Investigation in Estimating Vegetation Cover and Phytomass Production, Using Enhanced Landsat Data in A semi arid Region

The objective of this research is to survey the capability of ETM+ dedector s data in estimating
vegetation cover and phytomass production in the semi arid rangeland of Hanna Research Station in Isfahan Province. Initially the percentage of total vegetation, grass, forb cover and phytomass were estimated using 450 plots (1x1m) in 45 randomly chosenclusters. Global Positioning System (GPS) was employed to determine and adjust the sampling unit positions in corresponding unit characters in the image. After georefrencing the Thematic Mapper Image, dated June 2001, by 1:25000 digital topographic maps (dgn), developed by National Cartographic Center with RMS error of less than 0.5 pixel size, the ordinary vegetation indexes and soil line vegetation indexes as well as hybrid bands were evaluated through HIS and SR methods. Then by vector layer of sampling clusters, the digital number of indexes as well as origin bands in each cluster were extracted. The results of simple linear regression analysis between indexes (as independent variable) and biophysical parameters (as dependent variable) through SPSS software show that there is a significant relation ship between vegetation indexes and biophysical parameters. According to these results, the fourth original band of ETM+ has the highest Correlation coefficient (r = 0.66 and SE=4) with total
vegetation cover.
Likewise, ETM4 and VNIR2 indices with r= 056, SE=1.8 and 3.2 show highest relations with forb cover, while the fifth original band ofETM5 with r = 0.6 and SE=2.2 show the highest correlation with grass cover. As regards production the fourth band of fusion exhibits the highest r (052) with SE (6.3).According to the results in this research vegetation cover and production could be estimated through ETM4 , VNIR2 index and HYB4.