Pith. sign in

REVIEW 4 major objections 5 minor 36 references

Utilizing Earth Foundation Models to Enhance the Simulation Performance of Hydrological Models with AlphaEarth Embeddings

T0 review · 4 major / 5 minor · reviewed 2026-08-03 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A 64-dimensional satellite embedding of each basin, averaged over eight years of imagery, replaces hand-crafted catchment attributes and improves out-of-sample streamflow prediction.

desk verdict A useful first benchmark for Earth foundation model embeddings in CAMELS PUB, but the §2.2 'no leakage' claim is wrong as stated and the central OOS gain is less clean than the paper suggests. read the letter →

arxiv 2601.01558 v2 pith:G6A57VPG submitted 2026-01-04 cs.LG cs.AI

Pengfei Qu , Wenyu Ouyang , Chi Zhang , Yikai Chai , Shuolong Xu , Lei Ye , Yongri Piao , Miao Zhang
show 1 more author
Huchuan Lu
This is my paper · ORCID
classification cs.LGcs.AI
keywords predictioninungaugedbasinsEarthfoundationmodelssatelliteembeddingsstaticbasinattributesLSTMCAMELSsimilaritycross-regionalgeneralization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that basin descriptors learned by an Earth foundation model from satellite imagery are more transferable than the hand-crafted attributes traditionally used in large-sample hydrology. Using the 671-basin CAMELS benchmark, it finds that the two descriptor types perform about equally when predicting gauged basins in a new time period (median NSE 0.708 vs 0.706), but the satellite embeddings pull ahead when predicting basins never seen in training (0.612 vs 0.553). It also shows that choosing donor basins by cosine similarity in the embedding space yields more coherent, hydrologically consistent training sets, improving ungauged-basin prediction at small and moderate training sizes. If right, the result would give hydrologists a data-driven, task-agnostic way to characterize catchment similarity and would shift ungauged-basin prediction toward foundation-model representations.

What carries the argument

The load-bearing object is the AlphaEarth Foundation embedding: a 64-dimensional vector per location, produced by a self-supervised model trained on global satellite imagery, averaged here over pixels and over the 2017-2024 period to give one static descriptor per basin. It replaces the 17 hand-crafted CAMELS attributes as the static input to an LSTM rainfall-runoff model, and its cosine similarity defines basin relatedness for donor-basin selection. The work it does is to supply a dense, integrated land-surface signal — vegetation, terrain, surface moisture, land cover — that the paper argues is more transferable and more discriminative than sparse expert attributes.

What would settle it

Recompute the 5-fold out-of-sample comparison using only satellite embeddings from 2017 (or from a foundation model trained on imagery ending before 2010). If the median NSE advantage over CAMELS attributes (0.612 vs 0.553) shrinks or reverses, the reported cross-regional generalization is largely an artifact of post-test land-surface information embedded in the averaged 2017-2024 vectors.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that AlphaEarth Foundation (AEF) embeddings — 64-dimensional vectors summarizing multi-year satellite imagery of each basin — constitute a more transferable static basin representation than the 17 hand-crafted CAMELS attributes. In 5-fold spatial cross-validation the AEF-based LSTM reaches a median out-of-sample NSE of 0.612 against 0.553 for the CAMELS-attribute model, while in-sample performance is nearly identical (0.708 vs 0.706). Mutual-information analysis shows partial overlap with terrain and vegetation attributes plus additional environmental signal not present in CAMELS. The embeddings also define a similarity space with high globa

Load-bearing premise

The load-bearing premise is that a basin's 2017-2024 satellite embedding is a static description valid for the 1980-2014 modeling period; the paper's assertion that this setup 'inherently precludes' leakage is not logically forced, because the embeddings can contain land-surface information from after the test period.

