Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 21 of the 21 references it cites were found in a registry or an index; none of the 32 sources its claims cite is checked for support yet.

Historical hydrometric context and explainable multi-horizon evaluation of a GloFAS-derived discharge product near Babahoyo, Ecuador

Open access

Abstract

Open global hydrological products are often the only continuous source available for data-scarce locations, but their local use is stronger when the modelled signal is connected to documented hydrometric evidence. This study evaluates an explainable machine-learning workflow for anticipating the GloFAS v4-derived discharge series returned by the Open-Meteo Flood API near Babahoyo, Ecuador, and complements the modern 2017–2026 analysis with a historical hydrometric-context protocol based on official INAMHI yearbooks for the San Pablo-Babahoyo system. Direct models were assessed at t+1, t+3, t+7, and t+14 using leakage-aware features, rolling-origin backtesting, a 365-day holdout, prediction intervals, product-state classification, permutation importance, and SHAP. ElasticNetCV achieved R 2 /NSE of 0.900 at t+1 in the final holdout, while mean backtest skill weakened at longer horizons. The dominant role of current and recent discharge values shows that the main short-term contribution is efficient local post-processing of GloFAS persistence; meteorological and seasonal information becomes relatively more important as lead time increases. The study therefore contributes a transparent multi-horizon evaluation, an auditable pathway for historical gauge-level corroboration, and practical diagnostics for deciding when local emulation is informative and when the native global product should remain the primary reference.

Overview

CoChat Summary

TL;DR

This study evaluates an explainable machine-learning emulator for a GloFAS v4-derived discharge product near Babahoyo, Ecuador, at t+1, t+3, t+7, and t+14, alongside a historical INAMHI yearbook protocol. It finds strong short-horizon agreement with the model-derived target but no validation against observed river discharge, with skill and interval calibration degrading at longer leads.

Key findings

  • ElasticNetCV achieved final-holdout R²/NSE of 0.900 at t+1, 0.687 at t+3, 0.535 at t+7, and 0.424 at t+14; mean rolling-origin backtest R²/NSE was 0.777, 0.334, −0.065, and −0.609 at the same horizons.
  • At t+1, current GloFAS-derived discharge dominated permutation importance (0.6969), followed by flow_min_3d (0.0363) and precipitation_hours (0.0212), indicating short-horizon performance is principally autoregressive emulation of the source product.
  • Diagnostic 80% interval coverage was 74.5% at t+1, 59.7% at t+7, and 54.8% at t+14; average widths were 81.221 m³/s, 164.952 m³/s, and 135.189 m³/s, with t+3 omitted from the archived run.
  • The recursive emulator path diverged from the native Flood API trajectory by a mean difference near 160 m³/s; exploratory classification had 16 binary positives and only 3 Extreme cases in the holdout, so rare-state performance was inconclusive.

Methods

  • Quantitative, applied, non-experimental, retrospective longitudinal case study at latitude −1.80217 and longitude −79.53443, daily from 1 January 2017 through 28 March 2026; merged base series had 3,374 dates, with regression samples of 3,373 for t+1, 3,371 for t+3, 3,367 for t+7, and 3,360 for t+14.
  • Features used Open-Meteo Archive/Forecast meteorology and Open-Meteo Flood API discharge, with leakage-aware calendar, Fourier, rainfall, antecedent, lag, and interaction terms; targets were future GloFAS-derived discharge at t+1, t+3, t+7, and t+14. Models included ElasticNetCV, ExtraTreesRegressor, HistGradientBoostingRegressor, and XGBoost, evaluated with expanding-window rolling origin using three 90-day 2020 test blocks and a 365-day final holdout, plus persistence baselines, permutation importance, SHAP, and quantile prediction intervals.
  • A historical hydrometric-context protocol used official INAMHI yearbooks for station H371/H0371 San Pablo en Palmar, comparing candidate GloFAS cells and the original Babahoyo coordinate with matching Flood API dates using completeness, Spearman and Kendall association, cross-correlation across ±10-day lags, monthly-cycle similarity, and high-state event agreement.

