10 ans d'images satellites (Sentinel-2) pour étudier l'évolution des plages malgaches

10 ans d'images satellites (Sentinel-2) pour étudier l'évolution des plages malgaches

by Ranaivoson R Manoa, Révillion Christophe, Mollies Julie, Rakotondraompiana Solofo, Pennober Gwenna

Applied Sciences & Engineering Earth Sciences & Geography RGWGeographical information systems, geodata and remote sensing
🎙 Ranaivoson R Manoa, Révillion Christophe, Mollies Julie, Rakotondraompiana Solofo, Pennober Gwenna 👥 12K 📅 October 3, 2025 ⏱ 22 min 👁 77 📄 original study 🧭 2026-08-16
Available in: English (current) Français

Keywords

Sentinel-2shorelinecoastal vulnerabilitySAETDSAS

Summary

This presentation, given by a doctoral researcher from the University of La Réunion, details a study using 10 years of Sentinel-2 satellite imagery to analyze shoreline changes along the coast of Madagascar, focusing on the Mahajanga area. The research aims to address the lack of coastal monitoring in Madagascar, a country highly vulnerable to cyclones and coastal squeeze. The methodology involves pre-processing Sentinel-2 images using a Python-based chain, extracting shorelines with the SAET tool (which uses subpixel algorithms), and validating the results against high-resolution Pléiades imagery. The study found that recalculating cloud cover locally is more effective than relying on metadata, yielding 318 usable images for Mahajanga. Shoreline extraction achieved an RMSE of 5.78 meters, below the 10-meter pixel resolution. Analysis with DSAS revealed contrasting trends: stable areas with low urban pressure, persistent erosion in urbanized zones, and moderate variable erosion near river mouths. Major retreats coincided with cyclone seasons. The approach was also tested in Toliara, showing similar benefits of local cloud masking. The presentation concludes that this reproducible and cost-effective protocol provides a foundation for a future coastal observatory in Madagascar, with perspectives including tide correction and testing on other sites.

193 words

Critical Evaluation

Value of the Information & Strength of the Argument

The presentation provides valuable insights into the application of remote sensing for coastal monitoring in a data-sparse region. The argumentation is solid, based on a clear methodology and quantitative validation. The speaker justifies the choice of Sentinel-2 (free, 10m resolution, 5-day revisit) and the SAET tool (recent, flexible, subpixel accuracy). The results are presented with visual aids and statistical trends, and the limitations are acknowledged (e.g., need for tide correction, further testing). The discussion with the audience adds depth, addressing questions about data selection and shoreline definition.

Scientific Rigor, Source Quality, Title Accuracy

The scientific rigor is high: the study uses established tools (SAET, DSAS) and validates results against high-resolution imagery. The sources are not explicitly cited in the video, but the tools and methods are well-known in the literature. The title accurately reflects the content. The presentation is part of a doctoral research, indicating a peer-review process in the future. The video is a conference presentation, so it lacks formal citations, but the methodology is transparent and reproducible.

178 words

Title / Content Match

The title accurately reflects the content: the presentation focuses on using 10 years of Sentinel-2 imagery to study the evolution of Malagasy beaches.

Quality & Reliability

8/10

The presentation is based on a doctoral research project, with a clear methodology, validation of results (RMSE below pixel size), and use of established tools (SAET, DSAS). The speaker acknowledges limitations and future work. The video is a scientific presentation, not peer-reviewed, but the approach is rigorous.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

  • SAET (Shoreline Analysis and Extraction Tool) — Open-source Python tool used for shoreline extraction
  • DSAS (Digital Shoreline Analysis System) — Tool for analyzing shoreline changes
  • Sentinel-2 — Satellite mission providing 10m resolution imagery

Concurring Sources

  • Sentinel-2 mission — Provides the satellite data used in the study

Contribution & Novelties

The study contributes a reproducible and cost-effective protocol for generating shoreline time series from Sentinel-2 imagery in tropical contexts, addressing the lack of coastal monitoring in Madagascar. The local cloud masking approach improves image usability. The validation with Pléiades imagery demonstrates subpixel accuracy. This work is a step towards a national coastal observatory.

Pour aller plus loin :

86 words

Radar Profile

The radar profile shows high scores in quantity and quality of information, with a slightly lower technical level, indicating a well-balanced presentation accessible to a scientific audience. The reliability is high due to the validation steps.

Reliability 8/10

💬 No comments were provided for analysis.