Editorial extensions

If this is right

  • If AEF embeddings generalize as claimed, regional LSTM models can drop hand-crafted basin attributes without losing in-sample skill and with gains in spatial out-of-sample skill.
  • Similarity-based donor selection in the embedding space can make ungauged-basin prediction more data-efficient, reaching high skill with only 100-300 similar basins.
  • The observed performance decline when donor sets grow too large implies that pooling all available basins is not automatically best, supporting similarity-aware training-set construction.
  • The cross-regime results imply that a single global model may underperform a segmented, similarity-aware set of models at current data scales.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Inference: the 2017-2024 embeddings overlap the 2010-2014 test period, so the claimed 'no leakage' is not guaranteed; the OOS gain could partly reflect land-surface changes observed after the test years, and re-running with pre-2014 embeddings would clarify this.
  • Inference: the compact 64-dim descriptor may serve as a general-purpose catchment signature beyond streamflow — e.g., for water-quality, drought, or ecohydrological prediction in ungauged locations — though the paper does not test these.
  • Inference: the granularity-generalization trade-off identified for whole-cluster withholding suggests a conditional or mixture-of-experts architecture, rather than a single shared network, as the natural next step to exploit AEF precision.
  • Inference: because AEF embeddings are derived globally from publicly available satellite data, the approach transfers to regions without CAMELS-style attributes, which is testable by applying it to non-US large-sample datasets.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes using AlphaEarth Foundation (AEF) satellite embeddings, 64-dimensional vectors derived from 2017–2024 satellite imagery, as static basin descriptors for large-sample hydrological modeling with the CAMELS-US dataset. It compares LSTM models trained with AEF embeddings against models trained with 17 traditional CAMELS physiographic attributes under in-sample (IS) and out-of-sample (OOS) conditions, reporting similar IS performance (median NSE ≈ 0.708 vs 0.706) but higher OOS median NSE (0.612 vs 0.553). The paper also evaluates similarity-based donor-basin selection for prediction in ungauged basins (PUB) using attribute similarity, an MLP fusion-embedding similarity, and AEF embedding similarity across five target basins, concluding that AEF similarity identifies more coherent donor sets and improves performance for small-to-moderate training-set sizes. Additional analyses examine mutual information between AEF embeddings and CAMELS attributes and a cross-regime generalization experiment based on clustering basins by each representation.

Significance. If the OOS result is unbiased, the paper would make a useful contribution by demonstrating that Earth Foundation Model embeddings can serve as transferable basin descriptors for PUB, with public embeddings and code availability. The mutual-information analysis and similarity-space visualization are constructive, and the paper raises an important question about the trade-off between discriminative precision and cross-regime generalization. However, the central quantitative claim is currently vulnerable to temporal information leakage, and the PUB experiment is based on only five subjectively selected basins. These issues must be resolved before the paper's main conclusions can be accepted.

