A graph neural network (GNN) approach to basin-scale river network learning: the role of physics-based connectivity and data fusion

Journal Article · Fri Oct 14 00:00:00 EDT 2022 · Hydrology and Earth System Sciences (Online)

DOI:https://doi.org/10.5194/hess-26-5163-2022· OSTI ID:1892363

Sun, Alexander Y. ; Jiang, Peishi ; Yang, Zong-Liang ; Xie, Yangxinyu; Chen, Xingyuan

Abstract. Rivers and river habitats around the world are under sustained pressure from human activities and the changing global environment. Our ability to quantify and manage the river states in a timely manner is critical for protecting the public safety and natural resources. In recent years, vector-based river network models have enabled modeling of large river basins at increasingly fine resolutions, but are computationally demanding. This work presents a multistage, physics-guided, graph neural network (GNN) approach for basin-scale river network learning and streamflow forecasting. During training, we train a GNN model to approximate outputs of a high-resolution vector-based river network model; we then fine-tune the pretrained GNN model with streamflow observations. We further apply a graph-based, data-fusion step to correct prediction biases. The GNN-based framework is first demonstrated over a snow-dominated watershed in the western United States. A series of experiments are performed to test different training and imputation strategies. Results show that the trained GNN model can effectively serve as a surrogate of the process-based model with high accuracy, with median Kling–Gupta efficiency (KGE) greater than 0.97. Application of the graph-based data fusion further reduces mismatch between the GNN model and observations, with as much as 50 % KGE improvement over some cross-validation gages. To improve scalability, a graph-coarsening procedure is introduced and is demonstrated over a much larger basin. Results show that graph coarsening achieves comparable prediction skills at only a fraction of training cost, thus providing important insights into the degree of physical realism needed for developing large-scale GNN-based river network models.

• Research Organization: Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
• Sponsoring Organization: USDOE Office of Science (SC), Biological and Environmental Research (BER); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); USDOE Office of Science (SC), Biological and Environmental Research (BER). Earth and Environmental Systems Science Division; Texas Advanced Computing Center
• Grant/Contract Number: AC05-76RL01830; SC0022211
• OSTI ID: 1892363
• Alternate ID(s): OSTI ID: 1893622
• Report Number(s): PNNL-SA-177141
• Journal Information: Hydrology and Earth System Sciences (Online), Journal Name: Hydrology and Earth System Sciences (Online) Vol. 26 Journal Issue: 19; ISSN 1607-7938
• Publisher: Copernicus GmbH
• Country of Publication: Germany
• Language: English

Related Subjects

54 ENVIRONMENTAL SCIENCES