Simulation of bedrock groundwater dynamics in a distributed rainfall-runoff model

Abstract:

Recent field studies have shown that bedrock groundwater may play an important role to control the dynamics of the saturated subsurface flow in humid and steep mountainous regions. However many of existing hydrologic models typically assume impermeable bedrock and simulate saturated subsurface flow as a dominant runoff process. The objective of this study is to develop a simple groundwater model for mountainous bedrock aquifer incorporated to a distributed rainfall-runoff model to assess the importance of the bedrock groundwater for hydrologic simulations. The proposed model assumes the exponential-decline hydraulic conductivity in the bedrock to avoid defining the unknown bottom boundary. For the soil layer, it employs a stage-discharge relationship to simulate the unsaturated and saturated subsurface flows and surface flow. The developed model was applied to two Japanese catchments with weathered granite and sedimentary bedrock. We evaluated the simulation not only by the agreement with observed hydrographs but also in terms of the characteristics of their recessions and dynamic storage changes. Compared to the sedimentary rock catchment, the granite rock catchment showed stronger hysteresis in the recession and also dynamic storage change and discharge relationship estimated from the observed discharge. The original model with only surface soil layer failed to simulate these patterns, while the proposed model with groundwater component improved significantly the representation of the special patterns. Our study indicated the importance of bedrock groundwater for controlling lateral subsurface flow in particular the granite catchment we studied here.