Articles | Volume 26, issue 19
https://doi.org/10.5194/hess-26-5163-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/hess-26-5163-2022
© Author(s) 2022. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Research article
| 
14 Oct 2022
Research article |
| 14 Oct 2022
| 14 Oct 2022
A graph neural network (GNN) approach to basin-scale river network learning: the role of physics-based connectivity and data fusion
Bureau of Economic Geology, The University of Texas at Austin, Austin,
TX, USA
Peishi Jiang
Pacific Northwest National Laboratory, Richland, WA, USA
Zong-Liang Yang
Department of Geological Sciences, The University of Texas at Austin,
Austin, TX, USA
Yangxinyu Xie
Department of Computer Science, The University of Texas at Austin,
Austin, TX, USA
now at: The Department of Statistics and Data Science at The University
of Pennsylvania, Pennsylvania, USA
Xingyuan Chen
Pacific Northwest National Laboratory, Richland, WA, USA
Related authors
Peishi Jiang, Pin Shuai, Alexander Sun, Maruti K. Mudunuru, and Xingyuan Chen
Hydrol. Earth Syst. Sci., 27, 2621–2644, https://doi.org/10.5194/hess-27-2621-2023, https://doi.org/10.5194/hess-27-2621-2023, 2023
Short summary
Short summary
We developed a novel deep learning approach to estimate the parameters of a computationally expensive hydrological model on only a few hundred realizations. Our approach leverages the knowledge obtained by data-driven analysis to guide the design of the deep learning model used for parameter estimation. We demonstrate this approach by calibrating a state-of-the-art hydrological model against streamflow and evapotranspiration observations at a snow-dominated watershed in Colorado.
Maggi M. Laan, Stephanie G. Fulton, Vanessa A. Garayburu-Caruso, Morgan E. Barnes, Mikayla A. Borton, Xingyuan Chen, Yuliya Farris, Brieanne Forbes, Amy E. Goldman, Samantha Grieger, Robert O. Hall Jr., Matthew H. Kaufman, Xinming Lin, Erin L. M. Zionce, Sophia A. McKever, Allison Myers-Pigg, Opal Otenburg, Aaron C. Pelly, Huiying Ren, Lupita Renteria, Timothy D. Scheibe, Kyongho Son, Jerry Tagestad, Joshua M. Torgeson, and James C. Stegen
Biogeosciences, 22, 6137–6152, https://doi.org/10.5194/bg-22-6137-2025, https://doi.org/10.5194/bg-22-6137-2025, 2025
Short summary
Short summary
Respiration is a process that combines carbon and oxygen to generate energy for living organisms. Within a river, respiration in sediments and water makes variable contributions to the respiration of the whole river system. Contrary to conventional wisdom, we found that water column respiration did not increase strongly when moving from small streams to big rivers. Instead, it was locally influenced by temperature, nutrients, and suspended solids.
Yong-He Liu and Zong-Liang Yang
Geosci. Model Dev., 18, 3157–3174, https://doi.org/10.5194/gmd-18-3157-2025, https://doi.org/10.5194/gmd-18-3157-2025, 2025
Short summary
Short summary
NMH-CS 3.0 is a C#-based ecohydrological model reconstructed from the WRF-Hydro/Noah-MP model by translating the Fortran code of WRF-Hydro 3.0 and integrating a parallel river routing module. It enables efficient execution on multi-core personal computers. Simulations in the Yellow River basin demonstrate its consistency with WRF-Hydro outputs, providing a reliable alternative to the original Noah-MP model.
Mingjie Shi, Nate McDowell, Huilin Huang, Faria Zahura, Lingcheng Li, and Xingyuan Chen
Biogeosciences, 22, 2225–2238, https://doi.org/10.5194/bg-22-2225-2025, https://doi.org/10.5194/bg-22-2225-2025, 2025
Short summary
Short summary
Using Moderate Resolution Imaging Spectroradiometer data products, we quantitatively estimate the resistance and resilience of ecosystem functions to wildfires that occurred in the Columbia River basin in 2015. The carbon state exhibits lower resistance and resilience than the ecosystem fluxes. The random forest feature importance analysis indicates that burn severity plays a minor role in the resilience of grassland and a relatively major role in the resilience of forest and savanna.
