Continental-scale evaluation of a fully distributed coupled land surface and groundwater model ParFlow-CLM (v3.6.0) over Europe

Abstract

Abstract. High-resolution large-scale predictions of hydrologic states and fluxes are important for many multi-scale applications including water resource management. However, many of the existing global to continental scale hydrological models are applied at coarse resolution and or neglect lateral surface and groundwater flow, thereby not capturing smaller scale hydrologic processes. Applications of high-resolution and more complex models are often limited to watershed scales, neglecting the mesoscale climate effects on the water cycle. We implemented an integrated, physically-based coupled land surface groundwater model; Parflow-CLM version 3.6.0, over a pan-European model domain at 0.0275° ( 3 km) resolution. The model simulates three-dimensional variably saturated groundwater flow solving Richards equation and overland flow with a two-dimensional kinematic wave approximation, which is fully integrated with land surface exchange processes. A comprehensive evaluation of hydrologic states and fluxes, resulting from a 10 year (1997–2006) model simulation, was performed using in-situ and remote sensing observations including discharge, surface soil moisture (SM), evapotranspiration (ET), snow water equivalent and water table depth. Overall, the uncalibrated PF-CLM-EU3km model shows good agreement in simulating river discharge for 176 gauging stations across Europe. Comparison with satellite-based datasets of SM shows that PF-CLM-EU3km performs well in semi-arid and arid regions, but simulates overall higher SM in humid and cold regions. Comparisons with Global Land Evaporation Amsterdam Model(GLEAM) and Global Land Surface Satellite (GLASS) ET datasets show no significant differences, both, across the European domain (on average the difference is -0.09 and 0.30 mm d-1 for GLEAM and GLASS products, respectively), and within regions (R > 0.9). The large-scale high-resolution setup forms a basis for future studies, demonstrating the added value of capturing heterogeneities for improved water and energy flux simulations in physically-based fully distributed hydrologic models over very large model domains. This study also provides an evaluation reference for climate change impact projections and a climatology for hydrological forecasting, considering the effects of lateral surface and groundwater flows.

Estimating Groundwater Recharge in Fully Integrated pde‐Based Hydrological Models

“…Our results suggest that their ability to simulate lateral flow in the entire subsurface and on the land surface is key to identifying extreme values and distinct spatial and temporal patterns of groundwater recharge. With respect to the remaining high interest and recent development in large-scale simulations of fully integrated models (e.g., Hung et al, 2022;Naz et al, 2022;O'Neill et al, 2021), we see high potential for our method to extract further information from these computationally expensive model runs.…”

Impacts of Permeability Uncertainty in a Coupled Surface‐Subsurface Flow Model Under Perturbed Recharge Scenarios

“…Solvers and preconditioners continue to be improved and better parallelized (e.g., Gasper et al., 2014; Osei‐Kuffuor et al., 2014), allowing the heavy computational load to be dealt with in a shorter time. These advances in speed are pushing IHMs to the point where they could become predictive tools (Clark et al., 2015), rather than primarily descriptive or exploratory, but coupled models with up to billions of degrees of freedom or more (e.g., Abbaspour et al., 2015; Kollet et al., 2010; Maxwell & Condon, 2016; Naz et al., 2023; Zipper et al., 2019) have high degrees of non‐uniqueness and uncertainty. Outputs of such heavily parameterized simulations should not be considered as “predictive” without an assessment of their uncertainty (Meresa & Romanowicz, 2017), yet uncertainty quantification (UQ) in IHM‐based studies remains rare (Moges et al., 2021).…”

Integrated Hydrologic Model Calibration Under Non-Stationary Climates

“…Naz et al (2023) who developed high-resolution hydrologic models of continental Europe, coupling ParFlow with Common Land Model (CLM). In this work, the ParFlow-CLM model presented a relatively good performance in terms of estimating evapotranspiration, topsoil moisture, and groundwater storage(Naz et al, 2023).…”

“…Naz et al (2023) who developed high-resolution hydrologic models of continental Europe, coupling ParFlow with Common Land Model (CLM). In this work, the ParFlow-CLM model presented a relatively good performance in terms of estimating evapotranspiration, topsoil moisture, and groundwater storage(Naz et al, 2023). Other recent research includes work byYang et al (2023), who further advanced the ParFlow model developed for the contiguous United States (CONUS-ParFlow) to better represent continental-scale water source problems such as impacts of climate change on groundwater in the United States.…”

“…Other recent research includes work byYang et al (2023), who further advanced the ParFlow model developed for the contiguous United States (CONUS-ParFlow) to better represent continental-scale water source problems such as impacts of climate change on groundwater in the United States. Similar to the work byNaz et al (2023), this study also used the CLM model to provide the required atmospheric and environmental inputs to ParFlow(Condon & Maxwell, 2019). The model developed byYang et al (2023) provides a more accurate description of topography and hydrostratigraphy resulting in better simulation of surface water-groundwater interactions and variations.…”