major comments (4)
  1. [§2.2, §4.1] The claim that the setup 'inherently precludes any risk of data leakage' is not established. AEF embeddings are annual composites from 2017–2024, while the OOS test period is 2010–2014. Land-surface conditions observed after the evaluation window can encode information about slow hydrological and ecological processes—vegetation memory, land-cover change, drought recovery, groundwater trends—that are correlated with streamflow in the test period. The CAMELS attributes, by contrast, are constructed from more historical data. The OOS median NSE gap (0.612 vs 0.553) may therefore reflect future information in the feature set rather than superior cross-regional transferability. This is load-bearing for the paper's central claim. Providing an explicit test or a carefully reasoned stability analysis is necessary; asserting temporal precedence of model training is not sufficient.
  2. [§4.1, Figure 4] The statistical significance of the OOS comparison is overstated. The KS statistic for the OOS NSE distributions is only 0.0863, and the median difference is 0.059 NSE units. With 671 basins, a small KS statistic can nonetheless yield p ≈ 6×10⁻⁹, but this does not imply a practically large effect. The text repeatedly describes the AEF advantage as 'substantially' or 'much stronger,' which is not supported by the reported effect size. I recommend reporting bootstrap confidence intervals for the median difference (and for the KS statistic) and discussing the practical hydrological significance of a 0.06 median NSE gain.
  3. [§3, Experiment B; §4.2] The PUB scaling experiment is based on five target basins that are described only as 'selected from distinct geographic and ecological regions' with no explicit selection criteria. This small, subjectively chosen sample cannot support general conclusions about AEF-based donor selection across the 671-basin dataset. The mean trend in Figure 8(f) has uncertainty bands, but no formal statistical test is performed across targets, and there is no correction for multiple comparisons across the 21 configurations per basin. I recommend either substantially expanding the target set, or reframing Experiment B as an exploratory case study rather than a general demonstration.
  4. [§5.2, Figure 11] The cross-regime generalization experiment is confounded by the different number of clusters selected for each representation: 12 clusters for CAMELS attributes versus 9 for AEF embeddings. Leave-one-cluster-out therefore yields different training-set sizes and cluster compositions for the two models, so the observed difference in generalization cannot be attributed solely to the feature space. The text acknowledges this but does not control for it. A cleaner comparison would use the same number of clusters for both representations, or otherwise balance training-set size, to test whether the 'granularity-generalization trade-off' is a real property of AEF embeddings or an artifact of the clustering configuration.
minor comments (5)
  1. [§3, Experiment A] The description of the IS implementation says 'the train period of 531 catchments,' but the paper states that all 671 CAMELS catchments were used. Please clarify whether this is a typo and, if 531 is intentional, explain the discrepancy.
  2. [§3, Experiment A] The training period ends in December 2004 and the test period begins in January 2010. The five-year gap is not discussed; please state whether this was intentional and whether it affects the comparison.
  3. [§4.2, text near Figure 8] The text says 'The results of Experiment B, summarized in Figure 7,' but the NSE scaling results appear in Figure 8. Please correct the cross-reference.
  4. [§2.2] The sentence 'since the hydrological model training and testing phases predate the satellite observations, this setup inherently precludes any risk of data leakage' should be reworded even if the authors decide to keep the temporal-mismatch assumption, because 'inherently precludes' is too strong given the mechanism described above.
  5. [§2.3, 3] The MLP-embedding similarity is trained on the same streamflow prediction task, so it is a supervised, task-specific baseline rather than a task-agnostic representation. This should be stated explicitly when comparing it with AEF embeddings.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the central comparison is an external-embedding benchmark with held-out basins; the only flagged issue is a data-leakage validity concern, not circular reasoning.

full rationale

The paper's central claim—AEF embeddings provide stronger cross-regional generalization than CAMELS attributes—is an empirical benchmark result, not a derivation. The AEF embeddings are an externally pretrained, fixed 64-dimensional representation from AlphaEarth; they are not fitted to the CAMELS streamflow labels used for evaluation. Experiment A trains LSTM variants on 80% of basins and tests on held-out 20% (with a temporal split of 1980–2004 train / 2010–2014 test), so the OOS NSE comparison is a genuine out-of-sample evaluation. The mutual-information analysis and clustering are descriptive, not load-bearing derivations. The only near-circular element is the 'MLP/fusion embedding' similarity baseline (Section 2.3, 'Fusion embedding similarity'), which is trained on the streamflow task itself; however, the paper uses it only as a comparison baseline, and the central AEF-vs-CAMELS result does not depend on it, so it does not raise the circularity score. Self-citations (Ouyang et al. 2021, 2025) are not load-bearing. A validity concern that the review rule asks to flag is the Section 2.2 assertion: 'since the hydrological model training and testing phases predate the satellite observations, this setup inherently precludes any risk of data leakage.' Since AEF embeddings are annual composites from 2017–2024 while the test period is 2010–2014, this claim is not logically established and may reflect temporal information asymmetry; but that is an experimental-validity issue, not a circularity issue. Under the hard rules requiring a quoted reduction to the paper's own inputs, no circular step is present.

