Font Size: a A A

Study On Remote Sensing Quantitative Evaluation For Nenjiang Desertification Land

Posted on:2009-02-08Degree:MasterType:Thesis
Country:ChinaCandidate:L J LiuFull Text:PDF
GTID:2143360275967021Subject:Forest management
Abstract/Summary:PDF Full Text Request
Desertification is a serious environmental question at present. Duerbote mongolian nationality autonomous county is the only integrated desert county, the dominating type of desertification is wind-corroded. On April, May, September and October, sand and wind is frequent. Damaged by them, Meadow degraded year after year. For Keeping within limits the expand trend of desertification, correctly evaluats for deserted actuality, and makes the relevant measures, then improves the environment of desertification area are in urgent.The main researching conclusion and creative point are made from this paper:Liner spectral mixing model (LSMM) retrivals the occupation ratio of bare sand land, non-density model retrivals vegetation coverage, and then make correlation analysis with tasseled cap of the first and second variable. The result indicates the precision of retrival is good. Multiple regression analysis makes the biomass estimate model and the result is prominent, and R~2 is 0.70003. Then independent test is made. Predicted precision is 79.0067%, and meet this study need. Using improved wavelet transform fusion method based on IHS transform fuses multiple spectral and panchromatic images, and picks up the training area of soil texture and earth's surface status. Through the subjective and objective evaluation, the eye interpretation effect of fused image improves greatly, and increases the accuracy of training choice. The retrival result of five factors establishes the visual RS information model. Using this model to evaluate, the precision is 88.75%.Then the evaluated results enter GIS in raster and make thematic map to calculate acreage. The result indicates: slight regions account for 57.6%(the water is 13.23%),middle regions are 21.77%, and severe and extreme regions occupy 7.4%.All these show the desertification extent of duerbote mongolian nationality autonomous county is slight in total, the father measures win initial success.
Keywords/Search Tags:Desertification Evaluation, Remote Sensing, Retrival Evaluated Factors, Wavelet Fusion
PDF Full Text Request
Related items