Tuesday, April 30, 2013
Simultaneously calibrating regional hydraulic heads and springflow discharge(s) in a deterministic groundwater flow model is a significant modeling challenge due to uncertainty and non-uniqueness of the drain conductance values. To solve this problem, an analytical approach for drain conductance estimation, based on the physical analogy between well specific capacity and drain conductance, was simulated, using four pairs of simple generic steady-state groundwater-flow models. Each pair of models simulates either a “well” or a “drain” at an identical model design (hydrologic properties, shape, and size of the model domains) but at different values of a prescribed constant head boundary condition. The value of the drain conductance for all “drain” models was chosen to be equal to the value of the well specific capacity calculated from the simulation results of all “well” models. The identical results obtained for the spatial distribution of simulated heads for each model pair proved the interchangeability of the “well” and “drain” models with identical model design and boundary conditions. Therefore, it is reasonable to accomplish an effective calibration of deterministic groundwater flow models in two subsequent steps: 1) regional hydraulic head calibration by presenting the springflow as a “flowing well” with prescribed discharge at each model time step, and 2) by replacing the “flowing well” (after hydraulic head calibration) with a drain with drain conductance equal to the analytically estimated specific capacity of the “flowing well.” The prescribed methodology for analytical drain conductance estimation was applied in a case study of the Comal Springs (New Braunfels, Texas). This study was based on a preliminary statistical analysis of historical well water-level and springflow data for separation of both springflows discharge from the up-thrown (unconfined) and down-thrown (confined) blocks of the Balcones Fault Zone Edwards Aquifer.
