Explore New Methods, A Monitoring Typical Steppe Grassland Degradation

It is very significant for conservation of grassland ecosystem and sustainable development of the local agriculture and stock production precisely monitoring dynamic grassland degradation or health condition on large-scale order. Comparing to traditional community research methods, remote sensing technology has undoubted superiority for monitoring vegetation condition in the large area. Remote sensing technology has widely adopted in the research field of monitoring vegetation cover change in the past years. This paper systematically studies and reviews the past research concern and cases for grazing and grassland degradation without using remote sensing technology, introduced the basic theory, main methods (vegetation index) and involved research progress for remote sensing application of vegetation study. Particularly, this paper illustrates that grass vegetation degradation or health condition consists of several independent layers. Monitoring vegetation change in the past mainly depended on some vegetation index, especially NDVI that is just supposed to sensitive to the certain layer梘ross layer. We propose a new method to monitor grassland vegetation health condition combining community plot survey and remote sensing technology after lots of field survey works in Xilin River basin, Inner Mongolia, where the grassland degradation is very typical. Three specific Principal Component (PCs) with specific ecological meaning are extracted from a 12-variables data set that contains community information using principal component analysis (PCA). Based on these three PCs, we propose a new vegetation index桮HI, which is proved to be more qualified for monitoring grassland vegetation health condition, and sensitive to degradation~ From 6-band vegetation spectral reflection data, we extract perfectly?two PCs: visible light component and infrared light component. The PCs that indicate community gross and grazing degradation, as well as GHI, are correlated to plot spectral reflection data, and we get the regression model of GHI and visible light/infrared light. We apply this model when dealing with TM data, and get the GHI image. Compared with NDVI image on the same TM data, we find out GHI has drastic effect for indicating human disturbed influence on grassland, and is superior to NDVI. Moreover, GHI image can illustrate vegetation pattern, especially patches structure, better than other image. Applying GHI to the historical TM data, vegetation cover change and change mode of graze in the research region are also discussed in this paper.