Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Chine → 2007 → Research on Desertification Assessment in Naiman Kerqin Based on Remote Sensing

Northeast Forestry University (2007)

Research on Desertification Assessment in Naiman Kerqin Based on Remote Sensing

Li Ji Hong

Titre : Research on Desertification Assessment in Naiman Kerqin Based on Remote Sensing

Auteur : Li Ji Hong

Grade : Doctoral Dissertation 2007

Université : Northeast Forestry University

The study is the part of State Natural Science Funds project"Research on quantitative retrieval of the main factors of desertification and assessment of remote sensing information"(No.30371192).For the purpose of construct a desertification hazard assessment model to remote sensing quantitative monitoring, which can monitor and evaluate the potential desertification hazard situation in Inner Mongolia kerqin sandy areas quantitatively,we did a comprehensive analysis on the TM、MODIS and twice measured data from study area. Having used radioactive transfer model, linear spectrum mixed model and the integration of remote sensing data, We retrieved surface temperature what desertification evaluation needs, soil water content, leaf area index, vegetation index and the percentage distribution of sand on the base of remote sensing data processing. The research gets desertification evaluation factors such as soil texture from visual interpretation. Turn right on quantitative evaluation of kerqin sandy desertification degree ; we expect to make progress on quantitative remote sensing technology and innovation research methods of desertification research.Following are the content and result of this research:1. We established a suitable potential desertification risk evaluation index system for kerqin on the basis of the desertification monitoring and evaluation indicators identified by the system established by scholars domestic and foreign. Vegetation indicators, soil indicators and people pressure indicators were selected to evaluate the extent of desertification. It contains six desertification evaluation factors such as the soil moisture, leaf area index, vegetation index, soil texture, occupation ratio of bare sandy and population density.2. This paper adopts split window methods to build the Remote Sensing information model of the rate of soil water in Naiman Inner Mongolia, the RS information model based on the earth’s surface temperature. And it’s precision is 77.28%. Many factors such as the vegetation and the landform affect the land surface temperature, so the retrieval model based on the surface temperature is adapt to the plat, less vegetational or vegetationless regions. So terrain correction is important for it.3. Naiman EVI imaging data is reduced with CV-MVC after MODIS data processing. By comparing the vegetation index(Ⅵ) of TM, MODIS and image after integrating to those of measured data, it showed that TM and MODIS Vegetation Index are similar, but lower than the measured data. After integrating of both image, vegetation index data and ground spectra are closer to each other.4. Occupation ratio of bare sandy is one of desertification evalution important factors, which reflects desertification extent directly. We got occupation ratio of bare sandy with linear spectrum mixed model(LSMM) and the precision is about 75.13%5. After the quantitative retrieve to several important desertification remote sensing factors, we notes some evaluation factors such as soil texture and land use patterns still need to access through the combination of visual interpretation and classification. We used Brovey method integrate the two data together given the precision of TM, MODIS and visual interpretation. Fused image’s spatial resolution is 17.98 meters ; spectral resolution also reached nm level. The spatial and spectral resolution has been improved quite a lot.6. We evaluated Naiman potential desertification extent and made conclusions. Of which : severe desertification region accounted for 23.02%, moderate desertification region accounted for 19.3%, mildly desertification region accounted for 47.68%, potential desertification region accounted for 10%.The precision of the model was calculated by selecting 60 sample data. The result shows that the precision of the desertification evaluation is 91.7%.

Mots clés : Desertification Assessment Remote Sensing Naiman Quantitative Retrieval

Présentation (Dissertationtopic)

Page publiée le 4 mai 2013, mise à jour le 25 avril 2018