Research On The Tropical Cyclone Based On Spatial Analysis In GIS

The frequency of Tropical Cyclone on Northwest pacific is the highest in the world where suffered it, which occupied above 1/3. China is one of the most serious and frequent countries which suffered the Tropical Cyclone. Thus, it is meaningful for us to research where Tropical Cyclone occurred and how it grew, to forecast its path in order to prevent the disaster. We attach more importance to the research on it.People constantly explore the new technology and method in the application of TC forecasting and affecting analysis for its complexity. In this article, the author explores TC forecasting based on GIS spatial analysis and spatial interpolating about TC process precipitation, advances two methods: one is TC forecasting method based on Area-index similarity and its arithmetic; the other is data and spatial interpolating in the application of TC precipitation based on protruding polygon and Voronoi.It is always a hard problem to accurately forecast the TC path, thus meteorologists established all kinds of forecasting methods include dynamical modelx statistic model and dynamical-statistic model etc. The forecasting based on similarity is an important technology, thus similarity forecasting is explored based on GIS. We extend the geography similitude with geometry similitude and advance the area-index path similitude measurement that shows the similitude between history path and real path with the ratio between the area by history path and real path and reference area, and then select the forward n TC path according to similitude to build the forecasting model. The examination shows this model is helpful to the normal path.TC always leads to strongly raining. We need use spatial interpolating in research process precipitation and its effect. Now there are many research on spatial interpolating and this tool is supplied by GIS soft such as ArcGIS which has Kring% IDWx spline function etc. But these methods exist some problem: 1) the specialty of TC precipitation is distinct spatial extent because between the precipitation area and non-precipitation area there is an obvious breadline, so it is need to confirm the zero-value boundary except GIS soft. Otherwise it shows the incorrectprecipitation area, such as some non-precipitation area would be considered as precipitation area. We called the problem as the first data integrality problem. 2) During establishing database and analyzing data, we found there were some stations with on precipitation record existing in precipitation area. We couldn' t distinguish whether precipitation occurred in these station or not according data. We called this problem as the second data integrality problem.To show incorrectly process precipitation distribute, we advance the method to deal with the two problem: resolve the first data integrality problem with Voronoi and GIS neighbor analysis and the second data integrality problem with protruding polygon and maxmin neighbor-distance. Thus we obtain the complete data including non-precipitation station beside the precipitation area and use these data in interpolating.We develop the tool about TC forecasting based on area-index similarity measurement and the analysis of interpolating in TC process precipitation area and visualization.