Evaluating the importance of future data collection sites using parameter estimation and analytic element groundwater flow models

摘要

Given that resources available for model calibration data are limited, effort expended to determine the importance of future data collection for a given project objective can be extremely valuable if the cost in time and funding is not prohibitive. An approach for using parameter estimation sensitivity techniques with analytic element groundwater models to evaluate the importance of data collection sites for model calibration is presented here. Using this approach, it was found that head data collected away from surface-water features (i.e., lakes and streams) are more important than head data near surface-water features for estimating global conductivity, global recharge, and sub-basin recharge in both mature and immature drainage basins. Head data collected most distant from surface-water features can be more important for recharge estimation than information on streamflow in some cases. In basins where the terminus (most downstream site of a basin) is near a confluence, a mid-basin gage site away from the confluence was shown to be more important for estimating basin recharge than a gage site near the terminus. In the headwaters of an immature drainage network, a gage site that encompasses most of the basin of interest was most important. Finally, the determination of the optimal stream resistance is likely not tractable using the approach discussed in this paper because of low sensitivity to commonly collected head and flow information and the limitation of Dupuit-Forchheimer models for simulating heads near three-dimensional features such as streams and lakes.