Science Service System
You are here : Home : Proposals_Summary

Summary of Proposal LAN3114

TitleContribution of very high resolution terraSAR-X data for studying wooded ecological corridors.
Investigator HUBERT-MOY, Laurence - LETG Rennes COSTEL- Université Rennes 2, Geography
Team Member
Professor Pottier, Eric - IETR - Univ. Rennes 1, Electrical Engineering
Research Director Burel, Francoise - UMR Ecobio of Rennes - Univ. Rennes 1, Ecology
Reseach Director Baudry, Jacques - INRA SAD Paysage, Ecology
Professor Hubert-Moy, Laurence - LETG Rennes COSTEL - Univ. Rennes 2, Geography
SummaryHabitat fragmentation is considered as one of the most serious threats to biodiversity thus a major cause of the current extinction crisis. In fragmented landscapes, movement is a key process for the survival of plant and animal populations; landscape feature easing movement improves landscape connectivity. Among the elements that increase landscape connectivity, corridors are particularly interesting and effective. Many national and international policies promote ecological corridors for biodiversity protection. Thus, ecological corridors are often considered as a management tool in landscape planning and development policies. Deciphering them among landscape features and structure is of utmost importance. The vegetation structure of the different land cover types is a good surrogate of the quality as habitat for plants and animals. Few ecological studies use remote sensing data to assess the biophysical or structural properties of vegetation to understand species distribution. To date, most studies use optical images, while synthetic aperture radar (SAR) images are under-exploited for ecological applications. Indeed very few radar images have been used to study ecological corridors. We have evaluated TerraSAR-X images in a farming area with a quite sparse hedgerow network to extract wooded ecological corridors (Betbeder et al., 2014) and assessed the habitat quality using information derived from the TerraSAR-X images (Betbeder et al. 2015). The results show that hedgerow structure and hedgerow network patterns can be identified with a very good accuracy and also reveal the high correlation between one polarimetric parameter, the Shannon entropy, and the canopy structure measured in the field. Therefore, TerraSAR-X images can both precisely detect the presence of wooded hedgerow networks and characterize their structure, which cannot be achieved with optical images. Therefore, in comparison with passive optical remotely sensed data, the TerraSAR-X data on vegetation structure provide a better understanding of biodiversity patterns. The internal structure of hedgerow derived from the SAR image is a strong predictor of the abundance of the studied model species -forest carabid beetles- at two scales i.e. a local one and i.e. a landscape one. In this study, we aim at (i) evaluating the reproducibility of this method in different wooded landscape configurations, ranging from sparse to dense hedgerow networks and (ii) assessing the functionality of ecological corridors in integrating information derived from TerraSAR-X images in corridor connectivity model. For that purpose, polarization and polarimetric radar parameters will be firstly derived from radar images to extract landscape elements. More specifically, we will (1) evaluate the ability of TerraSAR-X images for habitat structure and hedgerow network mapping in different landscape configurations; (2) characterize the landscape connectivity with maps derived from the radar data using connectivity metrics, and (3) test the different metrics derived from the radar data to explain the spatial distribution of the studied species model.

Back to list of proposals

© DLR 2004-2016