Assessment of SWOT-RiverSP Data Reliability and Accuracy over Canadian Rivers
Notice bibliographique
Résumé
Launched in December 2022, the SWOT (Surface Water and Ocean Topography) mission represents a major breakthrough in global surface water monitoring. Leveraging Ka-band radar interferometry, SWOT can observe water surface elevation (WSE) in rivers wider than 50 m, covering more than 150,000 km of linear river across Canada. While the scientific potential of SWOT is substantial, its operational use hinges critically on their reliability and accuracy of the data. In this context, a study was conducted to assess the quality of SWOT data comparing the RiverSP product with multiple reference sources across Canadian territory. The first component of this work compares SWOT-derived WSE with ground truth across diverse hydrological contexts, using records from national hydrometric stations. Approximately 100 stations located on rivers with known elevation datum were selected and properly converted to the SWOT datum. Each station was then matched to the nearest SWOT RiverSP node within a 200-meter range, allowing for direct comparison between in situ and SWOT WSE measurements. The second component is based on field campaigns conducted between 2023 and 2025 on four rivers: the Nashwaak (New Brunswick), Saint-François (Québec), Chaudière (Québec), and Au Saumon (Québec). High-precision GNSS measurements were performed from riverbank or by boat to measure WSE along the rivers during SWOT overpasses. These ground-based measurements were then compared to the nearest SWOT nodes, providing an opportunity to validate SWOT outputs under known conditions, while considering factors such as river width, slope, discharge, and environmental influences including vegetation and infrastructure. The third component examines the reliability of the SWOT-RiverSP product by analyzing the factors that influence the availability of high quality data. It was observed that certain areas consistently yield high-quality nodes (node_q < 2), while others regularly produce poor quality nodes (nod_q > 1). These factors are divided into three categories: 1) those associated with the sensors, such as cross-track distance, flow angle, or cross-over calibration, 2) those associated with river characteristics, such as width, the presence of canopy on the bank or rock in the river, proximity to a bridge, meanders, and slope; and 3) finally, those related to hydroclimatic conditions, such as lack of wind (which affects dark water), precipitation, discharges, and ice cover. A Random Forest model is used to rank the relative importance of these variables in predicting node reliability. The analysis is performed on rivers in New Brunswick and Québec where discharge data are available (from stations or hydrological models), as well as LiDAR data for river characterization. Wind and precipitation conditions are derived from ERA5-Land datasets, whereas vegetation analysis is assessed from aerial imagery. In addition, field campaigns have also been carried out in specific areas of concern to better understand the factors that may affect the SWOT signal. Overall, this study aims to assess the reliability and accuracy of SWOT data across Canadian rivers by integrating satellite observations, hydrometric stations, and field measurements. It lays the groundwork for the informed use of SWOT in hydrological modeling, flood forecasting, and freshwater resource management.
Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.
Comment cette classification a été obtenuedéplier
Prédiction machine sur la base complète
Imitation des enseignantsNi prévalence calibrée, ni vérité terrain. Validation humaine à venir. Le volet Gemma est une étiquette directe du modèle pour chaque travail de la base, lue sur la notice réduite au titre. Le volet Codex est un classifieur appris des 10 348 étiquettes directes de Codex et calibré sur les taux pondérés de l'échantillon; les champs sans appui suffisant ne portent aucun appel Codex. Le mode candidate est l'union des deux volets; le consensus est leur intersection. Ces sorties portent le statut machine_predicted_unvalidated et ne sont pas des étiquettes humaines.
Scores du classifieur distillé par catégorie (deux têtes)
| Catégorie | Codex | Gemma |
|---|---|---|
| Métarecherche | 0,005 | 0,016 |
| Méta-épidémiologie (sens strict) | 0,001 | 0,000 |
| Méta-épidémiologie (sens large) | 0,000 | 0,000 |
| Bibliométrie | 0,003 | 0,006 |
| Études des sciences et des technologies | 0,002 | 0,001 |
| Communication savante | 0,002 | 0,001 |
| Science ouverte | 0,002 | 0,001 |
| Intégrité de la recherche | 0,001 | 0,000 |
| Charge utile insuffisante (le modèle a refusé de juger) | 0,001 | 0,000 |
Scores machine (provisoires)
Les deux têtes enseignantes du modèle étudiant, lues sur ce travail. Un score ordonne la base pour la relecture; il n'affirme jamais une catégorie, et le statut de validation accompagne chaque rangée tel quel.
Scores de référence d'un modèle non mature (critères de maturité non atteints, 7 itérations). Un score ordonne; il n'affirme jamais une catégorie.
score_only:v0-immature-baseline · tel quel depuis la passe de notation : score_only signifie que le nombre peut ordonner les travaux, et qu'aucune étiquette de catégorie n'en découleClassification
machine, non validéePrédiction automatique; un appel candidat d’une seule source (Gemma direct ou Codex distillé), pas un consensus.
Le détail, modèle par modèle et score par score, se trouve en fin de page sous « Comment cette classification a été obtenue ».