Physics-constrained Active Learning for Soil Moisture Estimation and Optimal Sensor Placement

TL;DR

This paper presents a physics-constrained deep learning framework combined with active learning to improve soil moisture estimation and optimize sensor placement, reducing the need for dense sensor deployment.

Contribution

It introduces the P-DAL framework that integrates physics-based modeling, deep learning, and active learning for accurate soil moisture prediction and strategic sensor placement.

Findings

Enhanced soil moisture prediction accuracy.

Effective sensor placement strategy reduces sensor count.

Improved field-scale soil moisture monitoring performance.

Abstract

Soil moisture is a crucial hydrological state variable that has significant importance to the global environment and agriculture. Precise monitoring of soil moisture in crop fields is critical to reducing agricultural drought and improving crop yield. In-situ soil moisture sensors, which are buried at pre-determined depths and distributed across the field, are promising solutions for monitoring soil moisture. However, high-density sensor deployment is neither economically feasible nor practical. Thus, to achieve a higher spatial resolution of soil moisture dynamics using a limited number of sensors, we integrate a physics-based agro-hydrological model based on Richards' equation in a physics-constrained deep learning framework to accurately predict soil moisture dynamics in the soil's root zone. This approach ensures that soil moisture estimates align well with sensor observations while obeying physical laws at the same time. Furthermore, to strategically identify the locations for sensor placement, we introduce a novel active learning framework that combines space-filling design and physics residual-based sampling to maximize data acquisition potential with limited sensors. Our numerical results demonstrate that integrating Physics-constrained Deep Learning (P-DL) with an active learning strategy within a unified framework--named the Physics-constrained Active Learning (P-DAL) framework--significantly improves the predictive accuracy and effectiveness of field-scale soil moisture monitoring using in-situ sensors.