Loading
Life sciences · Preprint
arXiv · September 10, 2026
Early or partial results. Treat as a signal, not a conclusion.
This preprint presents AmazonSWE, a large-scale spatiotemporal dataset for water surface elevation imputation across the Amazon basin, and proposes a bidirectional state-space model that reportedly outperforms one published baseline. The work addresses a real operational need for flood forecasting and water resource management but has not undergone peer review, lacks independent validation, and is not yet established as superior to operational methods in practice.
Dataset contribution with algorithmic validation on held-out test data. Amazon river basin; 19K+ river sections with satellite altimetry and in situ gauge observations.. Intervention: Bidirectional selective state space model for spatiotemporal graph imputation with topology-aware positional encodings. Compared with: State-of-the-art published SWOT-based WSE densification method integrating statistics with physical modeling. Amazon river basin.
Dataset covers over 19,000 river sections and 10 years (2016–2026) in the Amazon basin with < 1% daily observation coverage. Proposed model reduces RMSE against in situ gauges by 18–39% compared to the state-of-the-art published SWOT-based WSE densification method. Model produces predictions for every river section, whereas the baseline method covers only sections with sufficient nearby satellite coverage.
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
A novel dataset and proof-of-concept model for a real-world problem, but without clinical validation, peer review, or comparison to established operational methods beyond one published baseline.
As stated by the source record.
Quoted from the source exactly as published.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
Continuous monitoring of water surface elevation across river networks is critical for flood forecasting, water resource management, and understanding the global water cycle. Yet, the scarcity of in situ gauges across much of the globe constrains the development of reliable modeling frameworks. Satellite altimetry has the potential to alleviate this problem but its use is currently hindered by sparse temporal coverage. To this end, we introduce AmazonSWE, a dataset for training and evaluating large-scale spatiotemporal graph imputation methods that integrates processed satellite altimetry measurements from a range of sources, including the recent wide-swath SWOT sensor. The dataset covers over 19K river sections and 10 years (2016-2026) in the Amazon river basin, with in situ gauges held out for evaluation. Besides contributing a novel real-world use case with the potential for societal impact, AmazonSWE introduces significant technical challenges: with fewer than 1% of sections observed per day, the dataset is far sparser than existing imputation benchmarks, and its directed acyclic river topology is both structurally different from and larger than graphs in existing datasets. We show that prior spatiotemporal graph imputation methods are not adapted to this topology, scale and sparsity, and propose a simple bidirectional selective state space model that outperforms them by sampling connected subgraphs and flattening space and time into a single token sequence with topology-aware positional encodings. Compared to the state-of-the-art published method for SWOT-based WSE densification, which integrates statistics with physical modeling, our model reduces RMSE against in situ gauges by 18-39%, while producing predictions for every river section rather than only those with sufficient nearby satellite coverage.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.