Fuzzy Pattern Recognition Of Lake Trophic Status Base On Remote Sensing Technology

Lake eutrophication is one of global environment problems that human beings are facing with. In recent years, while the disturbance of climate warming and human activities, they make the eutrophic status of lakes located in Inner Mongolia and Sinkiang plateau area became worse and worse. The bad eutrophic status of lake has directly threaten on the fragile ecosystem and the life style of people around the lake. Frequent monitoring and in-tegrated management for lakes are necessary. However, conventional methods for water quality monitoring in lakes which are time & money-consumed can't provide water quality information of the whole lake. With spatial and temporal information,remote sensing technique is suitable to obtain such information quickly and frequently, and it'll be widely applied in water quality monitoring of lakes in future because of its advantage of fastness,economy and long-term dynamic monitoring.Based on the analysis of former theory, techniques and methods in water quality monitoring via remote sensing. This study aimed at developing and applying an operative method to estimate key water quality parameters related to assess the eutrophic status of lake,such as total N(TN),total P(TP),chlorophyll-a(chl-a),Chemical oxygen de-mand(COD) and Seechi depth(SD) using multi-spectral remote sensing data. Wuliangsuhai Lake, China was chosen as a case study, it's Landsat5 TM data from May to Oct, 2006, and the synchronous or quasi-synchronous observation on satellite and land was adopted in study. Some study issues involved in applying multi-spectral satellite imagery to water quality of lake, including atmospheric correction,methods for multi-spectral bands analy-sis were discussed.This paper focused on two key issues: one issue is the development and applications of the remote sensing retrieval models of such 5 water quality parameters base on the ra-dial basis function(RBF) network; another is assessment of the eutrophic status of lake us-ing concentration of water quality parameters retrieved from remote sensing data based on the cross iteration fuzzy pattern recognition model.The results indicated that Landsat5 TM band1～band4 data were suitable for retriev-ing of water quality parameters of lake, and comparing with single band, band ratio com-binations have better correlation with water quality parameters of lake; Mean errors of re-mote sensing retrieval models developed by using the radial basis function(RBF) network are less than±25%; The eutrophic status of the whole Wuliangsuhai lake recognized by the cross iteration fuzzy pattern recognition model match with the observed spatial trends of eutrophication.At the end of the papers, author summarizes the research contents and the achieve-ments and discusses the problem existing in the study process.