Representation of water table dynamics in a land surface scheme : observations, models, and analyses

Abstract

A recent regional-scale water balance analysis has indicated that the groundwater storage and groundwater runoff are significant terms in the monthly and annual water balance for areas with a shallow water table. However, most of the current land surface parameterization schemes lack any representation of regional groundwater aquifers. Such a simplified representation of subsurface hydrological processes would result in significant errors in the predicted land-surface states and fluxes especially for the shallow water table areas in humid regions. This study attempts to address this deficiency. To incorporate the water table dynamics into a land surface scheme LSX, a lumped aquifer model is developed to represent the regional unconfined aquifer as a nonlinear reservoir, in which the aquifer simultaneously receives the recharge from the overlying soils, and discharges runoff into streams. The dependence of groundwater runoff on the water table depth (WTD), i.e., groundwater rating-curve, is parameterized empirically based on the observations in Illinois. The unconfined aquifer model is linked to the soil model in a land surface scheme LSX through the groundwater recharge flux (i.e., soil drainage flux). The total thickness of the unsaturated zone varies in response to the water table fluctuations, thereby interactively couples the aquifer model with the soil model. The second issue to be addressed in this thesis is the representation of the sub-grid variability of water table depths (WTD) in the coupled model LSXGW. A statistical-dynamical (SD) approach is used to account for the effects of the unresolved sub-grid variability of WTD in the grid-scale groundwater runoff. The probability distribution function (PDF) of WTD is specified as a two-parameter Gamma distribution based on observations.

(cont.) The scale of this PDF is dynamic according to the varying grid-mean WTD at each time step. The shape parameter of the PDF describing the WTD is kept constant. The grid-scale groundwater rating-curve (i.e., aquifer storage-discharge relationship) is derived statistically by integrating a point groundwater runoff model with respect to the PDF of WTD. Next, a mosaic approach is utilized to account for the effects of sub-grid variability of WTD in the grid-scale groundwater recharge. According to the time-varying PDF, a grid-cell is categorized into different sub-grids based on WTD. The fraction describing each sub-grid can be determined from the WTD PDF; hence it varies with time. The grid-scale hydrologic fluxes are computed by averaging all the sub-grid fluxes weighted by their fractions. This new methodology combines the strengths of the SD approach and the mosaic approach. The developed model has been successfully tested in Illinois for an 11-year period (1984-1994). The results indicate that the simulated hydrologic variables (soil saturation and WTD) and fluxes (evaporation, runoff, and groundwater recharge) agree well with the observations in Illinois. Nevertheless, it is recognized that the excellent performance of LSXGW in the Illinois simulation is significantly attributed to the reliable estimation of the macro-scale groundwater rating-curve ...