Coupled statistical and deep-learning models link teleconnection indices with Yalong River runoff forecasts
Source check: Source checked. Not independent scientific confirmation.
Finding: Using monthly runoff records from four Yalong River Basin stations for 1953–2011 and atmospheric-circulation indices, the study reports a one-month index lag as most informative and identifies Northern Hemisphere Polar Vortex and East Asian Trough indices among predictive variables. SRA-MLPR reduced reported RMSE from 798.47 to 594.45 and MAPE from 34.79% to 22.90% versus the stated baseline, with NSE increasing from 0.67 to 0.76. The held-out data come from the same basin and time-series source; these predictive associations do not establish causal climate effects or transfer to other basins.
Technology: Stepwise regression, support vector regression, multilayer perceptron regression, and SHAP analysis
Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.
- AI contribution
- Stepwise regression selected atmospheric-index features and SVR/MLP models forecast runoff; SHAP was used to rank model predictors.
- Test and result
- Coupled models were evaluated on a held-out subset of four-station historical runoff data; the article reports 10-fold cross-validation and compares forecast metrics with baseline models.
- Limitation / still unknown
- The evaluation uses a single basin and historical time series; external basin transfer and causal climate attribution are not established.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://www.nature.com/articles/s41598-025-00115-1
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Full primary publisher article and primary bibliographic metadata, including Methods, Results, author affiliations and publication history.
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.