Groundwater is the world's most important freshwater resource. Despite this importance, groundwater flow and interactions between groundwater and other parts of the hydrological cycle are often neglected or simplified in large-scale hydrological models. One of the challenges in simulating groundwater flow and continental to global scales is the lack of consistent globally available hydrogeological data. These input data are needed for a more realistic physical representation of the groundwater system, enabling the simulation of groundwater head dynamics and lateral flows. A realistic representation of the subsurface is especially important as large-scale hydrological models move to finer resolutions and aim to provide accurate and locally relevant hydrologic information everywhere. In this study, we aim at improving and extending on current available large-scale data sets providing information of the subsurface. We present a detailed aquifer representation for the continental United States and Canada at hyper resolution (250 × 250 m). We integrate local hydrogeological information, including observations of aquifer layer thickness, conductivity, and vertical structure, to obtain representative aquifer parameter values applicable to the continental scale. The methods used are simple and can be expanded to other parts of the world. Hydrological simulations were performed using the integrated hydrological model ParFlow and demonstrated improved model performance when using the new aquifer parameterization. Our results support that more detailed and accurate aquifer parameterization will advance our understanding of the groundwater system at larger scales.
Plain Language SummaryGroundwater is the largest available freshwater resource humans can use for drinking and growing food. It is stored beneath our feet in thick layers of sand or rock; we call these layers aquifers. For a realistic estimation of how much groundwater is stored and flows through these aquifers, detailed information on the properties of these layers is essential. However, this information is often missing, or not detailed enough, at the larger scales. In this study, we introduce a new method to obtain this detailed data by including information from observations and local-scale studies what we upscaled to the continent of North America. We estimate aquifer parameters at very high spatial resolution over a large domain. These data were not available previously. We evaluate our newly obtained aquifer parameterization against other existing data sets and tested model sensitivity to different horizontal and vertical resolutions and conductivity values. We used the integrated hydrological model ParFlow.Our results show that estimates of water table depth improved using our new parameterization. The methods we introduce here can be used by other modelers looking to improve their aquifer parameterization. We hope that our study can be used as a road map towards improved and more detailed aquifer parameterization used in current large-scale hydrological mo...