Left River Valley Plant Climatic Productivity Of Remote Sensing Monitoring Methods

Left River Basin as a unique natural geographical unit, has a special structure of the natural environment, diverse ecosystem types and species diversity of unique, but it has been ignored. The scientific community has little research in this area. At present the national focus on the Development of Guangxi, this research made remote sensing of environment for the Left River Basin, thereby promoting the economic development of the left river basin to Guangxi's economic development, building the North Bay Economic Zone, for further developing the western region and lay a good basis. This paper studies the domestic and foreign remote sensing monitoring the progress of research content, development and the basis of analyzing problems, aiming at the left River Basin Ecosystem characteristics and existing problems, an integrated multi-source remote sensing, and carried out the left river basin ecology and productivity of monitoring techniques to explore. Main activities include: use of Landsat TM images and MODIS images as the information source, the Left River Basin as a test area, carried out research classification in remote sensing monitoring methods and analysis of remote sensing interpretation; using MODIS NDVI time series data as the main data, conduct left River Basin vegetation classification studies; using CASA model, MODIS remote sensing data and ground meteorological data integration methods, carry out Net Primary Productivity (NPP) for quantitative estimation in left River Basin; MODIS13Q1 data source for information from2004 to 2008, research MODIS time series data against the refined methods, including cloud and the noise of lowering the effective methods, again based on the use of linear trend and France decomposition method, analysis vegetation trends and spatial pattern. Through a series of studies, achieved the following results:(1) In the comparative analysis of the supervised classification, unsupervised classification, decision tree classification, and artificial intelligence neural network method and other methods, the use of supervised classification and unsupervised classification method of combining, on the left river basin object classes were analyzed to be left River Basin land feature in the spatial layout.(2) The use of process-analysis-based CASA (Carnegie-Ames-Stanford Approach) model, multi-temporal MODIS data, the method of combining weather information, comprehensive consideration of vegetation photosynthesis, temperature, radiation, water and different vegetation types the largest differences between light use efficiency based on the left of the river basin is located in the region for remote sensing of vegetation net primary productivity of quantitative estimation. Zuojiang River Basin region in 2008 net primary productivity changes with the seasons obvious seasonal characteristics, the year in the winter (December -2 months) vegetation NPP value is the lowest in the south and the north-western forest vegetation NPP On average, less than 100 gC.m2.M-1, the average time of NPP accounted for 0.78% of average annual NPP, from April to October of the left upper region of the NPP River Basin is the largest average, about year 85.39%(3) Data from 2004 to 2008 MODIS13Q1 as the main information source, using SG filtering method to study the NDVI time series data for the refinement of issues, including the noise of lowering clouds and effective way to extract the effective vegetation indices EVI trend. On this basis, the use of change vector method, linear trend and so on, on the left river valley vegetation cover change in intensity, trends and spatial pattern of in-depth analysis. From the analysis of geographical space, the left river basin covers eight counties, the greatest changes in vegetation index which EVI is Shangsi, EVI vegetation index vector changes in the slope of the minimum is -18.30, the maximum was 28.30, indicating the growth of vegetation condition Shangsi large regional differences, the specific performance in the east, the western region is relatively stable vegetation growth. EVI vegetation index in the smallest state for long, the minimum is -10.33, the maximum was 18.72, Longzhou more balanced growth changes within the vegetation.