PHYSICS-BASED CONTINUOUS SIMULATION OF LONG-TERM NEAR-SURFACE HYDROLOGIC RESPONSE FOR THE COOS BAY EXPERIMENTAL CATCHMENT

This study employed the physics-based model InHM to conduct continuous variably saturated three-dimensional hydrologic-response simulation from 1990 through 1996 for the Coos Bay (CB1) experimental catchment. The boundary-value problem (BVP) used in this study was developed and tested in a prior study via InHM simulation of three relatively low intensity sprinkling experiments. The uniqueness of the BVP was assessed in this study using model performance evaluation against piezometric and discharge data for 33 storms extracted from the continuous observed record. The BVP from the relatively low intensity sprinkling experiments did not perform well (i.e., is nonunique) for larger intensity storms. While the uniqueness of the integrated hydrologic response (i.e., discharge) could be greatly improved by incorporating further site characterization information into the BVP, the uniqueness of the distributed hydrologic response (i.e., piezometers) was not improved. None of the continuous InHM simulations in this study could adequately reproduce the observed pore-water pressures, suggesting that further characterization of the locations and connectivities of bedrock fractures is critical for future efforts. The simulations presented here, supplemented with bedrock piezometer observations, suggest the potential for interaction between the deeper water table and near-surface hydrologic response. The results reported herein indicate that uniqueness is a problem for physics-based models when employing a BVP used successfully for smaller magnitude storms to simulate larger storms. The findings of this study are pertinent to the emerging field of hydrogeomorphology, including landslide initiation.