Mission
Assessing wetland resilience to climate change using satellite imagery
Although wetlands are essential for maintaining water resources and biodiversity, their functioning is being affected by human pressures and climate change. However, there is still a lack of knowledge about the resilience of these ecosystems to the abrupt or gradual effects of climate change, whether in the short term (< 5 years) or in the long term (> 20 years). Continuous monitoring of wetland functioning, supported by freely available satellite archives, and new analysis methods based on deep learning provide an opportunity to better understand these processes. In the long term, the Landsat archive and SPOT World Heritage database have been open access since 2008 and 2015 respectively, but their combined use for monitoring wetland function has not yet been explored. In the short term, while Sentinel-2 optical and Sentinel-1 SAR time series have been extensively used to monitor wetlands, thermal data remain under-used, although analysis of the ECOSTRESS time series shows that they have great potential for identifying wetlands and assessing their functioning.
The aim of this thesis is to perform functional monitoring of wetlands throughout the Brittany region (France) using satellite time series to assess their resilience in the context of climate change.
Functional indicators reflecting seasonal variability of water, soil moisture and wetland vegetation will be generated annually at a regional scale from long (50 years) and short (5 years) optical (Landsat-5/8, SPOT-1/5, Sentinel-2), radar (Sentinel-1) and thermal (ECOSTRESS) time series. In situ hydrological and climatic measurements, available in open databases (SIGES, Hydroportail, ONDE, météo France …), will be used to validate these functional indicators. The temporal profiles of the functional indicators will then be analyzed to distinguish the long-term changes associated with climate change from the intrinsic seasonal variability of wetlands. Finally, deep learning (e.g. transforming neural networks) classification of these temporal profiles will be used to estimate the resilience of wetlands to climate change. The results will help to assess the impact of climate change (recurrent droughts, intense rainfall episodes, etc.) on wetland spatial extent and ecosystem services (flood regulation, flood control, carbon storage, biodiversity reservoir, etc.).
This approach is original and its development would position French teams in this still unexplored international field of research. This work requires a conceptual development and its implementation in the Brittany region, which is co-funding this thesis, will represent a pilot project for a more widespread application. This thesis will support the TOSCA ‘RESZH’ project submitted in September 2024 in the framework of the ‘CNES APR’.
This thesis will highlight the value of SPOT World Heritage data, as well as thermal data in preparation for the THRISNA mission, for monitoring natural environments.
The methodological and algorithmic developments resulting from this work could be integrated into the THEIA centre as part of the development of the WETLANDS theme in the CNES Earth Sciences programme.
=================
For more Information about the topics and the co-financial partner (found by the lab !);
contact Directeur de thèse – laurence.hubert@univ-rennes2.fr
Then, prepare a resume, a recent transcript and a reference letter from your M2 supervisor/ engineering school director and you will be ready to apply online before March 14th, 2025 Midnight Paris time !
Profil
Master in remote sensing and environmental sciences (geography, ecology …)
Laboratoire
LETG
Message from PhD team
More details on CNES website : https://cnes.fr/fr/theses-post-doctorats