James Stegen, Amy J. Burgin, Michelle H. Busch, Joshua B. Fisher, Joshua Ladau, Jenna Abrahamson, Lauren Kinsman-Costello, Li Li, Xingyuan Chen, Thibault Datry, Nate McDowell, Corianne Tatariw, Anna Braswell, Jillian M. Deines, Julia A. Guimond, Peter Regier, Kenton Rod, Edward K. P. Bam, Etienne Fluet-Chouinard, Inke Forbrich, Kristin L. Jaeger, Teri O'Meara, Tim Scheibe, Erin Seybold, Jon N. Sweetman, Jianqiu Zheng, Daniel C. Allen, Elizabeth Herndon, Beth A. Middleton, Scott Painter, Kevin Roche, Julianne Scamardo, Ross Vander Vorste, Kristin Boye, Ellen Wohl, Margaret Zimmer, Kelly Hondula, Maggi Laan, Anna Marshall, and Kaizad F. Patel
Biogeosciences, 22, 995–1034, https://doi.org/10.5194/bg-22-995-2025, https://doi.org/10.5194/bg-22-995-2025, 2025
Short summary
Short summary
The loss and gain of surface water (variable inundation) are common processes across Earth. Global change shifts variable inundation dynamics, highlighting a need for unified understanding that transcends individual variably inundated ecosystems (VIEs). We review the literature, highlight challenges, and emphasize opportunities to generate transferable knowledge by viewing VIEs through a common lens. We aim to inspire the emergence of a cross-VIE community based on a proposed continuum approach.
Katherine A. Muller, Peishi Jiang, Glenn Hammond, Tasneem Ahmadullah, Hyun-Seob Song, Ravi Kukkadapu, Nicholas Ward, Madison Bowe, Rosalie K. Chu, Qian Zhao, Vanessa A. Garayburu-Caruso, Alan Roebuck, and Xingyuan Chen
Geosci. Model Dev., 17, 8955–8968, https://doi.org/10.5194/gmd-17-8955-2024, https://doi.org/10.5194/gmd-17-8955-2024, 2024
Short summary
Short summary
The new Lambda-PFLOTRAN workflow incorporates organic matter chemistry into reaction networks to simulate aerobic respiration and biogeochemistry. Lambda-PFLOTRAN is a Python-based workflow in a Jupyter notebook interface that digests raw organic matter chemistry data via Fourier transform ion cyclotron resonance mass spectrometry, develops a representative reaction network, and completes a biogeochemical simulation with the open-source, parallel-reactive-flow, and transport code PFLOTRAN.
Stephanie G. Fulton, Morgan Barnes, Mikayla A. Borton, Xingyuan Chen, Yuliya Farris, Brieanne Forbes, Vanessa A. Garayburu-Caruso, Amy E. Goldman, Samantha Grieger, Robert Hall Jr., Matthew H. Kaufman, Xinming Lin, Erin McCann, Sophia A. McKever, Allison Myers-Pigg, Opal C. Otenburg, Aaron C. Pelly, Huiying Ren, Lupita Renteria, Timothy D. Scheibe, Kyongho Son, Jerry Tagestad, Joshua M. Torgeson, and James C. Stegen
EGUsphere, https://doi.org/10.5194/egusphere-2023-3038, https://doi.org/10.5194/egusphere-2023-3038, 2024
Preprint archived
Short summary
Short summary
This research examines oxygen use in rivers, which is central to the carbon cycle and water quality. The study focused on an environmentally diverse river basin in the western United States and found that oxygen use in river water was very slow and influenced by factors like water temperature and concentrations of nutrients and carbon in the water. Results suggest that in the study system, most of the oxygen use occurs via mechanisms directly or indirectly associated with riverbed sediments.
Cenlin He, Prasanth Valayamkunnath, Michael Barlage, Fei Chen, David Gochis, Ryan Cabell, Tim Schneider, Roy Rasmussen, Guo-Yue Niu, Zong-Liang Yang, Dev Niyogi, and Michael Ek
Geosci. Model Dev., 16, 5131–5151, https://doi.org/10.5194/gmd-16-5131-2023, https://doi.org/10.5194/gmd-16-5131-2023, 2023
Short summary
Short summary
Noah-MP is one of the most widely used open-source community land surface models in the world, designed for applications ranging from uncoupled land surface and ecohydrological process studies to coupled numerical weather prediction and decadal climate simulations. To facilitate model developments and applications, we modernize Noah-MP by adopting modern Fortran code and data structures and standards, which substantially enhance model modularity, interoperability, and applicability.
