analysis of the climate drought and its effects on spatial patterns of location in rural settlement (case study villages in mazandaran province)

1. introduction drought is one of the most complex problems not only in the management of water resources, but also creates economic and social issues . but the problem definition, identification and prediction of unresolved. the occurrence of drought can be caused by several factors, mainly rooted in the general circulation of the atmosphere and space physics, and earth's energy balance is sought. continuous and prolonged drought and water shortages said. drought indices to evaluate various aspects of the invention are wipe the index is often based on a measure of deviation from the mean long-term rainfall amounts over a specified period of time is based. .one of the most effective profiles, standardized precipitation index spi the first time by mackay (1993 ) was used . feature of this index is that it allows the researcher of drought and rain events occurred for each stage of defining the desired time. the drought has five main features are as follows: beginning and end of the period of persistence, intensity, frequency and scope. .these features are related to the drought is examined in each study. 2. material and methods the research and application of analytical and statistical is an approach to governing. the present study is based on annual rainfall of 22 -year drought in the province from 1988 to 2010 was performed. for this purpose, using data from the survey of precipitation is 22 synoptic stations and rain. because some data are incomplete meteorological stations using the ratio method, incomplete statistics were estimated . the zoning of the models and methods used in kriging idw the fitting includes three power idw 1, 2,3 , and kriging with spherical model , circular , exponential , gaussian and is and the use of radial basis function rbfdone . also for spatial analysis and modeling of spatial autocorrelation moran drought indicators and statistics will be used for general gto the spatial pattern of drought and rural location in the province mazandaran of space analysis placed. 3. result and discussion based on interpolation models were extracted from the optimized model for zoning drought province spline model is quite regular . studies show that a very dry area 284,208.5 square kilometers, 704,854 square kilometers little dry, dry, normal 483,280.5 kilometers, 535,492.1 kilometers normal wet and very wet zone and framrtvb a total of 210,837.01 square kilometers have been allocated to the results of application moran index on the spatial distribution of drought, this coefficient is positive and equal moran`s i = 0.994307 good indicator of the spatial distribution of drought . because the ratio of the number 1 indicates the focus tends to be higher. since the z score = 6.9462 calculated sig.i = 0.01 confidence level greater than the expected value (ei = 2.58) is, therefore, welcome the spatial distribution of precipitation moran coefficient is confirmed. welcome to the spatial distribution of rainfall as a result of the moran coefficient is confirmed. general g statistic analysis results show that parts of west and west coast and parts of the province is a hot area, the z score of 6.518083 , which is significant at a confidence level of 01/0 .from the results obtained , the correlation matrix of the table , we can say that the level of trust between area clusters 0/01 hot and cold , drought and rural location, there mazandaran. however, this amount is equal to 0/914 direct (positive) population. 4. conclusion the results are quite regular completely regularized spline determine the best model 0/45. the results of the application of moran's index of spatial distribution of drought, the coefficient is positive and equal moran's = 0/99 representing the cluster spatial distribution of drought. according to moran index can identify a variety of statistical analyzes using the general classification of spatial patterns should g this shortcoming was capped. general statistics g analysis results show that drought areas located on the other side of the cluster is focusing on drought enjoy. the area of each cluster of villages was calculated based on. calculations show very hot clusters about% 35/21 of the land area is just over 753 villages. slightly cool to slightly warm the total area covers 1,245 villages while clustering is % 37/17. the results indicate that the correlation between the number of clusters and villages confidence level of 99 /0% error level 004 / 0, which is a direct relationship, but much weaker .with increasing cluster size is relatively low added.