In china, especially in the North and Northwest, many cities suffer sand-dust or sand-storm attacks in winter and spring. There are two sand sources forming the bad weather, local sand source and other source out of local area. The second kind source needs state level activities to control sand movement and recover local ecological environment. For the local government, it should pay much attention to decreasing its inner sand & dust source, because the sand-stormy or sand-dusty weather causing by the local sand source usually comes abruptly and brings much damage frequently, and it is also hard to forecast it. However, people always emphasize all-year bare land controlling and pay less attention to seasonal bare land (especially in winter and spring seasons) which caused by unreasonable agricultural pattern. In this paper, taking Beijing as the test area, using MODIS vegetation index time-series data, all-year and seasonal bare land had been classified. The data set used was 16-day composited EVI time-series with a 250m spatial resolution. After filtered and reconstructed, this paper applied the parallelepiped classification algorithm to the data set, and emphasized the all-year bare land with lower EVI value and seasonal bare land with lower EVI value just in spring and winter. Taking accord of local terrain, infield was the main part of seasonal bare land. The experiment result showed that all-year bare land mainly distributes in northwest Beijing, the joint area of Beijing, Shanxi and Inner Mongolia, especially in Inner Mongolia. Seasonal bare land mainly distributes in northwest and west Beijing, gathering in northwest Hebei, east Shanxi; there are also some ones in Daxing, fangshan, changpin, yanqing, miyun, shunyi, and tongzhou areas of Beijing. These two kind bare lands were all possible contributor to sand storm or sand dust weather in Beijing. Considering wind direction and terrain information of Beijing area, some possible sand source could be found. This paper emphasized the research of seasonal bare land, which is easily neglected in common work for sand and dust prevention. The paper considered that much attention should be paid to change the original cultivation pattern through the reasonable guidance, to realize the balance between the economic efficiency and environment benefit, and finally, reduce the local sand dust harm gradually by such development strategy. For the governors in north China, it is also essential to take accord of this factor when designing city development plan.
Crop leaf water content can be a valuable biochemical parameter to diagnose crop water stress. The leaf water content
characterizes some spectral absorption features in NIR band. Some researchers have proved that it was feasible to
retrieve leaf water content utilizing those spectral absorption features. Measured leaf water content data of several sorts
of winter wheat and the corresponding reflectance, CHRIS images were collected. Even 974 nm, 1160 nm and 1440 nm
were absorption feature bands of foliar water, the predicted accuracies of leaf water content only using these bands were
not satisfying. Four popular indices of vegetation water content including NDWI, SR, WI and REP were used to build
the predict model and evaluated though relativity analysis. SR and REP opposed the stronger predicted accuracy of leaf
water content than other spectral indices. Limited to the band position setting and band spectral function of CHRIS, SR
and REP were modified fit to the band setting of CHRIS. The predicted model based on ground reflectance were made to
adaptable to the band spectral function of CHRIS sensors, and the results showed that wide band spectral function indeed
caused a lower accuracy of crop water content than narrow band spectral function.