Spatiotemporal analysis of hydroclimatic variability in Crete

The assessment of extremes in hydrological processes is crucial for a wide range of human activities, from engineering design to risk management. In recent years, the island of Crete has been affected by extreme weather events due to its geographical position, as it is located in the south-eastern Mediterranean, a region considered to be one of the most vulnerable to climate change, given the consequences of both extreme weather events (floods, droughts, heat waves) and extensive urbanization. However, in recent decades, hydro-meteorological processes in Crete have been monitored by an extensive network of meteorological stations. The aim of this study is to investigate the spatio-temporal analysis of the hydroclimatic variability for precipitation and temperature processes on the island of Crete for the period 1950 - 2021. The study focuses on the description of the climate throughout the island, by interpreting the temporal variations and spatial analysis of the distribution of the processes studied, in order to identify the vulnerable areas. The seasonality as well as the geographical properties developed by the processes are also examined, based on measurements obtained from the monitoring stations. Through descriptive statistical analysis of hydroclimatic processes, several temporal properties in the data are examined, while correlation analysis of precipitation and temperature variability provides relations between stations and regions for spatial patterns identification. Since many geophysical phenomena are characterized by randomness and cannot be described in a deterministic way, the investigation of the temporal variability of precipitation and temperature was chosen, using stochastic methods. To estimate long-term persistence, the climacogram method (i.e., variance of the time averaged process over averaging time scale) was applied, studying processes at annual and interannual scales, which is of particular interest for the management of water resources. In this way, the Hurst parameter is quantified, as well as presenting less uncertainty compared to other stochastic tools. Finally, the spatial distribution of the climatic variables of precipitation and temperature in Crete is presented, through mapping, with spatial interpolation methods that incorporate Geographic Information Systems (GIS). From these maps the maximum and minimum precipitation and temperature levels for each case were obtained in order to study their spatial variability.