Validation and comparison of approaches to mapping hydraulic conductivity at the continental scale using the Upper Colorado River Basin as a testbed, A

Abstract

Groundwater represents the world’s largest accessible freshwater reserve and constitutes a major component of the hydrology in many regions including the United States. Accurate, continental-scale hydrologic and groundwater modeling is necessary for realistic estimates of water availability at large scales. Computational demand makes the automated calibration of high-resolution, continental-scale models infeasible. This study improves the modeling of continental-scale hydrology by providing a synthesis of methods for estimating and mapping hydraulic conductivity. We derive, model, and validate a suite of approaches to determining hydraulic conductivity over the contiguous United States, as well as compare the results with published hydraulic conductivity sets from literature. We also compare the performance of validation simulations with those of geologically informed approaches. Finally, we study the impact of model thickness on effective transmissivity and impose a spatially variable vertical flow barrier at an estimated depth of bedrock. For validation, hydraulic conductivity datasets are used to inform an integrated hydrologic model of the Upper Colorado River Basin. Simulated daily streamflows are compared to daily flow data from 10 USGS stream gages in the domain, and annually averaged simulated groundwater depths are compared to observations from nearly 2,000 monitoring wells. We find that hydraulic conductivity estimates from methods derived in this study compare well with those of analytical approaches from literature. We also find that analytically derived hydraulic conductivity estimates validate well in comparison to geologically informed approaches.