H11A-1319
Robust quantitative parameter estimation by advanced CMP measurements for vadose zone hydrological studies
Monday, 14 December 2015
Poster Hall (Moscone South)
Christian Koyama1, Honghua Wang2, Tseedulam Khuut3, Takayuki Kawai4 and Motoyuki Sato1, (1)Tohoku University, Center for Northeast Asian Studies, Sendai, Japan, (2)Central South University, School of Geosciences and Information Physics, Changsha, China, (3)Mongolian University of Science and Technology, School of Geology and Petroleum Engineering, Ulaanbaatar, Mongolia, (4)Niigata University, Research Institute for Natural Hazards and Disaster Recovery, Niigata, Japan
Abstract:
Soil moisture plays a crucial role in the understanding of processes in the vadose zone hydrology. In the last two decades ground penetrating radar (GPR) has been widely discussed has nondestructive measurement technique for soil moisture data. Especially the common mid-point (CMP) technique, which has been used in both seismic and GPR surveys to investigate the vertical velocity profiles, has a very high potential for quantitaive obervsations from the root zone to the ground water aquifer. However, the use is still rather limited today and algorithms for robust quantitative paramter estimation are lacking. In this study we develop an advanced processing scheme for operational soil moisture reetrieval at various depth. Using improved signal processing, together with a semblance - non-normalized cross-correlation sum combined stacking approach and the Dix formula, the interval velocities for multiple soil layers are obtained from the RMS velocities allowing for more accurate estimation of the permittivity at the reflecting point. Where the presence of a water saturated layer, like a groundwater aquifer, can be easily identified by its RMS velocity due to the high contrast compared to the unsaturated zone. By using a new semi-automated measurement technique the acquisition time for a full CMP gather with 1 cm intervals along a 10 m profile can be reduced significantly to under 2 minutes. The method is tested and validated under laboratory conditions in a sand-pit as well as on agricultural fields and beach sand in the Sendai city area. Comparison between CMP estimates and TDR measurements yield a very good agreement with RMSE of 1.5 Vol.-%. The accuracy of depth estimation is validated with errors smaller than 2%. Finally, we demonstrate application of the method in a test site in semi-arid Mongolia, namely the Orkhon River catchment in Bulgan, using commercial 100 MHz and 500 MHz RAMAC GPR antennas. The results demonstrate the suitability of the proposed method for advanced operational vadose zone research.
