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

Accueil du site → Master → Chine → 2004 → Studies Of Arid Land Classifacation And Lake Monitoring Based On MODIS Remote Data

Fujian Agriculture and Forestry University (2004)

Studies Of Arid Land Classifacation And Lake Monitoring Based On MODIS Remote Data

刘惠英; Liu Hui Ying

Titre : Studies of Arid Land Classifacation and Lake Monitoring Based on MODIS Remote Data

Auteur : 刘惠英; Liu Hui Ying

Grade : Master’s Theses 2004

Université : Fujian Agriculture and Forestry University

The Moderate Resolution Imaging Spectroradiometer (MODIS), flown on board the Terra Earth Observing System (EOS) platform launched in December 1999, produces a land covered change product. This product is expected to be of better quality than LCC (Land Cover Change) products based on operational satellites (notably GOES and AVHRR), due both to improved spectral resolution and higher spatial resolution of the MODIS instrument .The plans for the MODIS instrument include the generation of improved global data sets. Vegetation Indices are simple, robust, empirical measures of vegetation activity at the land’s surface. They are widely utilized for global monitoring of both spatial and temporal variations in vegetation at high precision. The MODIS VI products will improve upon currently available indices and will be used to more accurately monitor and detect changes in the state and condition of the Earth’s vegetative cover. Vegetation indices will also play a role, as ’intermediaries’ in the derivation of vegetation biophysical parameters, such as leaf area index (LAI), percent green cover, and fractional absorption of photosynthetically active radiation (fAPAR).MODIS as one of sensors of TERRA, begin numerous to pay close attention to the thing as soon as lifting off. Its preferential policy and one’s own wave band characteristic used free, covered and studied ecological environmental monitoring , climatic change analysis and the earth’s surface which launch the whole world and area by the change, the past NOAA-AVHRR that use there are more advantages relatively. EOS/MODIS data turn of the century most attractive star year datum source one of in the world, its extensive regional coverage rate and resolution ratio of good time make it have wide prospect in using. Therefore a lot of MODIS ground that country set up in succession accept and stand. The study on MODIS of our country utilizes beginning in recent years, it is less to utilize on its, mostly still at the starting stage. The form of the foundation EOS/MODIS data of this text and wave band characteristic, have described MODIS data source, using current situation at home and abroad at present in detail . Utilizing stand datum and datum in stock of IB of MODIS, got to accept according to the time slot every 8 days, supported by remote sensing software, selecting the data through the data , form changing, concatenation, projection again, a series of materials pretreatment procedure of cutting out etc., is it study to can use and assign to MODIS remote sensing data of the wave band area to get, utilize number one ( red wave band ) its , second wave band (near and infrared) formate space resolution ratio 250 meters, the time resolution ratio is NDVI image of 8 days. Adopt NDVI many time phase maximum synthetic methods (MVC ),produce the area NDVI data products of Xinjiang of the corresponding resolution ratio, choose it as from 2001 to 2003 time slot, formate 138 area NDVI products of Xinjiang altogether. What has been done the following two respects to NDVI image that is handled well is studied:1.Utilize the image to superpose the method of analysing with the main composition etc. in ERDAS, superpose the image in January - December of 2003, produce NDVI time series image ; Utilizing this time sequence image to draw out 2003 Xinjiang NDVI first main composition ; using unsupervise NDVI first main composition to classify in ISODATA algorithm ; Combine 1:400 ten thousand vegetation type vectograph with DEM datum classify to combine ; The categorised resul : is good finally.2.As Xinjiang ground form have model arid characteristic of district, for studying arid district environment it have good representativeness to take the place of to perform. So, utilizing the categorised result to the water body while classifying above, a model tothe studying area is clicked—Ebinur lake. Having used the time series analyticalmethod of the array finally, NDVI of resolution ratio 250 used for three years in succession formates the data, have made time sequence analysis.

Mots clés : modis remote sensing NDVI; land cover classification; Xinjiang; Ebinur lake; time series; lake monitoring;

Présentation (CNKI)

Page publiée le 19 avril 2013, mise à jour le 14 septembre 2017