Reserch On Spatial Distribution And Habitat Monitoring Of Oncomelania Snails In Marshlands Of Dongting Lake Area Based On RS And GIS

As the unique intermediate host of schistosome, the geographical distribution of Oncomelania snails determined the spatial distribution of schistosome. At the same time, snails’ distribution closely related to environmental factors. Therefore, as to study on the control of snail and schistosome, it is very significant to understand the relationship between Oncomelania snails’ distribution and environmental factors. Through using RS, GIS, artificial neural network(ANN) and related technique ,we make effective monitoring of snails’spatial distribution and habitat in Dongting lake area based on snails’distribution data in 2004 .In order to have a global analysis for spatial distribution characteristic in Dongting lake area, we construct geographical information system and make a spatial analysis for Oncomelania snails in Dongting lake area,which based on snails conditions as attribute database and 1:50000 digital factor figures for Dongting lake area as structure database. The results show that the distribution of snail have a certain trend and spatial aggregation.most snails, about 77%, distribute by the range of 300m distance from water source.In order to make further analysis of special relationship between Oncomelania snails ’distribution and its environment factors, we extract remote sensing replacement index of marshland snails habitat from Landsat ETM+ image, the results show that 80% normalized difference vegetation index(NDVI) of Oncomelania snails habitat is from 0.061 to 0.278, moisture index is from 0.161 to 0.3999, Other index fluctuating within a certain range.Meanwhile, we analysis the relationship of each environment factor and living Oncomelania snails density, living snail density is positively correlated with NDVI, SAVI and TC-G.Vegetation distribution and type is a indispensable factor in many environment factors, We investigate the application of remote sensing image in vegetation monitoring for snails habitat through using unsupervised classification, maximum likelihood classification and neural network classification. The results show that precision of BP ANN is the highest , reached to 86.5%,kappa coefficient is 0.8070,we can conclude that BP ANN has a better efficient on vegetation classification with remote sensing. After further analysis ,we find that the main vegetation types of snails habitat are reed and nutsedge ,which account for 58% snail habitat.Finally, make use of preceding research result ,we construct remote sensing expert system for snail distribution in Dongting lake area with the support of software ERDAS8.6, which contained qualitative determination for Oncomelania snails habitat distribution and quantitative monitoring of snails’suspected habitat through using remote sensing replacement index of environment factors. The result is in accord with the practical situation, which can monitor distribution of snails effectively.