Peishi Jiang, Pin Shuai, Alexander Sun, Maruti K. Mudunuru, and Xingyuan Chen
Hydrol. Earth Syst. Sci., 27, 2621–2644, https://doi.org/10.5194/hess-27-2621-2023, https://doi.org/10.5194/hess-27-2621-2023, 2023
Short summary
Short summary
We developed a novel deep learning approach to estimate the parameters of a computationally expensive hydrological model on only a few hundred realizations. Our approach leverages the knowledge obtained by data-driven analysis to guide the design of the deep learning model used for parameter estimation. We demonstrate this approach by calibrating a state-of-the-art hydrological model against streamflow and evapotranspiration observations at a snow-dominated watershed in Colorado.
Hui Zheng, Wenli Fei, Zong-Liang Yang, Jiangfeng Wei, Long Zhao, Lingcheng Li, and Shu Wang
Earth Syst. Sci. Data, 15, 2755–2780, https://doi.org/10.5194/essd-15-2755-2023, https://doi.org/10.5194/essd-15-2755-2023, 2023
Short summary
Short summary
An ensemble of evapotranspiration, runoff, and water storage is estimated here using the Noah-MP land surface model by perturbing model parameterization schemes. The data could be beneficial for monitoring and understanding the variability of water resources. Model developers could also gain insights by intercomparing the ensemble members.
Pin Shuai, Xingyuan Chen, Utkarsh Mital, Ethan T. Coon, and Dipankar Dwivedi
Hydrol. Earth Syst. Sci., 26, 2245–2276, https://doi.org/10.5194/hess-26-2245-2022, https://doi.org/10.5194/hess-26-2245-2022, 2022
Short summary
Short summary
Using an integrated watershed model, we compared simulated watershed hydrologic variables driven by three publicly available gridded meteorological forcings (GMFs) at various spatial and temporal resolutions. Our results demonstrated that spatially distributed variables are sensitive to the spatial resolution of the GMF. The temporal resolution of the GMF impacts the dynamics of watershed responses. The choice of GMF depends on the quantity of interest and its spatial and temporal scales.
Huiying Ren, Erol Cromwell, Ben Kravitz, and Xingyuan Chen
Hydrol. Earth Syst. Sci., 26, 1727–1743, https://doi.org/10.5194/hess-26-1727-2022, https://doi.org/10.5194/hess-26-1727-2022, 2022
Short summary
Short summary
We used a deep learning method called long short-term memory (LSTM) to fill gaps in data collected by hydrologic monitoring networks. LSTM accounted for correlations in space and time and nonlinear trends in data. Compared to a traditional regression-based time-series method, LSTM performed comparably when filling gaps in data with smooth patterns, while it better captured highly dynamic patterns in data. Capturing such dynamics is critical for understanding dynamic complex system behaviors.
Cited articles
Alfieri, L., Burek, P., Dutra, E., Krzeminski, B., Muraro, D., Thielen, J., and Pappenberger, F.: GloFAS – global ensemble streamflow forecasting and flood early warning, Hydrol. Earth Syst. Sci., 17, 1161–1175, https://doi.org/10.5194/hess-17-1161-2013, 2013. a
Andersson, T. R., Hosking, J. S., Pérez-Ortiz, M., Paige, B., Elliott, A.,
Russell, C., Law, S., Jones, D. C., Wilkinson, J., Phillips, T., et al.:
Seasonal Arctic sea ice forecasting with probabilistic deep learning, Nat.
Commun., 12, 1–12, 2021. a
Barnett, T. P., Adam, J. C., and Lettenmaier, D. P.: Potential impacts of a
warming climate on water availability in snow-dominated regions, Nature, 438,
303–309, 2005. a
Battaglin, W. A., Hay, L. E., and Markstrom, S. L.: Watershed Scale Response to
Climate Change-East River Basin, Colorado, US Geological Survey Fact Sheet,
3126, 6 pp., 2011. a
Bauer, P., Stevens, B., and Hazeleger, W.: A digital twin of Earth for the
green transition, Nat. Clim. Change, 11, 80–83, 2021. a
Beck, H. E., De Roo, A., and van Dijk, A. I.: Global maps of streamflow
characteristics based on observations from several thousand catchments,
J. Hydrometeorol., 16, 1478–1501, 2015. a
Best, J.: Anthropogenic stresses on the world's big rivers, Nat. Geosci.,
12, 7–21, 2019. a
Biancamaria, S., Lettenmaier, D. P., and Pavelsky, T. M.: The SWOT mission and
its capabilities for land hydrology, in: Remote sensing and water resources,
Springer, 117–147, 2016. a
Blöschl, G., Hall, J., Parajka, J., Perdigão, R. A., Merz, B.,
Arheimer, B., Aronica, G. T., Bilibashi, A., Bonacci, O., Borga, M., et al.:
Changing climate shifts timing of European floods, Science, 357, 588–590,
2017. a
Bronstein, M. M., Bruna, J., LeCun, Y., Szlam, A., and Vandergheynst, P.:
Geometric deep learning: going beyond Euclidean data, IEEE Signal Proc.
