Research On Spatial Distribution Of Oncomelania Hupensis And The Snail Control Method In Mountainous Regions

Sichuan province is a typical mountainous schistosomiasis-endemic region. Oncomelania hupensis is the only intermediate host snail of Schistosoma japonicum, and the endemic region of schistosomiasis is consistent with the distribution of O. hupensis. Hence it is of important significance to control snail through analyzing the spatial distribution of snail and predicting snail distribution via related environmental factors. Snail control is the important component of the integrated schistosomiasis prevention and control strategy. The environment of snail habitats in mountainous regions is more complex than that in lake type endemics. It is necessary to explore better snail control methods in these areas. In this study, snail distribution, the topographic map and remotely sensed images of Puge county were collected, and the environmental indicators were extracted from the images. Based on the geographic database, the spatial statistics were employed to analyze the spatial characteristics of snail distribution and its relationship with environmental covariates. As well we determined the concentration of niclosamide and observe its dynamic change after soil heaping mixed with niclosamide in sods with water. All of these aim to provide information for the better surveillance and prediction of snail habitats and for snail control and the prevention of schistosomiasis in mountainous regions. The whole study includes three parts.PartⅠSpatial characteristics of distribution of Oncomelania hupensis in mountainous regionsObjective To analyze the spatial characteristics of distribution of Oncomelania hupensis, and to predict the snail distribution in mountainous regions. Methods Snail surveys were carried out in all Schistosomiasis-endemic towns, Puge county, Sichuan province, and the geographic database of snail distribution was established. The nearest neighbor analysis, the spatial autocorrelation analysis and spatial scan statistics were applied to analyze the spatial correlation and clustering of snail distribution. The variogram model was built to describe the spatial variation of snail distribution and then ordinary Kriging model was constructed to predict the snail distribution in the overall Schistosomiasis-endemics in Puge county. Results There was a spatial clustering model for the distribution of snail habitat locations. With regard to the rate of frame with snails, the global Moran's I indicator was 0.095(P<0.05), and the General G indicator was 0.067(P=0.405). The local spatial autocorrelation analysis showed that there were 28 snail habitats with statistically significant LISA value(P<0.05), among which existed high-high, low-low, low-high and high-low four types of correlation model. The spatial scan statistics totally detected 24 snail habitat clusters (P<0.05), including 14 high rate clusters and 10 low rate clusters, and the result was similar to that of LISA analysis. The spatial variation of snail distribution fitted in with the spherical model, and the spatial autocorrelation changed with the distance when h was shorter than 0.2542. The cross-over validation showed that the ordinary Kriging model achieved good unbiasedness and optimality.Conclusion There were spatial autocorrelation and clustering of snail distribution in mountainous regions, meanwhile existed spatial heterogeneity of snail distribution. The spatial variation of snail distribution fitted the spherical model. The Kriging modeling could better predict the spatial distribution of snails.Part II Spatial regression analysis to predict the distribution of Oncomelania hupensis in mountainous regionsObjective To predict the snail distribution by employing the spatial autoregressive model. Methods Snail surveys were carried out, and the remotely sensed images were collected. Related environmental factors were derived from a Landsat 7 ETM+image, and the relationship between environmental covariates and the snail density was analyzed by correlation analysis, ordinary linear regression model and spatial autoregressive model. Results Descriptive analysis found that all environmental factors in snail habitats varied in the small range. Spearman rank correlation analysis indicated that snail density was positively correlated with normalized difference vegetation index(NDVI) and moisture index(MI)(P<0.05), negatively correlated with LST and elevation but with no statistical significance (P>0.05). The linear regression model was performed, and the residuals of this model were spatially autocorrelated through spatial dependence diagnosing (Moran's 1=0.242, P<0.05). So it was not appropriate to use the linear regression model for the analysis. Lagrange multiplier statistics indicated that the spatial lag model (SLM) should be employed. The spatial parameter p of SLM model was highly significant (ρ=0.249, P0.05). After five months, niclosamide still could be determined in groups with dosage of 4g/m2 and the larger. The mortality rate of snails decreased as the concentration of niclosamide decreased(P<0.05). After 5 months, in the group of 4 g/m2 and the larger dosage, the 3d,7d mortality rate of snails were 5.33% and 9.33% in the surface soil layer, significantly higher than that of the control group(P<0.05). Conclusions The heaping method was an efficacious measure of controlling snails and its recommended dosage was 4g/m2.