Yunes Khosravi; Hassan Lashkari; Aliakbar Matkan; Hossein Asakareh
Abstract
Introduction
Survey of spatial relationships of environmental data is considered as one of themost important goalsof spatial statistics for analyzing the spatial patterns and understanding the spatial dependencies. In this context, the Exploratory Spatial Data Analysis (ESDA) could well provide methods ...
Read More
Introduction
Survey of spatial relationships of environmental data is considered as one of themost important goalsof spatial statistics for analyzing the spatial patterns and understanding the spatial dependencies. In this context, the Exploratory Spatial Data Analysis (ESDA) could well provide methods for distinguishing betweenspatialrandomandnon-random patterns. Using the ESDA for analyzing the spatial autocorrelation of climatic elements is necessary to distinguish their changes and spatial distribution. The present research is aimed at explaining the use of ESDA to describethe spatial patterns ofwater vapor pressureas one of the most important climatic parameters. Water vapor pressure plays a crucial role in climate system as an important feedback variable associated with the earth’s energy balance and hydrologic cycle. This climatic parameter has an important rolein explaining the climate change or changes in climatic parameters, because: 1) It is the main sourceof rainfall in allweathersystems, 2) It suppliesthe latent heatin this process and controls the heat inthetroposphere, 3) It is the booster of the storm's speed and 4) It plays a major role in the dynamics of atmospheric circulation. So, determination and interpretation of the likely reasons of Water vapor pressure changes and its variability are vitally important for human as well as other living-beings.
Materials & Methods
The studied area, with about 360,200 km2 area, is located in the South and the Southwest of Iran and approximately between 25° 00'N and 34° 25'N latitudes and between 45° 38'E and 59° 17'E longitudes. Southern and southwestern parts of the studied area are located beside two massive sources of moisture, i.e. Persian Gulf and Oman Sea. The main mountain chain in the studied area is Zagros that extends from the northwest to the Southern part of the studied area. The Zagros mountainrange is responsible for the major portion of rain-producing air masses that enter the region from the Western and Northwestern sides, with relatively high amounts of rainfall. In this study, water vapor pressure data in pixels (dimension of 9×9 km) inthe time interval of 1981-2010 were collected by the Iranian Meteorological data website (http://www.weather.ir).To interpolate the water vapor pressure, Kriging Inverse Distance Weighting (IDW) and Radial Basis Functions (RBF) were tested and so after theerror validation, the optimum method (Ordinary Kriging with Gaussian method) was chosen. Considering the aim of this study, analyzing the spatial variability of WV in regional and local scale, the most important geographical features such as elevation, longitude, latitude, slope and other aspects were chosen. Topographical maps of the studied area were collected by the Geological Survey of Iran (http://www.gsi.ir). The Digital Elevation Model (DEM)with a 10 Km cell size was derived by mosaicking, geo-referencing, and editing these maps in Arc GIS 10.2 software, and the geographical features were prepared based on it. Moran's I, local Moran'sAnselin, and LISA were used asESDA’s approaches to analyze the spatial autocorrelation of water vapor pressure patterns based on climate parameters.
Results & Discussion
According to the cross validation, it was cleared thatthe optimum method for interpolation of water vapor pressureis Ordinary Kriging with Gaussian method. The results of Moran’s Istatistic showed that the water vapor pressure hasspatial structure and is distributed in cluster patternin the South and the Southwest of Iran. The monthly surveys showed thatthe spatial autocorrelation of water vapor pressure in warm months is higher than the cold months and therefore hasa greater tendency to cluster. The results alsoshowedthat the water vapor pressure is tending to disperse and non-clusterinspace in the South and SouthWest of Iran. The bivariate Moran's Istatistic for relation of water vapor pressure and longitude showed thestrong and positive spatial autocorrelation and also clustered pattern.
Conclusion
The monthly surveys showed that the spatial autocorrelation of water vapor pressure in warm months is higher than the cold months and is more tending towards clustering. The existence of such situation in most regions of the studied area in the warm seasons reflects the consistency and homogeneity in this seasons in relation to other seasons. The main reason for these circumstances may be the lack of non-diversification of input pressure systems in these seasons, climate uniformity and sustainability and effects of local systems. Over the time, the water vapor pressure in the South and Southwest of Iran has tended to be more dispersed and less clustering in space. The reason for this incident is not fully revealed but it may be attributed to topographical effects, changes in system positioning, land use changes, etc.Investigating the relationship between spatial distribution of water vapor pressure and geographical parameters showed that the relationship betweenwater vapor pressureand latitude,elevation and slope suggested adispersed and heterogeneousspatial distribution between them. The results of the bivariaterelationship betweenwater vapor pressureand other aspects suggested a discontinuous and random relation.
