Vegetation coverage is an important ecology parameter and used in many climatic and ecological models. Field measurement and remote sensing measurement are rudimental approaches to get vegetation coverage. As far as field measurement, currently common methods include sampling, instruments and visible estimating. Field measurement plays a crucial role in ground vegetation investigation, which provides a background for interpreting and quantifying remote sensing data. However, field measurement has much self-limitation, which does not satisfy exhibiting vegetation features and variation in a large area. Referring to remote sensing measurement, experiential models, sub-pixel models and vegetation indices approaches are three prime methods used for vegetation coverage estimation, which are restricted by some factors such as ground measurement precision and image spatial resolution. Vegetation indices mostly used in estimating vegetation coverage involve DVI,ARVI,ASVI,GEMI,SAVI,MSAVI and SAVI, which have various suitable conditions. Corresponding to different spatial scales, actual remote sensing imagines can be divided into low spatial resolution images such as NOAA/AVHRR and MODIS, middling spatial resolution images such as TM,MSS and SPOT, and high spatial resolution images such as aerial photograph and IKONOS. Remote sensing measurement in grass vegetation coverage has close relation with field measurement data, so consummate design for field measurement is very essential for improving measuring precision of grassland vegetation coverage. Only fast combining these two kinds of data, we are having chances to get perfect grass vegetation coverage measuring results.
This article aims at analyzing and discussing measurement of grassland vegetation coverage, synthetically studying methods of filed measurement and remote sensing measurement, and prospecting possible methods to improve measuring precision of grassland vegetation coverage. Undoubtedly, with the development and mature of sensor technology and various mathematics models, it is possible for us to get remote sensing imagine of high spatial resolution, great spatial scales and credible performance. Digital camera, hyperspectral remote sensing and comprehensive use of multi-scale remote sensing data are possible development trends for improving measurement precision to grassland vegetation coverage. It is true that remote sensing imagine will play more and more important role in studying vegetation characters in the future.