Design And Development Of Production And Service System For Qinghai Lake Basin's Eco-environmental Remote Sensing Products

Qinghai Lake is the largest inland and salt lake of China, which is located in the eastern part of Qinghai Province. Qinghai Lake plays a very important role in regulating regional climate, stabilizing the local ecology and promoting the local economic growth.However, the ecological environment of Qinghai Lake Basin is relatively fragile, and the influence of human activities on the region is constantly increasing. Therefore, it is necessary to monitor the ecological environment of Qinghai Lake Basin.Ecological environment remote sensing products can provide data support and scientific basis for dynamic monitoring of large-scale ecological environment. The inversion algorithms of ecological environment remote sensing products are mainly based on physical model and empirical model. The former is more accurate and has a good universality, but time-consuming. The latter is just the opposite. In order to pursue a higher accuracy, inversion algorithms based on physical model are adopted in this paper. Meanwhile, the time-consuming problem derived from physical model must be solved. Combined with the actual demands of ecological environment monitoring in Qinghai Lake Basin, it is of great practical significance to design and develop a remote sensing product production and service system to provide a variety of ecological environment remote sensing products for ecological environment monitoring of Qinghai Lake Basin. The main work of this paper including the following three aspects:1) This paper analyzes the complex inversion algorithms of six kinds of ecological environment remote sensing products: fractional vegetation cover, canopy chlorophyll content, leaf area index, canopy water content, biomass, and soil moisture. According to the modular programming idea, these algorithms were broken into a number of small function modules and realized using JAVA programming.2) Parallel programs of the six kinds of ecological environment remote sensing products are designed and implemented. For inversion algorithms based on physical model are time-consuming, on the basis of analyzing the production process of the six products deeply, parallel programs of the six kinds of products are designed and implemented according to the principle of parallel programming and the characteristics of MapReduce model in order to improve the production efficiency. The performances of these parallel programs are tested and results show that the production efficiency of the six kinds of products is greatly improved.3) A set of Qinghai Lake Basin ecological environment remote sensing products production and service system based on B/S structure is designed and developed. The system consists of two subsystems: the production subsystem and the service subsystem.The production subsystem integrates the parallel algorithms of the six kinds of ecological environment remote sensing products. It also produces the six kinds of products rapidly and publishes the results to the service subsystem automatically. The service subsystem is responsible to provide ordinary users product services, such as searching, previewing and downloading. Users can use these products for their further applications.