spatial moisture patterns and distinguish drainage features from natural variations. Verified through case studies in Utah and Ohio, this method provides precise insights into subsurface water flow and drainage network configurations.

 

This technology introduces a flexible and rapid approach for mapping of subsurface drainage networks and holds promise in agricultural land management.