Mag., 34, 18–42, 2017. a
Bronstein, M. M., Bruna, J., Cohen, T., and Veličković, P.: Geometric
deep learning: Grids, groups, graphs, geodesics, and gauges, arXiv [preprint], https://doi.org/10.48550/arXiv.2104.13478, 2021. a
Bruna, J., Zaremba, W., Szlam, A., and LeCun, Y.: Spectral networks and locally
connected networks on graphs, arXiv [preprint], https://doi.org/10.48550/arXiv.1312.6203, 2013. a
Bureau of Reclamation: Taylor Park Reservoir,
https://www.usbr.gov/projects/index.php?id=236, last access: 22 March 2022. a
Camps-Valls, G., Tuia, D., Zhu, X. X., and Reichstein, M.: Deep learning for
the Earth Sciences: A comprehensive approach to remote sensing, climate
science and geosciences, John Wiley & Sons, ISBN: 9781119646143, 2021. a
Dai, A. and Trenberth, K. E.: Estimates of freshwater discharge from
continents: Latitudinal and seasonal variations, J. Hydrometeorol.,
3, 660–687, 2002. a
Dauphin, Y. N., Fan, A., Auli, M., and Grangier, D.: Language modeling with
gated convolutional networks, Proceedings of the 34th International Conference on Machine Learning, PMLR, 70, 933–941, Sydney, Australia, 2017. a
Davenport, F. V., Herrera-Estrada, J. E., Burke, M., and Diffenbaugh, N. S.:
Flood size increases nonlinearly across the western United States in response
to lower snow-precipitation ratios, Water Resour. Res., 56,
e2019WR025571, https://doi.org/10.1029/2019WR025571, 2020. a
De Groot, R. S., Wilson, M. A., and Boumans, R. M.: A typology for the
classification, description and valuation of ecosystem functions, goods and
services, Ecol. Econom., 41, 393–408, 2002. a
Defferrard, M., Bresson, X., and Vandergheynst, P.: Convolutional neural
networks on graphs with fast localized spectral filtering, Adv. Neur. Info. Proc. Syst., 29, 3844–3852, 2016. a
Döll, P., Kaspar, F., and Lehner, B.: A global hydrological model for
deriving water availability indicators: model tuning and validation, J. Hydrol., 270, 105–134, 2003. a
Donges, J. F., Zou, Y., Marwan, N., and Kurths, J.: The backbone of the climate
network, Europhys. Lett., 87, 48007, https://doi.org/10.1209/0295-5075/87/48007, 2009a. a
Dottori, F., Szewczyk, W., Ciscar, J.-C., Zhao, F., Alfieri, L., Hirabayashi,
Y., Bianchi, A., Mongelli, I., Frieler, K., Betts, R. A., and Feyen L.: Increased
human and economic losses from river flooding with anthropogenic warming,
Nat. Clim. Change, 8, 781–786, 2018. a
Feng, D., Fang, K., and Shen, C.: Enhancing streamflow forecast and extracting
insights using long-short term memory networks with data integration at
continental scales, Water Resour. Res., 56, e2019WR026793, https://doi.org/10.1029/2019WR026793, 2020. a
Feng, D., Lawson, K., and Shen, C.: Mitigating prediction error of deep
learning streamflow models in large data-sparse regions with ensemble
modeling and soft data, Geophys. Res. Lett., 48, e2021GL092999, https://doi.org/10.1029/2021GL092999,
2021. a
Gochis, D., Barlage, M., Dugger, A., FitzGerald, K., Karsten, L., McAllister,
M., McCreight, J., Mills, J., RafieeiNasab, A., Read, L., Sampson, K., Yates, D., and Yu, W.: The
WRF-Hydro modeling system technical description (Version 5.0), NCAR
Technical Note, 107 pp., https://ral.ucar.edu/sites/default/files/public/WRFHydroV5TechnicalDescription.pdf (last access: 22 March 2022), 2018. a
Grill, G., Lehner, B., Thieme, M., Geenen, B., Tickner, D., Antonelli, F.,
Babu, S., Borrelli, P., Cheng, L., Crochetiere, H., et al.: Mapping the
world's free-flowing rivers, Nature, 569, 215–221, 2019. a
Hagberg, A., Swart, P., and Schult, D.: Exploring network structure, dynamics, and function using NetworkX (No. LA-UR-08-05495; LA-UR-08-5495), Los Alamos National Lab.(LANL), Los Alamos, NM (United States), https://networkx.org (last access: 22 March 2022), 2008. a