Assumptions & free parameters 4 free parameters · 3 assumptions · 0 invented entities

No new entities are introduced. The main non-empirical assumptions are the temporal transferability of the satellite embeddings and the spatial aggregation method, plus the choice of hyperparameters. These are the most significant unverified inputs to the central claim.

free parameters (4)
  • LSTM hidden size = 128
    Hand-selected, not swept; affects model capacity and could influence the comparison.
  • Dropout rate = 0.4
    Hand-selected, not swept; a regularization choice.
  • Batch size = 256
    Hand-selected; affects training dynamics.
  • Sequence length = 365 days
    Hand-selected; long enough for annual memory but arbitrary.
assumptions (3)
  • ad hoc to paper AEF embeddings computed over 2017–2024 are valid static descriptors for the 1980–2014 hydrology.
    Introduced to make the CAMELS comparison work; the paper acknowledges the mismatch but claims no leakage (Section 2.2).
  • domain assumption Averaging pixel-level AEF embeddings across space and time yields a representative basin-scale vector.
    Used to obtain one vector per basin (Section 2.2); the aggregation could wash out sub-basin heterogeneity.
  • domain assumption The 17 CAMELS attributes used are the standard/adequate attribute baseline.
    Follows Feng et al. 2020; not independently verified here, and the choice of baseline affects the comparison.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Utilizing Earth Foundation Models to Enhance the Simulation Performance of Hydrological Models with AlphaEarth Embeddings." pith.science (2026). https://pith.science/paper/G6A57VPG

@misc{pith2026260101558,
  author       = {Pith},
  title        = {Pith review of: Utilizing Earth Foundation Models to Enhance the Simulation Performance of Hydrological Models with AlphaEarth Embeddings},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/G6A57VPG}},
  note         = {Machine review of arXiv:2601.01558}
}
read the original abstract

Predicting river flow in places without streamflow records is challenging because basins respond differently to climate, terrain, vegetation, and soils. Traditional basin attributes describe some of these differences, but they cannot fully represent the complexity of natural environments. This study examines whether AlphaEarth Foundation embeddings, which are learned from large collections of satellite images rather than designed by experts, offer a more informative way to describe basin characteristics. These embeddings summarize patterns in vegetation, land surface properties, and long-term environmental dynamics. We find that models using them achieve higher accuracy when predicting flows in basins not used for training, suggesting that they capture key physical differences more effectively than traditional attributes. We further investigate how selecting appropriate donor basins influences prediction in ungauged regions. Similarity based on the embeddings helps identify basins with comparable environmental and hydrological behavior, improving performance, whereas adding many dissimilar basins can reduce accuracy. The results show that satellite-informed environmental representations can strengthen hydrological forecasting and support the development of models that adapt more easily to different landscapes.

Figures

Figures reproduced from arXiv: 2601.01558 by the authors.

Figure 9
Figure 9. Basin clusters derived from (a) CAMELS attributes, (b) [PITH_FULL_IMAGE:figures/full_fig_p015_9.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 15 canonical work pages

  1. [1]

    • AEF similarity identifies highly coherent basins, enabling more effective PUB transfer

    manuscript submitted to Water Resources Research Utilizing Earth Foundation Models to Enhance the Simulation Performance of Hydrological Models with AlphaEarth Embeddings Pengfei Qu1, Wenyu Ouyang2, Chi Zhang2, Yikai Chai2, Shuolong Xu1, Lei Ye2, Yongri Piao3, Miao Zhang4, Huchuan Lu5 1 School of Computer Science and Technology, Dalian University of Techn...

