Tychon, Bernard ; Université de Liège - ULiège > Département des sciences et gestion de l'environnement > Département des sciences et gestion de l'environnement
Knowledge-based approaches for degradation mapping in arid rangeland
Alternative titles :
[en] Utilisation d'un système expert pour la cartographie de la dégradation des parcours arides
Publication date :
28 September 2010
Number of pages :
A0
Event name :
International Conferense on ADVANCED SIENTIFIC TOOLS FOR DESERITIFICATION POLICY, IFAD, ROME
Event organizer :
IFAD, UNCCD, DeSURVEY project, The glocal mechanism, CSIC Agencia Estatal Consejo Superior de Investigaciones Científicas, Spain
Event place :
Rome, Italy
Event date :
28 et 29 septembre 2010
Audience :
International
Name of the research project :
Desurvey project
Commentary :
Rangeland covers about 82 % of the arid lands of Morocco. It is largely recognized that these areas are threatened by the desertification. It’s associated with biodiversity loss and contrib¬utes to global climate change. Fighting against desertification involves having an accurate knowledge on a current land degradation status. Remote Sensing is a widely used technique for mapping land degradation. However, arid rangelands are difficult to map accurately from satellite data by means of parametric classification because of the spectral similarity between soil and several perennial vegetation and sparse vegetation contributes little to total radiation detected by satellite sensors.
The objective of this study was to assess the spatial extent and severity of the rangeland degradation in the highlands of eastern Morocco, using a Knowledge-based approach.
The Knowledge Engineer of Erdas imagine was used to build a knowledge base. It’s based on the combination of data obtained by Landsat TM satellite images of summer and spring 2007, field investigation and other ancillary data mainly digital elevation model and climatic data. The different degrees of degradation were evaluated using several indicators such as vegetation cover, plant diversity, bare soil area, sand dune localisation, overgrazing and cultivation of rangeland.
The results indicate that the total degraded area is about 25000 km2, making up 74% of the land in the study area. Degraded lands were classified into four levels: slight, moderate, severe and very severe.
This study reveals that the combination of remote sensing data and Knowledge base provides a high accuracy level in mapping degradation of arid rangeland.
