|
Description
|
Mangroves are highly productive ecosystems that provide important ecosystem services, are strategic allies in carbon capture and storage, conserve different plant and wildlife species, are producers of aquatic species such as crabs and shrimp, and local communities have developed strong economic, cultural and identity ties. Despite their great ecological, economic, and social importance, mangroves are threatened by natural and anthropogenic factors, hence the importance of their constant monitoring. Remote sensing technology has demonstrated its ability to map changes in mangroves and technological advances allow faster application of mapping methodologies, optimizing costs and time. To facilitate the sustainable management of mangroves, an open tool based on remote sensing data and machine learning was developed on the Google Earth Engine platform (MANGLEE). MANGLEE was tested in the mangroves of Guayas, Ecuador. Mangrove cover maps were obtained for the years 2018, 2020 and 2022 as well as the mangrove change maps for the two periods 2018-2020 and 2020-2022.
The maps generated in this service are also available on the APP MANGLEE:
The methodology and the diffusion workshops are available at the following link.
The source code can be found in GitHub MANGLEE:
This publication is possible by the support of the people of the United States through the United States Agency for International Development .(USAID). The content of this publication is the responsibility of its authors and does not necessarily reflect the views of USAID or the Government of the United States of America.
NOTE: See the updated version for all Ecuador in the following link: https://doi.org/10.7910/DVN/RDTUZC .
|
|
Keyword
|
carbon sequestration, biodiversity, ecosystem services, biomass, ecology, land degradation, agroforestry systems, Mangrove cover, Mangrove change map, mapping, Random Forest, vegetation index, carbon sink, deforestation, Shrimp farms, Cloud computing, ecosystem services, ecosystem, classification, Sentinel 1, Sentinel 2, SAR, Google Earth Engine, GEE, SERVIR Amazonia, USAID, Ecuador, Guayas, Latin America and the Caribbean, Mangrove ecosystem |
|
Notes
| Methodology:
MANGLEE compiles and preprocesses optical (Sentinel-2) and synthetic aperture radar (SAR) Sentinel-1 data for a given year, calculates vegetation indexes, then with a training file and the Sentinels 1 and 2 stack, it uses a Random Forest classification to obtain the binary mangrove and non-mangrove map for the selected year, of 10 m spatial resolution. Using two coverage maps MANGLEE detects the changes greater than one-half hectare (gain and loss) pixel by pixel cataloging them in degradation or gain and presents all the results in a viewer that allows to compare the optical image of two different years. These results were evaluated and improved with ground truth data and visual interpretation. The process was repeated for both periods.
Fundación Ecuatoriana de estudios Ecológicos Ecociencia (Ecociencia).
Centro Internacional para la investigación del Fenómeno del Niño (CIIFEN). MANGLEE's methodology is yet to be published. |