Hamilton, W. L.: Graph representation learning, Synthesis Lectures on Artifical
Intelligence and Machine Learning, 46, Morgan & Claypool Publishers, Williston, VT, ISBN: 1681739631, 2020. a
Hochreiter, S. and Schmidhuber, J.: Long short-term memory, Neural Comput.,
9, 1735–1780, 1997. a
Hrachowitz, M., Savenije, H., Blöschl, G., McDonnell, J., Sivapalan, M.,
Pomeroy, J., Arheimer, B., Blume, T., Clark, M., Ehret, U., et al.: A decade
of Predictions in Ungauged Basins (PUB) – a review, Hydrol. Sci. J., 58, 1198–1255, 2013. a
Hu, Z., Dong, Y., Wang, K., Chang, K.-W., and Sun, Y.: Gpt-GNN: Generative
pre-training of graph neural networks, in: Proceedings of the 26th ACM SIGKDD
International Conference on Knowledge Discovery & Data Mining,
1857–1867, 2020. a
Hubbard, S. S., Williams, K. H., Agarwal, D., Banfield, J., Beller, H.,
Bouskill, N., Brodie, E., Carroll, R., Dafflon, B., Dwivedi, D., Falco, N., Faybishenko, B., Maxwell, R., Nico, P., Steefel, C., Steltzer, H., Tokunaga, T., Tran, P. A., Wainwright, H., and Varadharajan, C.: The
East River, Colorado, Watershed: A mountainous community testbed for
improving predictive understanding of multiscale hydrological–biogeochemical
dynamics, Vadose Zone J., 17, 1–25, 2018. a
Jia, J. and Benson, A. R.: A unifying generative model for graph learning
algorithms: Label propagation, graph convolutions, and combinations, arXiv [preprint], https://doi.org/10.48550/arXiv.2101.07730, 2021. a
Jia, X., Zwart, J., Sadler, J., Appling, A., Oliver, S., Markstrom, S.,
Willard, J., Xu, S., Steinbach, M., Read, J., and Kumar, V.: Physics-guided
recurrent graph model for predicting flow and temperature in river networks,
in: Proceedings of the 2021 SIAM International Conference on Data Mining
(SDM), Society for Industrial and Applied Mathematics, 612–620, 2021. a, b, c, d
Johnson, M. and Blodgett, D.: NOAA National Water Model Reanalysis Data at
RENCI, HydroShare,
https://doi.org/10.4211/hs.89b0952512dd4b378dc5be8d2093310f (last access: 22 March 2022),
2020. a, b
Kashinath, K., Mustafa, M., Albert, A., Wu, J., Jiang, C., Esmaeilzadeh, S.,
Azizzadenesheli, K., Wang, R., Chattopadhyay, A., Singh, A., Manepalli, A., Chirila, D., Yu, R., Walters, R., White, B., Xiao, H., Tchelepi, H. A., Marcus, P., Anandkumar, A., Hassanzadeh, P., and Prabhat:
Physics-informed machine learning: case studies for weather and climate
modelling, Philos. T. Roy. Soc. A, 379,
20200093, https://doi.org/10.1098/rsta.2020.0093, 2021. a, b
Kitzmiller, D. H., Wu, W., Zhang, Z., Patrick, N., and Tan, X.: The analysis of
record for calibration: a high-resolution precipitation and surface weather
dataset for the united states, in: AGU Fall Meeting Abstracts, Vol. 2018, H41H–06, 2018. a
Knoben, W. J. M., Freer, J. E., and Woods, R. A.: Technical note: Inherent benchmark or not? Comparing Nash–Sutcliffe and Kling–Gupta efficiency scores, Hydrol. Earth Syst. Sci., 23, 4323–4331, https://doi.org/10.5194/hess-23-4323-2019, 2019. a
Kratzert, F., Klotz, D., Herrnegger, M., Sampson, A. K., Hochreiter, S., and
Nearing, G. S.: Toward improved predictions in ungauged basins: Exploiting
the power of machine learning, Water Resour. Res., 55,
11344–11354, 2019a. a
Kratzert, F., Klotz, D., Shalev, G., Klambauer, G., Hochreiter, S., and Nearing, G.: Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets, Hydrol. Earth Syst. Sci., 23, 5089–5110, https://doi.org/10.5194/hess-23-5089-2019, 2019b. a
Kratzert, F., Klotz, D., Hochreiter, S., and Nearing, G. S.: A note on leveraging synergy in multiple meteorological data sets with deep learning for rainfall–runoff modeling, Hydrol. Earth Syst. Sci., 25, 2685–2703, https://doi.org/10.5194/hess-25-2685-2021, 2021. a
