ATTENTION/WARNING - NE PAS DÉPOSER ICI/DO NOT SUBMIT HERE

Ceci est la version de TEST de DIAL.mem. Veuillez ne pas soumettre votre mémoire sur ce site mais bien à l'URL suivante: 'https://thesis.dial.uclouvain.be'.
This is the TEST version of DIAL.mem. Please use the following URL to submit your master thesis: 'https://thesis.dial.uclouvain.be'.
 

Tropical forest type and land cover mapping from SAR-C multi-year time series (Sentinel-1)

Collet, Thibauld
(2022)

Files

Collet_27391700_2022.pdf
  • Open access
  • Adobe PDF
  • 19.35 MB
Download

Details

Supervisors
Defourny, Pierre
Faculty
Faculté des bioingénieurs
Degree label
Master [120] : bioingénieur en sciences et technologies de l'environnement, à finalité spécialisée
Abstract
The Congo Basin Forest is one of the most preserved forest ecosystems on the planet in terms of species diversity and density (Gillet and al., 2016). Besides its role as a biodiversity reserve, the Congo Basin has an important function as long-term carbon storage especially in the wetlands and peatlands from the Cuvette Centrale. The Cuvette Central area is the biggest peatland complex in the world and is mostly lying in the Republic of Congo and the Democratic Republic of Congo. It gathers all the high organic content wetlands of the Congo River and its tributaries. The quantity of carbon contained within the forest and forest soil differs depending on the forest density, soil water content, tree species and tree heights (Biddulph and al., 2021). This highlights the importance of forest type mapping for carbon stock monitoring. Those considerations make the region one of the most critical forest areas to monitor in the climate change context and the recent decrease in biodiversity. The Copernicus programme provides a large number of continuous data that are available in radar (Sentinel-1) imagery both very high resolution. Those data have the potential to constitute a good data source for forest type and land cover remote sensing. This master thesis will explore the regional Congo Basin regional context and Sentinel-1 remote sensing capabilities. The materials and the methodology used will be described. The results and limitations will be discussed. Finally, a complete land cover map will be presented. The final product of this study is a comprehensive land cover map at ten meter resolution based only on the use of SAR-C images. A map of forest cover change was also produced and was quantitatively and qualitatively compared to optical image-based classifications. Based on these maps and their validation, several conclusions are presented.