Non-invasive estimation of soil hydraulic parameters in boreal podzolic soils using ground-penetrating radar and hyrological models

Abstract

Obtaining information on soil hydraulic parameters (SHPs), including their spatiotemporal variation, is fundamental to understanding flow and transport processes in the vadose zone. This research examined the potential of coupling time-lapse ground-penetrating radar direct ground wave (GPR-DGW) information with hydrological models for non-invasive estimation of SHPs in a boreal podzolic soil using two experimental studies. The first research study focused on estimating soil water content (SWC) variation in podzolic soil using GPR-DGW and predicting field-saturated hydraulic conductivity (Kfₛ) by integrating the estimated SWC variations into the Green-Ampt (GA) model. The second study investigated the potential of integrating SWC variation derived from GPR-DGW in the Hydrus-1D model for obtaining Mualem-van Genuchten (M-vG) parameters and simulating soil water dynamics. Both studies involved short-term Beerkan infiltration experiments monitored using a surface GPR system of 500 MHz center frequency, collecting time-lapsed traces every 5 s. The first study revealed that GPR-DGW can successfully provide continuous, reliable SWC variation data during a short-term infiltration, which proved useful in obtaining probable Kfₛ values from the GA model. The second study also found that reliable M-vG parameters can be obtained through the inversion of SWC variations from GPR-DGW in the Hydrus-1D model. In addition, good correlations were observed between measured SWC variation and those simulated using the inverted M-vG parameters. Overall, this research demonstrates a significant advancement in using GPR as a reliable and robust technique for large-scale SHPs prediction in boreal podzolic soils, with the possibility of extending to other soil types in future research.