Le, X.-H., Nguyen, D.-H., Jung, S., Yeon, M., and Lee, G.: Comparison of deep
learning techniques for river streamflow forecasting, IEEE Access, 9,
71805–71820, 2021. a
Lehner, B. and Grill, G.: Global river hydrography and network routing:
baseline data and new approaches to study the world's large river systems,
Hydrol. Proc., 27, 2171–2186, 2013. a
Li, Q., Han, Z., and Wu, X.-M.: Deeper insights into graph convolutional
networks for semi-supervised learning, in: Proceedings of the AAAI Conference
on Artificial Intelligence, Vol. 32, ISBN 978-1-57735-800-8, 2018. a
Lin, P., Pan, M., Beck, H. E., Yang, Y., Yamazaki, D., Frasson, R., David,
C. H., Durand, M., Pavelsky, T. M., Allen, G. H., Gleason, C. J., and Wood, E. F.: Global
reconstruction of naturalized river flows at 2.94 million reaches, Water
Resour. Res., 55, 6499–6516, 2019. a
Loshchilov, I. and Hutter, F.: Fixing weight decay regularization in adam,
arXiv [preprint], https://doi.org/10.48550/arXiv.1711.05101, 2018. a
Lu, D., Konapala, G., Painter, S. L., Kao, S.-C., and Gangrade, S.: Streamflow
simulation in data-scarce basins using Bayesian and physics-informed machine
learning models, J. Hydrometeorol., 22, 1421–1438, 2021. a
Malik, N., Bookhagen, B., Marwan, N., and Kurths, J.: Analysis of spatial and
temporal extreme monsoonal rainfall over South Asia using complex networks,
Clim. Dynam., 39, 971–987, 2012. a
Mizukami, N., Clark, M. P., Gharari, S., Kluzek, E., Pan, M., Lin, P., Beck,
H. E., and Yamazaki, D.: A Vector-Based River Routing Model for Earth System
Models: Parallelization and Global Applications, J. Adv. Model. Earth Syst., 13, e2020MS002434, https://doi.org/10.1029/2020MS002434, 2021. a, b, c
Moore, R. B. and Dewald, T. G.: The Road to NHDPlus–Advancements in Digital
Stream Networks and Associated Catchments, J. Am. Water Resour. Assoc., 52, 890–900, 2016. a
Newman, A. J., Clark, M. P., Sampson, K., Wood, A., Hay, L. E., Bock, A., Viger, R. J., Blodgett, D., Brekke, L., Arnold, J. R., Hopson, T., and Duan, Q.: Development of a large-sample watershed-scale hydrometeorological data set for the contiguous USA: data set characteristics and assessment of regional variability in hydrologic model performance, Hydrol. Earth Syst. Sci., 19, 209–223, https://doi.org/10.5194/hess-19-209-2015, 2015. a
Nonnenmacher, M. and Greenberg, D. S.: Deep emulators for differentiation,
forecasting, and parametrization in Earth science simulators, J. Adv. Model. Earth Syst., 13, e2021MS002554, https://doi.org/10.1029/2021MS002554, 2021. a
Palmer, M. A., Reidy Liermann, C. A., Nilsson, C., Flörke, M., Alcamo, J.,
Lake, P. S., and Bond, N.: Climate change and the world's river basins:
anticipating management options, Front. Ecol. Environ., 6,
81–89, 2008. a
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen,
T., Lin, Z., Gimelshein, N., Antiga, L., Desmaison, A., Kopf, A., Yang, E., DeVito, Z., Raison, M., Tejani, A., Chilamkurthy, S., Steiner, B., Fang, L., Bai, J., and Chintala, S.: Pytorch: An imperative
style, high-performance deep learning library, Adv. Neural Info. Proc. Syst., 32, , ISBN: 9781713807933, 2019. a
Pathak, J., Subramanian, S., Harrington, P., Raja, S., Chattopadhyay, A.,
Mardani, M., Kurth, T., Hall, D., Li, Z., Azizzadenesheli, K., Hassanzadeh, P., Kashinath, K., and Anandkumar A.:
FourCastNet: A Global Data-driven High-resolution Weather Model using
Adaptive Fourier Neural Operators, arXiv [preprint], https://doi.org/10.48550/arXiv.2202.11214, 2022. a, b
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel,
O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J.,
Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.:
Scikit-learn: Machine Learning in Python, J. Mach. Learn.
