Comparative Analysis Of Seagrass Biophysical Properties Mapping Using Multi-Resolution Satellite Ima
| Source: Mastodon | Original article
A new peer‑reviewed study has demonstrated that high‑resolution satellite imagery, when paired with machine‑learning algorithms, can accurately map the biophysical properties of seagrass beds in the shallow waters of Teluk Pandan, Lampung, Indonesia. The research, published in *Remote Sensing Applications: Society and Environment* (doi 10.1016/j.rsase.2026.102002), compared several multi‑resolution datasets—including Sentinel‑2, PlanetScope and WorldView‑3—against an extensive field‑collected database of seagrass biomass, leaf‑area index and species composition. By training convolutional neural networks on the calibrated field data, the authors produced spatially explicit maps that outperformed traditional object‑based image analysis in both precision and processing speed.
The breakthrough matters because seagrass meadows are among the world’s most productive carbon sinks and serve as critical nurseries for fisheries, yet they remain under‑monitored due to the difficulty of surveying turbid, shallow coastal zones. Remote sensing that can resolve fine‑scale variations in canopy density and health offers a cost‑effective, repeatable tool for national agencies and NGOs tasked with protecting these habitats. In Indonesia, where seagrass covers an estimated 2 million hectares, the ability to track changes from coastal development, dredging or climate‑driven bleaching could inform adaptive management and bolster commitments under the UN Decade on Ecosystem Restoration.
The next steps will test the workflow’s scalability across the archipelago’s diverse reef‑lagoon systems and integrate near‑real‑time data streams from emerging constellations such as Planet’s daily global coverage. Stakeholders will watch for collaborations between Indonesian research institutes, satellite providers and AI firms that could turn the methodology into an operational service, potentially feeding into regional blue‑carbon accounting frameworks and early‑warning systems for habitat loss.
Sources
Back to AIPULSEN