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

Accueil du site → Master → Chine → 2006 → The Study of the Application of MODIS Data for Monitoring Desert Steppe Biomass

Chinese Academy of Agricultural Sciences (2006)

The Study of the Application of MODIS Data for Monitoring Desert Steppe Biomass

何涛; He Tao

Titre : The Study of the Application of MODIS Data for Monitoring Desert Steppe Biomass

Auteur : 何涛; He Tao

Grade : Master’s Theses 2006

Université : Chinese Academy of Agricultural Sciences

Résumé
The warm desert steppe mainly distributes on climate arid grassland district. Its area is large and distributes widely, and it plays an very important part in the grassland stockbreeding economy of our country. The warm desert steppe becomes the homes of a lot of property stability, the breed fine, rare , characterized animals and plants, and plays an important role in the northern people’s lives and meadow ecosystem at the same time. In order to monitor the desert grassland vegetation growing situation, estimate the biomass on the ground, and provide the basis which dynamically monitors the balance of domestic animal and the grass, this text is based on science of grassland, environmental science, and chooses the new generation satellite remote sensing datum of MODIS (the Moderate Resolution Imaging Spectroradiometer), and combine the ground census data, regard " 3S " (remote sensing , geographical information system , global positioning system ) comprehensive research such as the technology, statistical analysis of mathematics and physics as the means. The article makes three inferior kinds of the warm desert steppe as the typical research district, and develops the model of assessing separately.The inferior kinds of the desert steppe of plain and hills, sand ground lie in the central and west regions of the Inner Mongolia, the inferior kinds of the desert steppe of mountain lies within the boundaries of Xinjiang. First of all, this research has dealt with covering MODIS remote sensing image of studying the area, including bow-tie correction of MODIS data, geometry correcting. Secondly, according to forefathers’ research results, this article chooses the Normalized Difference Vegetation Index, soil regulate vegetation index, soil of revision regulate vegetation index , ratio vegetation index as vegetation index that research employs. To calculate the vegetation index file from MODIS images which were projected, compare vegetation index and compose the image day by day, make vegetation index large time phases formate and pursue. Pick up vegetation index of the point, according to longitude and latitude.The article analyzes the correlation of different vegetation indexes and biomass. According to three inferior kind of desert steppe, the article separately develops the estimate model which is different curve type. And the results survey that the regression equation of parabola form which use SAVI as independent variable has the highest fitting accuracy. The regression model of fitting accuracy of the inferior kind of the desert steppe of plain and hills, sand ground, and mountain is 0.627 , 0.757 , 0.764 respectively. And the precision of three model is 75.93%,68.57%,72.38%. For the investigation of large area desert steppe, three kinds of models basically reach the requirement of the macroscopic monitors.Set up the yield picture which is based on the MODIS-SAVI picture of the inferior kind of the desert steppe of plain and hills, sand ground, and mountain. According to yield level, the picture is divided into several grades. The study show that the yield level distribution of the study area is basically consistent with the precipitation change of the study area.

Mots clés : The desert steppe; biomass; The MODIS image; Vegetation index; The yield estimating model;

Présentation (CNKI)

Page publiée le 16 mai 2013, mise à jour le 14 octobre 2017