Newswise — Mangrove forests are an integral component of coastal zones in tropical and subtropical regions, providing a wide range of goods and ecosystem services that play an important role in ecology. , disappearing and deteriorating.
One way to promote effective mangrove conservation and encourage conservation policies is to carefully assess mangrove habitat and its changes and identify fragmented areas. However, obtaining this kind of information is not always easy.
“Mangrove forests are mostly inaccessible because they are located on tidal flats and marshes,” said Dr. Neda Bihamta Toosi, a postdoctoral fellow at Isfahan University of Technology, Iran, who uses remote sensing to address changes in landscape patterns. I’m here.In a recent study in the journal nature conservationtogether with a team of authors, she explored ways to use machine learning to classify these fragile ecosystems.
By comparing the performance of different combinations of satellite imagery and classification methods, researchers investigated how well each method performed in mapping mangrove ecosystems.
“We developed a new method focused on landscape ecology to map spatial disturbances in mangrove ecosystems,” she explains. “The disturbance maps provided will efficiently facilitate future management and planning activities of mangrove ecosystems and support sustainable conservation of these coastal areas.”
The results of this study showed that object-oriented classification of fused Sentinel images can significantly improve the accuracy of mangrove land use/land cover classification.
“Using model-based landscape metrics and principal component analysis techniques to assess and monitor the state of such ecosystems is a time- and cost-effective approach. Generating detailed land cover maps is essential, and the freely available Sentinel-2 data ensures future continuity,” explains Dr. Bihamta Toosi.
The research team hopes that this approach can be used to provide information on trends in land cover change that affect the development and management of mangrove ecosystems, supporting better planning and decision-making.
“Our results on mapping mangrove ecosystems can contribute to improved management and conservation strategies for these ecosystems affected by human activity,” they wrote in the study.
Research papers:
Soffianian AR, Toosi NB, Asgarian A, Regnauld H, Fakheran S, Waser LT (2023) Resampled and fused Sentinel-2 data and machine learning algorithms for mangrove mapping on the northern coast of Qeshm Island, Iran. evaluation. Nature Conservation 52: 1-22. https://doi.org/10.3897/natureconservation.52.89639
