Accelerating the Lagrangian Particle Tracking in Hydrologic Modeling to Continental‐Scale

Abstract

Unprecedented climate change and anthropogenic activities have induced increasing ecohydrological problems, motivating the development of large‐scale hydrologic modeling for solutions. Water age/quality is as important as water quantity for understanding the terrestrial water cycle. However, scientific progress in tracking water parcels at large‐scale with high spatiotemporal resolutions is far behind that in simulating water balance/quantity owing to the lack of powerful modeling tools. EcoSLIM is a particle tracking model working with ParFlow‐CLM that couples integrated surface‐subsurface hydrology with land surface processes. Here, we demonstrate a parallel framework on distributed, multi‐Graphics Processing Unit platforms with Compute Unified Device Architecture‐Aware Message Passing Interface for accelerating EcoSLIM to continental‐scale. In tests from catchment‐, to regional‐, and then to continental‐scale using 25‐million to 1.6‐billion particles, EcoSLIM shows significant speedup and excellent parallel performance. The parallel framework is portable to atmospheric and oceanic particle tracking models, where the parallelization is inadequate, and a standard parallel framework is also absent. The parallelized EcoSLIM is a promising tool to accelerate our understanding of the terrestrial water cycle and the upscaling of subsurface hydrology to Earth System Models.

The Role of Topography in Controlling Evapotranspiration Age

“…It was developed to calculate water ages (groundwater, ET, and outflow) and diagnose source water composition (snow, rainfall, and historical groundwater). It was recently accelerated by multi-GPU (Yang et al, 2021;Yang et al, 2023aYang et al, , 2023bYang, Maxwell, & Valent, 2022), thus enabling particle tracking at the regional scale, which is computationally expensive and previously infeasible. Results of the third-year spin-up from the PF-CLM model were repeatedly used to spin-up the EcoSLIM model for a further 42 years at an hourly time step.…”

“…ParFlow-CLM (ParFlow Developers, 2022) is open source at https://github.com/parflow, https://doi.org/10.5281/ zenodo.6413322; EcoSLIM (Yang et al, 2023a(Yang et al, , 2023b is also open source at https://github.com/aureliayang/ EcoSLIM_CONUS, https://doi.org/10.5281/zenodo.7302297. The inputs scripts for simulations and the Figure data (including Figure data in Supporting Information S1) (Yang, Maxwell, McDonnell, et al, 2022;Yang, Maxwell, & Valent, 2022)…”

“…ParFlow‐CLM (ParFlow Developers, 2022) is open source at https://github.com/parflow, https://doi.org/10.5281/zenodo.6413322; EcoSLIM (Yang et al., 2023a, 2023b) is also open source at https://github.com/aureliayang/EcoSLIM_CONUS, https://doi.org/10.5281/zenodo.7302297.…”

“…As a result, we leverage the latest parallel architecture of the multi-GPU (Yang et al, 2021;Yang et al, 2023aYang et al, , 2023bYang, Maxwell, McDonnell, et al, 2022;Yang, Maxwell, & Valent, 2022) to develop a GPU-accelerated particle tracking approach linked to the integrated hydrologic modeling. We explicitly simulate flow paths and ET age at the regional scale with an explicit treatment of three-dimensional groundwater flow and ET at an hourly time step.…”

Spatial aggregation effects on the performance of machine learning metamodels for predicting transit time to baseflow

“…Results from these models can be calibrated to available observations of groundwater level or streamflow, thus alleviating the need to obtain high frequency tracer data to gain insights on transit times. Continental scale transit time assessment using high resolution, integrated hydrological modeling paired with particle tracking has been demonstrated in recent years, but such applications remain relatively rare due to their substantial computational requirements (Maxwell et al 2016, Yang et al 2023.…”