Limitations

  • The paper states that the target is a GloFAS-derived grid-cell series near Babahoyo, not a confirmed river reach; the returned cell was not verified against a named Babahoyo/San Pablo reach, upstream drainage area, or channel identifier, and continuous local gauge discharge was unavailable. It also states that INAMHI yearbooks report historical water levels, not directly comparable discharge without a rating curve.
  • The paper states that temporal validation used only three 90-day blocks within January–September 2020, one final holdout, one random seed, and fixed model configurations; archived materials lack complete split-level ablations, repeated-run dispersion, the t+3 prediction interval, and historical native-forecast archives, and API-based products may evolve.
  • Evident from the text: classification labels used full-series quantiles before the temporal split, introducing mild label-definition leakage; quantile models were fitted independently without a non-crossing constraint; the prospective comparison against the native Flood API was a single-issuance consistency check, not a backtest, and no Diebold-Mariano comparison of predictive accuracy was available.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3389/frai.2026.1901591

Cites

21 of 21 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Image 1
CC BY · From the PDF
Image 2
CC BY · From the PDF
Image 3
CC BY · From the PDF
Image 4
CC BY · From the PDF
Image 5
CC BY · From the PDF
Image 6
CC BY · From the PDF
Image 7
CC BY · From the PDF
Image 8
CC BY · From the PDF
Image 9
CC BY · From the PDF
Residual diagnostics for the t+1 product-emulation model: residual distribution and residuals versus predicted GloFAS-derived discharge. The defective Q-Q panel was removed.
Residual diagnostics for the t+1 product-emulation model: residual distribution and residuals versus predicted GloFAS-derived discharge. The defective Q-Q panel was removed.CC BY · From the PDF
Image 11
CC BY · From the PDF
Image 12
CC BY · From the PDF
Image 13
CC BY · From the PDF
Image 14
CC BY · From the PDF
Image 15
CC BY · From the PDF
Image 16
CC BY · From the PDF
TABLE 6 Exploratory binary classification of the ≥ 0.90-quantile GloFAS-derived next-day state.
TABLE 6 Exploratory binary classification of the ≥ 0.90-quantile GloFAS-derived next-day state.CC BY · From the PDF
Image 18
CC BY · From the PDF
Image 19
CC BY · From the PDF

Verification

Claims the paper makes

0 Valid · 0 Not valid · 17 Not checked · 109 Own results and interpretations

44 more claims

Create a free account to see all 126 claims and to verify the 26 references still open.

See all claims — free account
  • It may state two separate claims as one. 65
  • Probably not a substantial claim. 24

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 21 references

Research works printed but not linked: 6

  1. 2024, GloFAS v4 Calibration Methodology and Parameters
  2. 1990, Anuario Hidrológico 1990
  3. 1991, Anuario Hidrológico 1991
  4. 1992, Anuario Hidrológico 1992
  5. 1993, Anuario Hidrológico 1993
  6. 2026, Publicaciones de la Dirección de la Información Hidrometeorológica

Cited by

Still being checked.

Metadata

Authors
Haro-Sarango, A., Cachay-Salcedo, S., Saavedra-Vasconez, J., Salcedo-Dávalos, R., & Proaño-Altamirano, E.
Affiliations
Universidad Peruana Unión (PE)
Corresponding author
Haro-Sarango, Alexander
Venue
Frontiers in Artificial Intelligence
Published
2026-08-21
Volume
9
Pages
1901591-1901591
ISSN
2624-8212 (ISSN-L)
OpenAlex
W7203865128
Cited by
0
References
21
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
  • Frontiers in Artificial Intelligence (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
Funders
Universidad Peruana Unión
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
GloFAS, river discharge prediction, explainable machine learning, SHAP, permutation importance, backtesting, data-scarce regions, Babahoyo, lead time

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.

A language model drew the claims from the paper on 2026-11-19, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-19 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

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

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Historical hydrometric context and explainable multi-horizon evaluation of a GloFAS-derived discharge product near Babahoyo, Ecuador. https://app.cochat.ai/papers/10.3389/frai.2026.1901591
The paper
Haro-Sarango, A., Cachay-Salcedo, S., Saavedra-Vasconez, J., Salcedo-Dávalos, R., & Proaño-Altamirano, E. (2026). Historical hydrometric context and explainable multi-horizon evaluation of a GloFAS-derived discharge product near Babahoyo, Ecuador. Frontiers in Artificial Intelligence. https://doi.org/10.3389/frai.2026.1901591

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.