A Field Study on GPR for Non-Invasive Measurement of Soil Moisture and a Preliminary Investigation of GPR-Remote Sensing Imagery Correlations

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

0087802 Rubin Information on soil moisture is vital in several areas of scientific research. In hydrology and meteorology, the near-surface soil moisture content is needed for accurately estimating the exchange of heat and vapor fluxes between the surface and the atmosphere, which is a crucial component of global circulation models. Frequent monitoring of soil moisture over large regions could significantly improve our ability to predict soil erosion and mass movement. Shallow soil moisture content is extremely important in determining the migration of contaminants from the ground surface. Soil moisture is highly variable in both space and time. Current instruments for measuring soil moisture, such as neutron probe or time domain reflectometry (TDR), offer reasonable solutions in terms of efforts and accuracy for measurement at a fixed location. These methods, however, become practically useless for collecting data over large areas since they require drilling and sample only small volumes of soil. Additionally, one cannot discount the effects of drilling on the data. Currently, no technique is available to accurately provide soil water content measurements over the spatial and temporal scales necessary for estimating, monitoring and modeling moisture movement in the vadose zone. The focus of the project is to investigate the potential of Ground Penetrating Radar (GPR) as a tool for large-scale monitoring of shallow subsurface soil moisture (down to depths of 2-3 meters). The work includes a series of experiments to determine optimal GPR data acquisition parameters, simple interpretation methods, effects of soil texture and moisture variability on the GPR response, and criteria for assessing accuracy and reliability. The experiments will be conducted both under controlled conditions at a constructed test site, at outcrops, and at a naturally heterogeneous field site in Napa Valley, CA. The work will be used to evaluate GPR technology and interpretation methods for various soil types and saturations. The controlled experiments will address (1) the sensitivity of GPR to variations in saturation and to variations (vertical and lateral) in soil types; (2) the effect of erosional surfaces and boundaries between soil horizons on the GPR response; (3) sensitivity to clay; and (4) optimal data acquisition and interpretation techniques. The field experiments will test the method under natural conditions at a geologically heterogeneous site at different times of year. The multidisciplinary team from U.C. Berkeley and Lawrence Berkeley National Laboratory (LBNL) is well poised to make significant progress in developing methods that would enable more accurate and extensive estimates of soil water content for use in global circulation models, drought management, vadose zone contaminant transport and geotechnical problems.