The DOI matches its Crossref record; 51 of the 52 references it cites were found in a registry or an index.
Mitigating Prediction Error of Deep Learning Streamflow Models in Large Data‐Sparse Regions With Ensemble Modeling and Soft Data
Abstract
Predicting discharge in contiguously data‐scarce or ungauged regions is needed for quantifying the global hydrologic cycle. We show that prediction in ungauged regions (PUR) has major, underrecognized uncertainty and is drastically more difficult than previous problems where basins can be represented by neighboring or similar basins (known as prediction in ungauged basins). While deep neural networks demonstrated stellar performance for streamflow predictions, performance nonetheless declined for PUR, benchmarked here with a new stringent region‐based holdout test on a US data set with 671 basins. We tested approaches to reduce such errors, leveraging deep network's flexibility to integrate “soft” data, such as satellite‐based soil moisture product, or daily flow distributions which improved low flow simulations. A novel input‐selection ensemble improved average performance and greatly reduced catastrophic failures. Despite challenges, deep networks showed stronger performance metrics for PUR than traditional hydrologic models. They appear competitive for geoscientific modeling even in data‐scarce settings.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“While the LSTM's performance declines from temporal to PUB tests, it obtains better results than established process-based models calibrated on the test basins (Feng et al., 2021;Kratzert et al., 2019a).”
From The suitability of differentiable, physics-informed machine learning hydrologic models for ungauged regions and climate change impact assessment (2023) · cites this paper as Feng et al., 2021
In the citing paper
“For PUR, the LSTM's performance further declines significantly (Feng et al., 2021).”
From The suitability of differentiable, physics-informed machine learning hydrologic models for ungauged regions and climate change impact assessment (2023) · cites this paper as Feng et al., 2021
In the citing paper
“For both kinds of tests, regionalized LSTM models hold the performance record on daily hydrograph metrics (Feng et al., 2021;Kratzert et al., 2019a).”
From The suitability of differentiable, physics-informed machine learning hydrologic models for ungauged regions and climate change impact assessment (2023) · cites this paper as Feng et al., 2021
In the citing paper
“best PUR results in region 1 (northeast) and region 7 (northwest), with NSE/KGE medians larger than or close to 0.6 (Fig. B2), which are consistent with our previous PUR study using LSTM (Feng et al., 2021).”
From The suitability of differentiable, physics-informed machine learning hydrologic models for ungauged regions and climate change impact assessment (2023) · cites this paper as Feng et al., 2021
Citations
Cites
Hochreiter, S., & Schmidhuber, J. (1997) · Neural Computation · doi:10.1162/neco.1997.9.8.1735 · Cited by 102,318
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
River flow forecasting through conceptual models part I — A discussion of principles
Nash, J. E., & Sutcliffe, J. V. (1970) · Journal of Hydrology · doi:10.1016/0022-1694(70)90255-6 · Cited by 24,401
Model Evaluation Guidelines for Systematic Quantification of Accuracy in Watershed Simulations
Moriasi, D. N., Arnold, J. G., Liew, M. W. V., Bingner, R. L., Harmel, R. D., & Veith, T. L. (2007) · Transactions of the ASABE · doi:10.13031/2013.23153 · Cited by 13,446
Gupta, H. V., Kling, H., Yilmaz, K. K., & Martinez, G. F. (2009) · Journal of Hydrology · doi:10.1016/j.jhydrol.2009.08.003 · Cited by 6,758
Dropout as a bayesian approximation: Representing model uncertainty in deep learning
Gal, Y., & Ghahramani, Z. (2015) · arXiv · doi:10.48550/arxiv.1506.02142 · Cited by 6,471
The Soil Moisture Active Passive (SMAP) Mission
Entekhabi, D., Njoku, E. G., O'Neill, P. E., Kellogg, K. H., Crow, W. T., Edelstein, W. N., Entin, J. K., Goodman, S. D., Jackson, T. J., Johnson, J. T., et al. (2010) · Proceedings of the IEEE · doi:10.1109/jproc.2010.2043918 · Cited by 3,793
Constraints on future changes in climate and the hydrologic cycle
Allen, M. R., & Ingram, W. J. (2002) · Nature · doi:10.1038/nature01092 · Cited by 3,009
Rainfall–runoff modelling using Long Short-Term Memory (LSTM) networks
(2018) · doi:10.5194/hess-22-6005-2018 · Cited by 2,114
Shen, C. (2018) · Water Resources Research · doi:10.1029/2018wr022643 · Cited by 1,522
Self-citation
Research works printed but not linked: 4
- Addor N., 2017, Catchment attributes for large-sample studies
- Global Runoff Data Center (GRDC), 2020, BfG‐River discharge data‐Station maps
- Newman A. J., 2014, A large‐sample watershed‐scale hydrometeorological dataset for the contiguous USA
- O’Neill P. E., 2020, SMAP Enhanced L3 Radiometer Global Daily 9 km EASE‐Grid Soil Moisture, Version 4 . April 2015 to March 2020
Cited by
Sorted by citations
Differentiable modelling to unify machine learning and physical models for geosciences
Shen, C., Appling, A., Gentine, P., Bandai, T., Gupta, H. V., Tartakovsky, A. M., Baity‐Jesi, M., Fenicia, F., Kifer, D., Li, L., et al. (2023) · Nature Reviews Earth & Environment · doi:10.1038/s43017-023-00450-9 · Cited by 594
Self-citationFeng, D., Liu, J., Lawson, K., & Shen, C. (2022) · Water Resources Research · doi:10.1029/2022wr032404 · Cited by 386
Self-citationHybrid forecasting: blending climate predictions with AI models
Slater, L. J., Arnal, L., Boucher, M.-A., Chang, A. Y.-Y., Moulds, S., Murphy, C., Nearing, G., Shalev, G., Shen, C., Speight, L., et al. (2023) · Hydrology and Earth System Sciences · doi:10.5194/hess-27-1865-2023 · Cited by 281
Self-citationKumar, V., Kedam, N., Sharma, K. V., Mehta, D., & Caloiero, T. (2023) · Water · doi:10.3390/w15142572 · Cited by 270
Feng, D., Beck, H., Lawson, K., & Shen, C. (2023) · Hydrology and Earth System Sciences · doi:10.5194/hess-27-2357-2023 · Cited by 197
Self-citation
Metadata
- Authors
- Feng, D., Lawson, K., & Shen, C.
- Affiliations
- Pennsylvania State University (US)
- Corresponding author
- Shen, Chaopeng
- Venue
- Geophysical Research Letters
- Publisher
- American Geophysical Union (AGU)
- Published
- 2021-07-19
- Volume
- 48
- Issue
- 14
- ISSN
- 0094-8276 (ISSN-L)
- 1944-8007
- OpenAlex
- W3174646361
- Cited by
- 162
- References
- 52
- Citations by year
- 2026: 20 · 2025: 27 · 2024: 38 · 2023: 33 · 2022: 43 · 2021: 1
- FWCI
- 9.30
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- Locations
- Geophysical Research Letters (publishedVersion)
- Funders
- National Science Foundation (EAR#1832294)
- Topics
- Physical Sciences › Environmental Science › Environmental Engineering › Hydrological Forecasting Using AI
- Physical Sciences › Environmental Science › Water Science and Technology › Hydrology and Watershed Management Studies
- Physical Sciences › Environmental Science › Management, Monitoring, Policy and Law › Environmental Monitoring and Data Management
- Keywords
- streamflow forecasting, deep neural network, ensemble learning, feature selection, soft information, satellite soil moisture, low flow simulation, data-scarce basins, forecast uncertainty, ungauged basins
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