    2003

  2. [3]

    https://doi.org/10.3334/ORNLDAAC/1328 Virtanen, P., Gommers, R., Oliphant, T.E., Haberland, M., Reddy, T., Cournapeau, D., Burovski, E., Peterson, P., Weckesser, W., Bright, J., Van Der Walt, S.J., Brett, M., Wilson, J., Millman, K.J., Mayorov, N., Nelson, A.R.J., Jones, E., Kern, R., Larson, E., Carey, C.J., Polat, İ., Feng, Y ., Moore, E.W., VanderPlas,...

    work page doi:10.3334/ornldaac/1328

  3. [5]

    nearest manuscript submitted to Water Resources Research neighbors

    Mutual information matrix between the 64-dimensional AEF embedding and the CAMELS static attributes (left), and distributions (histograms and boxplots) of two representative static attributes (right). The relative robustness of the AEF model raises questions about how deep learning models leverage physiographic attributes. Our results echo concerns raised...

    2022

  4. [10]

    dissimilar

    Boxplots of CAMELS Attributes Across AEF-Based Basin Clusters The broader structure shown in Figure 10 also illustrates that the AEF embeddings encode much more than simple visual similarity. They arrange basins along environmental gradients that correspond closely to dominant runoff-generation mechanisms. Although the embeddings were obtained through sel...

    2021

  5. [12]

    Water Resources Res

    The Data Synergy Effects of Time-Series Deep Learning Models in Hydrology. Water Resources Res. 58, e2021WR029583. https://doi.org/10.1029/2021WR029583 Feng, D., Fang, K., Shen, C.,

    work page doi:10.1029/2021wr029583

  6. [14]

    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 Gupta, H.V ., Kling, H., Yilmaz, K.K., Martinez, G.F.,

    work page doi:10.1029/2021gl092999

  7. [15]

    Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling. J. Hydrol. 377, 80–91. https://doi.org/10.1016/j.jhydrol.2009.08.003 He, K., Chen, X., Xie, S., Li, Y ., Dollár, P., Girshick, R.,

    doi:10.1016/j.jhydrol.2009.08.003 2009

  8. [18]

    Neural Comput

    Long Short-Term Memory. Neural Comput. 9, 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735 Hodges, J.L.,

    doi:10.1162/neco.1997.9.8.1735 1997

Show all 36 references
  1. [21]

    AAAI Press

    SatCLIP: global, general-purpose location embeddings with satellite imagery, in: Proceedings of the Thirty-Ninth AAAI Conference on Artificial Intelligence and Thirty-Seventh Conference on Innovative Applications of Artificial Intelligence and Fifteenth Symposium on Educationa...

    doi

  2. [22]

    Kratzert, F., Klotz, D., Herrnegger, M., Sampson, A.K., Hochreiter, S., Nearing, G.S., 2019a

    StefaLand: An efficient geoscience foundation model that improves dynamic land-surface predictions. Kratzert, F., Klotz, D., Herrnegger, M., Sampson, A.K., Hochreiter, S., Nearing, G.S., 2019a. Toward Improved Predictions in Ungauged Basins: Exploiting the Power of Machine Lea...

    2019 doi

  3. [23]

    Water Resources Res., 58, e2021WR031794

    Regionalization in a Global Hydrologic Deep Learning Model: From Physical Descriptors to Random Vectors. Water Resources Res., 58, e2021WR031794. https://doi.org/10.1029/2021wr031794 Mai, G., Huang, W., Sun, J., Song, S., Mishra, D., Liu, N., Gao, S., Liu, T., Cong, G., Hu, Y ...

    doi

  4. [24]

    https://doi.org/10.1145/3653070 Nai, C., Liu, X., Tang, Q., Liu, L., Sun, S., Gaffney, P.P.J.,

    doi

  5. [25]

    Water Resources Research 60, e2023WR035051

    A Novel Strategy for Automatic Selection of Cross-Basin Data to Improve Local Machine Learning-Based Runoff Models. Water Resources Research 60, e2023WR035051. https://doi.org/10.1029/2023WR035051 Nash, J.E., Sutcliffe, J.V .,

    doi

  6. [29]