Res., 12, 2825–2830, 2011. a
Phillips, J. D., Schwanghart, W., and Heckmann, T.: Graph theory in the
geosciences, Earth-Sci. Rev., 143, 147–160, 2015. a
Pokhrel, Y., Hanasaki, N., Koirala, S., Cho, J., Yeh, P. J.-F., Kim, H., Kanae,
S., and Oki, T.: Incorporating anthropogenic water regulation modules into a
land surface model, J. Hydrometeorol., 13, 255–269, 2012. a
PRISM Climate Group: 30-year normals,
https://prism.oregonstate.edu/normals/, last access: 6 September 2022. a
Qin, Y., Abatzoglou, J. T., Siebert, S., Huning, L. S., AghaKouchak, A.,
Mankin, J. S., Hong, C., Tong, D., Davis, S. J., and Mueller, N. D.:
Agricultural risks from changing snowmelt, Nat. Clim. Change, 10,
459–465, 2020. a
Rasmussen, C. E. and Williams, C. K. I.: Gaussian processes for machine
learning, The MIT Press, ISBN: 9780262182539, Cambridge, MA, 2006. a
Rong, Y., Huang, W., Xu, T., and Huang, J.: Dropedge: Towards deep graph
convolutional networks on node classification,
arXiv [preprint], https://doi.org/10.48550/arXiv.1907.10903, 2019. a, b
Salas, F. R., Somos-Valenzuela, M. A., Dugger, A., Maidment, D. R., Gochis,
D. J., David, C. H., Yu, W., Ding, D., Clark, E. P., and Noman, N.: Towards
real-time continental scale streamflow simulation in continuous and discrete
space, J. Am. Water Resour. Assoc., 54, 7–27,
2018. a
Saunders, W.: Preparation of DEMs for use in environmental modeling analysis,
Hydrologic and Hydraulic Modeling Support, Redlands, CA: ESRI, 29–51,
2000. a
Schulthess, T. C., Bauer, P., Wedi, N., Fuhrer, O., Hoefler, T., and Schär,
C.: Reflecting on the goal and baseline for exascale computing: a roadmap
based on weather and climate simulations, Comput. Sci. Eng., 21, 30–41, 2018. a
Shen, C.: A transdisciplinary review of deep learning research and its
relevance for water resources scientists, Water Resour. Res., 54,
8558–8593, 2018. a
Simley, J. D. and Carswell Jr., W. J.: The national map–hydrography, US
Geological Survey Fact Sheet, 2009–3054, 4 p., 2009. a
Sun, A. Y., Wang, D., and Xu, X.: Monthly streamflow forecasting using Gaussian
process regression, J. Hydrol., 511, 72–81, 2014. a
Tejedor, A., Longjas, A., Passalacqua, P., Moreno, Y., and Foufoula-Georgiou,
E.: Multiplex networks: A framework for studying multiprocess multiscale
connectivity via coupled-network theory with an application to river deltas,
Geophys. Res. Lett., 45, 9681–9689, 2018. a
Thornton, P. E., Thornton, M. M., Mayer, B. W., Wilhelmi, N., Wei, Y.,
Devarakonda, R., and Cook, R.: Daymet: Daily surface weather on a 1 km grid
for North America, 1980–2008, Oak Ridge National Laboratory (ORNL)
Distributed Active Archive Center for Biogeochemical Dynamics (DAAC), https://doi.org/10.3334/ORNLDAAC/1219, 2012. a
Todini, E.: A model conditional processor to assess predictive uncertainty in
flood forecasting, Int. J. River Basin Manage., 6,
123–137, 2008. a
USGS: Watershed Boundary Dataset,
https://www.usgs.gov/national-hydrography/watershed-boundary-dataset, last access: 6 September 2022b. a
Van Beek, L., Wada, Y., and Bierkens, M. F.: Global monthly water stress: 1.
