Characterizing soil water content variability across spatial scales from optimized high-resolution distributed temperature sensing technique

Abstract

• A new method to calibrate DTS soil moisture ( θ ) field installations was tested. • The new calibration method resulted in a mean soil moisture RMSE of 0.03 m 3 m −3 . • The spatial variability of θ across scales was characterized. • Number of samples needed to capture the θ variability across scales was estimated. Fiber-optic Distributed Temperature Sensing, when combined with the Single-probe Heat-pulse technique can measure soil moisture ( θ ) across spatial scales. The key limitation of this system is in obtaining the relationship between soil thermal conductivity ( λ ) and θ for a specific field. Using the Department of Energy Atmospheric Radiation Measurement (ARM) site, this study tested a new methodology to account for the spatial variability in the λ - θ relationship using a Gaussian processes model. The resulting accurate θ measurements (RMSE = 0.03 m 3 m −3 ) were used to characterize the spatial variability of θ across scales and to develop an empirical equation that can correct for the changes in the θ spatial variability observed at different spatial resolutions. In addition, the number of required samples to accurately characterize θ and its variability over scales ranging from 5 m and 350 m were estimated. These findings provide key information to scale soil moisture from centimeters to hundreds of meters for process understanding.