    Water Resour

    Exploring hydrological variable interconnections and enhancing predictions for data-limited basins through multi-task learning. Water Resour. Res. 61, e2023WR036593. https://doi.org/10.1029/2023WR036593 Ouyang, W., Lawson, K., Feng, D., Ye, L., Zhang, C., Shen, C.,

    doi

  7. [30]

    Continental-scale streamflow modeling of basins with reservoirs: Towards a coherent deep-learning-based strategy. J. Hydrol. 599, 126455. https://doi.org/10.1016/j.jhydrol.2021.126455 Pedregosa, F., Pedregosa, F., Varoquaux, G., Varoquaux, G., Org, N., Gramfort, A., Gramfort, ...

    2021

  8. [31]

    Water Resources Research 57, e2021WR030437

    Regionalization for Ungauged Catchments — Lessons Learned From a Comparative Large-Sample Study. Water Resources Research 57, e2021WR030437. https://doi.org/10.1029/2021wr030437 Sivapalan, M., Takeuchi, K., Franks, S.W., Gupta, V .K., Karambiri, H., Lakshmi, V ., Liang, X., Mc...

    doi

  9. [33]

    Earth-Science Reviews 252, 104739

    Catchment characterization: Current descriptors, knowledge gaps and future opportunities. Earth-Science Reviews 252, 104739. https://doi.org/10.1016/j.earscirev.2024.104739 Thornton, P.E., Thornton, M.M., Mayer, B.W., Wei, Y ., Devarakonda, R., V ose, R.S., Cook, R.B.,

    2024

  10. [35]

    Nat Methods 17, 261–272

    SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 17, 261–272. https://doi.org/10.1038/s41592-019-0686-2 Yu, Q., Jiang, L., Schneider, R., Zheng, Y., Liu, J.,

    doi

  11. [36]

    Water Resources Research 60, e2023WR035876

    Deciphering the Mechanism of Better Predictions of Regional LSTM Models in Ungauged Basins. Water Resources Research 60, e2023WR035876. https://doi.org/10.1029/2023WR035876

    doi

  12. [256]

    Training period was January 1980–December 2004, testing period was January 2010– December 2014, sequence length was 365 days

    For prediction, the best epoch’s model state was used. Training period was January 1980–December 2004, testing period was January 2010– December 2014, sequence length was 365 days. We ran two different variants thereof. We trained two distinct models, each incorporating differ...

    1980

  13. [1958]

    The significance probability of the smirnov two-sample test. Ark. Mat. 3, 469–486. https://doi.org/10.1007/BF02589501 Hrachowitz, M., Savenije, H.H.G., Blöschl, G., McDonnell, J.J., Sivapalan, M., Pomeroy, J.W., Arheimer, B., Blume, T., Clark, M.P., Ehret, U., Fenicia, F., Fre...

    doi

  14. [1970]

    River flow forecasting through conceptual models part I — A discussion of principles. J. Hydrol. 10, 282–290. https://doi.org/10.1016/0022-1694(70)90255-6 Omernik, J.M., Griffith, G.E.,

    doi

  15. [1997]

    mutual noise

    trained on large collections of basins have demonstrated strong information sharing and generalization capacity, achieving impressive performance even in ungauged settings (Fang et al., 2022; Kratzert et al., 2019a). This success has led to the prevailing view that multi-basin...

    2022

  16. [2003]

    Hydrological Sciences Journal 48, 857–880

    IAHS Decade on Predictions in Ungauged Basins (PUB), 2003–2012: Shaping an exciting future for the hydrological sciences. Hydrological Sciences Journal 48, 857–880. https://doi.org/10.1623/hysj.48.6.857.51421 Szwarcman, D., Roy, S., Fraccaro, P., Gíslason, Þ.E., Blumenstiel, B...

