A multi-scale IoT and remote sensing framework for field-level crop water use and water stress assessment
TL;DR
A multi-scale sensing framework that integrates continuous in situ IoT measurements with satellite- and UAS-derived observations to enable cross-scale assessment of crop water dynamics demonstrates how integrating continuous field sensing with remote sensing observations enhances characterization of crop water use and water stress beyond what either sensing modality can achieve independently.
Abstract
Effective characterization of crop water use and water stress at the field scale remains a key limitation in precision agriculture. This study presents a multi-scale sensing framework that integrates continuous in situ IoT measurements with satellite- and UAS-derived observations to enable cross-scale assessment of crop water dynamics. The system integrates soil and open-field microclimate, and actual crop evapotranspiration (ET) sensors within a WiFi-enabled Arduino-based IoT architecture for real-time data acquisition and cloud storage. Performance was evaluated in a soybean field by comparing LI-710–derived ET with METRIC EEFLUX satellite products and the FAO-56 crop coefficient method. Results showed overestimation of crop ET by remote sensing and empirical approaches, with strong agreement in temporal patterns (R² = 0.67–0.95, p < 0.05), reflecting scale-dependent differences between continuous field measurements and temporally discrete or generalized estimates. A Water Stress Index derived from in situ measurements exhibited a strong negative correlation with soil volumetric water content and consistent relationships with NDVI from both UAS and Landsat 8/9 imagery, linking soil moisture dynamics, crop physiological response, and canopy condition. Overall, the framework demonstrates how integrating continuous field sensing with remote sensing observations enhances characterization of crop water use and water stress beyond what either sensing modality can achieve independently, providing a foundation for improved irrigation decision support under field conditions.