Download PDFOpen PDF in browserStatistical analysis and Spatial correlation of rainfall in Greece for a 20-year time periodEasyChair Preprint 7376 pages•Date: January 18, 2019AbstractRainfall is a major natural process, with severe impact on human population. Recent changes in rainfall are directly linked with climate change. This study aims at the spatial and statistical analysis of rainfall in Greece during the last two decades of the 20th century. Its main objective is to investigate current conditions and evaluate emerging trends. Analysis relies on data from 74 meteorological stations all over the country and covers the years 1981-2000. Methodologically, the paper is divided into two parts. In the first part, Thiessen polygons have been created to determine their influence zones, and spatial correlation has been estimated using Moran's I index, weighted by appropriate spatial proximity matrices. Analysis has been performed using GeoDa software. At a second level, and having selected a representative sample from those stations, descriptive statistics were applied to determine current status and identify future trends. For each one of the years studied, analysis refers to two sub-periods, the dry and the wet one, as well as to the year as a whole. Differences detected between dry and wet seasons, as well as the similarities between wet and year period imply that the characteristics of annual rainfall are mainly conditioned by those of the wet season. This approach, if applied for multiple and continuous time-periods, could serve as a very useful tool for researching and analyzing future trends and spatial patterns, in different areas worldwide. Thus, important conclusions can come of about rainfall (or climatic in general) differentiations, possibly due to climate change. Keyphrases: GIS, Rainfall, climatology, spatial analysis, spatial statistics
|