Search for: All records

Abstract Hydrogeologic models generally require gridded subsurface properties, however these inputs are often difficult to obtain and highly uncertain. Parametrizing computationally expensive models where extensive calibration is computationally infeasible is a long standing challenge in hydrogeology. Here we present a machine learning framework to address this challenge. We train an inversion model to learn the relationship between water table depth and hydraulic conductivity using a small number of physical simulations. For a 31M grid cell model of the US we demonstrate that the inversion model can produce a reliable K field using only 30 simulations for training. Furthermore, we show that the inversion model captures physically realistic relationships between variables, even for relationships that were not directly trained on. While there are still limitations for out of sample parameters, the general framework presented here provides a promising approach for parametrizing expensive models. 

Abstract Integrated hydrologic models can simulate coupled surface and subsurface processes but are computationally expensive to run at high resolutions over large domains. Here we develop a novel deep learning model to emulate subsurface flows simulated by the integrated ParFlow‐CLM model across the contiguous US. We compare convolutional neural networks like ResNet and UNet run autoregressively against our novel architecture called the Forced SpatioTemporal RNN (FSTR). The FSTR model incorporates separate encoding of initial conditions, static parameters, and meteorological forcings, which are fused in a recurrent loop to produce spatiotemporal predictions of groundwater. We evaluate the model architectures on their ability to reproduce 4D pressure heads, water table depths, and surface soil moisture over the contiguous US at 1 km resolution and daily time steps over the course of a full water year. The FSTR model shows superior performance to the baseline models, producing stable simulations that capture both seasonal and event‐scale dynamics across a wide array of hydroclimatic regimes. The emulators provide over 1,000× speedup compared to the original physical model, which will enable new capabilities like uncertainty quantification and data assimilation for integrated hydrologic modeling that were not previously possible. Our results demonstrate the promise of using specialized deep learning architectures like FSTR for emulating complex process‐based models without sacrificing fidelity. 

Abstract Groundwater is by far the largest unfrozen freshwater resource on the planet. It plays a critical role as the bottom of the hydrologic cycle, redistributing water in the subsurface and supporting plants and surface water bodies. However, groundwater has historically been excluded or greatly simplified in global models. In recent years, there has been an international push to develop global scale groundwater modeling and analysis. This progress has provided some critical first steps. Still, much additional work will be needed to achieve a consistent global groundwater framework that interacts seamlessly with observational datasets and other earth system and global circulation models. Here we outline a vision for a global groundwater platform for groundwater monitoring and prediction and identify the key technological and data challenges that are currently limiting progress. Any global platform of this type must be interdisciplinary and cannot be achieved by the groundwater modeling community in isolation. Therefore, we also provide a high‐level overview of the groundwater system, approaches to groundwater modeling and the current state of global groundwater representations, such that readers of all backgrounds can engage in this challenge.