Water balance and water availability, Water Resour. Res., 47, 7, https://doi.org/10.1029/2010WR009791, 2011. a
Wang, Y., Jin, J., Zhang, W., Yu, Y., Zhang, Z., and Wipf, D.: Bag of tricks
for node classification with graph neural networks, arXiv [preprint], https://doi.org/10.48550/arXiv.2103.13355, 2021. a
Weerts, A. H., Winsemius, H. C., and Verkade, J. S.: Estimation of predictive hydrological uncertainty using quantile regression: examples from the National Flood Forecasting System (England and Wales), Hydrol. Earth Syst. Sci., 15, 255–265, https://doi.org/10.5194/hess-15-255-2011, 2011. a
Weiler, M., McGlynn, B. L., McGuire, K. J., and McDonnell, J. J.: How does
rainfall become runoff?, A combined tracer and runoff transfer function
approach, Water Resour. Res., 39, 11, https://doi.org/10.1029/2003WR002331, 2003. a
Willard, J., Jia, X., Xu, S., Steinbach, M., and Kumar, V.: Integrating
physics-based modeling with machine learning: A survey,
ACM Computing Surveys (CSUR), https://dl.acm.org/doi/abs/10.1145/3514228, 2020. a
Winnick, M. J., Carroll, R. W., Williams, K. H., Maxwell, R. M., Dong, W., and
Maher, K.: Snowmelt controls on concentration-discharge relationships and the
balance of oxidative and acid-base weathering fluxes in an alpine catchment,
East River, Colorado, Water Resour. Res., 53, 2507–2523, 2017. a
Winsemius, H. C., Aerts, J. C., Van Beek, L. P., Bierkens, M. F., Bouwman, A.,
Jongman, B., Kwadijk, J. C., Ligtvoet, W., Lucas, P. L., Van Vuuren, D. P.,
and Ward, P. J.: Global drivers of future river flood risk, Nat. Clim. Change, 6,
381–385, 2016. a
Xia, Y., Mitchell, K., Ek, M., Sheffield, J., Cosgrove, B., Wood, E., Luo, L.,
Alonge, C., Wei, H., Meng, J., et al.: Continental-scale water and energy
flux analysis and validation for the North American Land Data Assimilation
System project phase 2 (NLDAS-2): 1. Intercomparison and application of model
products, J. Geophys. Res.-Atmos., 117, D3, https://doi.org/10.1029/2011JD016048, 2012. a
Yang, T., Sun, F., Gentine, P., Liu, W., Wang, H., Yin, J., Du, M., and Liu,
C.: Evaluation and machine learning improvement of global hydrological
model-based flood simulations, Environ. Res. Lett., 14, 114027, https://doi.org/10.1088/1748-9326/ab4d5e,
2019. a
Ye, A., Duan, Q., Yuan, X., Wood, E. F., and Schaake, J.: Hydrologic
post-processing of MOPEX streamflow simulations, J. Hydrol., 508,
147–156, 2014. a
Yeo, I.-K. and Johnson, R. A.: A new family of power transformations to improve
normality or symmetry, Biometrika, 87, 954–959, 2000. a
Yuval, J. and O'Gorman, P. A.: Stable machine-learning parameterization of
subgrid processes for climate modeling at a range of resolutions, Nat. Commun., 11, 1–10, 2020. a
Zhang, L., Shi, Z., Han, J., Shi, A., and Ma, D.: Furcanext: End-to-end
monaural speech separation with dynamic gated dilated temporal convolutional
networks, in: International Conference on Multimedia Modeling,
Springer, 653–665, 2020. a
Zhao, F., Veldkamp, T. I., Frieler, K., Schewe, J., Ostberg, S., Willner, S.,
Schauberger, B., Gosling, S. N., Schmied, H. M., and Portmann, F. T.: The
critical role of the routing scheme in simulating peak river discharge in
global hydrological models, Environ. Res. Lett., 12, 075003, https://doi.org/10.1088/1748-9326/aa7250,
2017. a
Zhu, X. and Ghahramani, Z.: Learning from labeled and unlabeled data with label
propagation, https://doi.org/10.1.1.14.3864, 2002. a
Short summary
High-resolution river modeling is of great interest to local governments and stakeholders for flood-hazard mitigation. This work presents a physics-guided, machine learning (ML) framework for combining the strengths of high-resolution process-based river network models with a graph-based ML model capable of modeling spatiotemporal processes. Results show that the ML model can approximate the dynamics of the process model with high fidelity, and data fusion further improves the forecasting skill.
High-resolution river modeling is of great interest to local governments and stakeholders for...