    2003 doi

  17. [2008]

    Water Resources Research 44, 2007WR006240

    Spatial proximity, physical similarity, regression and ungaged catchments: A comparison of regionalization approaches based on 913 French catchments. Water Resources Research 44, 2007WR006240. https://doi.org/10.1029/2007WR006240 Ouyang, W., Gu, X., Ye, L., Liu, X., Zhang, C.,

    doi

  18. [2009]

    To account for uncertainty arising from stochastic model weight initialization, each model configuration was trained and evaluated using five different random seeds

    were calculated during the test period. To account for uncertainty arising from stochastic model weight initialization, each model configuration was trained and evaluated using five different random seeds. The computation of NSE and KGE was further subjected to a bootstrap pro...

    1958

  19. [2011]

    A review of regionalisation for continuous streamflow simulation. Hydrol. Earth Syst. Sci. 15, 3539–3553. https://doi.org/10.5194/hess-15-3539-2011 Heudorfer, B., Gupta, H.V ., Loritz, R.,

    2011 doi

  20. [2013]

    Hydrological Sciences Journal 58, 1198–1255

    A decade of Predictions in Ungauged Basins (PUB)—a review. Hydrological Sciences Journal 58, 1198–1255. https://doi.org/10.1080/02626667.2013.803183 Klemmer, K., Rolf, E., Robinson, C., Mackey, L., Rußwurm, M.,

    2013

  21. [2014]

    Environmental management 54, 1249–1266

    Ecoregions of the conterminous United States: evolution of a hierarchical spatial framework. Environmental management 54, 1249–1266. https://doi.org/10.1007/s00267-014-0364-1 Oudin, L., Andréassian, V ., Perrin, C., Michel, C., Le Moine, N.,

    doi

  22. [2016]

    These features were subsequently fed into two model variants utilized in this study

    data set in the CAMELS dataset. These features were subsequently fed into two model variants utilized in this study. One model variant utilized a set of 17 physiographic catchment attributes derived from the CAMELS dataset. This feature set is consistent with that employed by ...

    2020

  23. [2017]

    Hydrology and Earth System Sciences 21, 5293–5313

    The CAMELS data set: catchment attributes and meteorology for large-sample studies. Hydrology and Earth System Sciences 21, 5293–5313. https://doi.org/10.5194/hess-21-5293-2017 Brown, C.F., Kazmierski, M.R., Pasquarella, V .J., Rucklidge, W.J., Samsikova, M., Zhang, C., Shelha...

    2017 doi

  24. [2020]

    Water Resources Res

    Enhancing Streamflow Forecast and Extracting Insights Using Long-Short Term Memory Networks With Data Integration at Continental Scales. Water Resources Res. 56, e2019WR026793. https://doi.org/10.1029/2019wr026793 Feng, D., Lawson, K., Shen, C.,

    doi

  25. [2021]

    Notable examples include SatMAE (Cong et al., 2022), which leverages masked autoencoders (MAE) (He et al.,

    that describe each location with a rich, transferable representation unavailable to supervised, task-specific models. Notable examples include SatMAE (Cong et al., 2022), which leverages masked autoencoders (MAE) (He et al.,

    2022

  26. [2022]

    to learn representations from temporal and multi-spectral satellite imagery, and Prithvi-EO-2.0 (Szwarcman et al., 2025), a multi-temporal Geospatial Foundation Model (GFM) trained on millions of samples from Harmonized Landsat and Sentinel-2 (HLS) data. These GFMs provide hig...

    2025

  27. [2024]

    We acknowledge the temporal mismatch between this window and the standard CAMELS period (1980–2014). However, we treat these embeddings as static attributes based on the assumption that the fundamental physiographic characteristics they encode, including topography and dominan...

    1980

  28. [2025]

    https://doi.org/10.1029/2024GL113036 Hochreiter, S., Schmidhuber, J.,

    Are deep learning models in hydrology entity aware? Geophysical Research Letters 52, e2024GL113036. https://doi.org/10.1029/2024GL113036 Hochreiter, S., Schmidhuber, J.,

    doi

Pith tools

Reviewed August 3, 2026 · model on record in the stance chip above.