K. Tolika

Recent past and future patterns of the Etesian winds based on regional scale climate model simulations

The aim of this work is to investigate the recent past and future patterns of the Etesian winds, one of the most persistent localized wind systems in the world, which dominates the wind regime during warm period over the Aegean Sea and eastern Mediterranean. An objective classification method, the Two Step Cluster Analysis (TSCA), is applied on daily data from regional climate model simulations carried out with RegCM3 for the recent past (1961–1990) and future periods (2021–2050 and 2071–2100) constrained at lateral boundaries either by ERA-40 reanalysis fields or the global circulation model (GCM) ECHAM5. Three distinct Etesian patterns are identified by TSCA with the location and strength of the anticyclonic action center dominating the differences among the patterns. In case of the first Etesian pattern there is a ridge located over western and central Europe while for the other two Etesian patterns the location of the ridge moves eastward indicating a strong anticyclonic center over the Balkans. The horizontal and vertical spatial structure of geopotential height and the vertical velocity indicates that in all three Etesian patterns the anticyclonic action center over central Europe or Balkan Peninsula cannot be considered as an extension of the Azores high. The future projections for the late 21st century under SRES A1B scenario indicate a strengthening of the Etesian winds associated with the strengthening of the anticyclonic action center, and the deepening of Asian thermal Low over eastern Mediterranean. Furthermore the future projections indicate a weakening of the subsidence over eastern Mediterranean which is rather controlled by the deepening of the south Asian thermal Low in line with the projected in future weakening of South Asian monsoon and Hadley cell circulations.

Climate change effects on the marine characteristics of the Aegean and Ionian Seas

This paper addresses the effects of estimated climate change on the sea-surface dynamics of the Aegean and Ionian Seas (AIS). The main aim is the identification of climate change impacts on the severity and frequency of extreme storm surges and waves in areas of the AIS prone to flooding. An attempt is made to define design levels for future research on coastal protection in Greece. Extreme value analysis is implemented through a nonstationary generalized extreme value distribution function, incorporating time harmonics in its parameters, by means of statistically defined criteria. A 50-year time span analysis is adopted and changes of means and extremes are determined. A Regional Climate Model (RegCM3) is implemented with dynamical downscaling, forced by ECHAM5 fields under 20C3M historical data for the twentieth century and the SRES-A1B scenario for the twenty-first century. Storm surge and wave models (GreCSSM and SWAN, respectively) are used for marine climate simulations. Comparisons of model results with reanalysis and field data of atmospheric and hydrodynamic characteristics, respectively, are in good agreement. Our findings indicate that the dynamically downscaled RegCM3 simulation adequately reproduces the present general circulation patterns over the Mediterranean and Greece. Future changes in sea level pressure and mean wind fields are estimated to be small, yet significant for marine extremes. In general, we estimate a projected intensification of severe wave and storm surge events during the first half of the twenty-first century and a subsequent storminess attenuation leading to the resettlement of milder extreme marine events with increased prediction uncertainty in the second half of the twenty-first century.

Drought Index Over Greece as Simulated by a Statistical Downscaling Model

Drought is the least understood of all weather phenomena, since it differs from other natural hazards in several ways. The hazardous of drought can be better understood by analyzing drought indices. The Standardized Precipitation Index (SPI) has been one of the most widely used indices for drought studies, as it can provide satisfactorily results for the appearance, variability and intensity of drought. Moreover, SPI has been developed in order to quantify and record drought episodes on multiple time scales (3 months, 6 months, 1 year, 2 years). In the present study a statistical downscaling technique based on artificial neural network was employed for the estimation of SPI over Greece. The aim of the study is to evaluate the simulated SPI index against the observational one. The performance of the statistical downscaling model is assessed utilizing biases, standard deviation and correlation coefficient between observed and simulated indices. An overestimation of the simulated mean SPI indices accompanied by a decrease in standard deviation is evident for all stations and all time periods. The reproduction of SPI3 and SPI6 for winter, spring seasons display rather promising results.

Comparison of an automated classification system with an empirical classification of circulation patterns over the Pannonian basin, Central Europe

The aim of the study is to compare the performance of the two classification methods, based on the atmospheric circulation types over the Pannonian basin in Central Europe. Moreover, relationships including seasonal occurrences and correlation coefficients, as well as comparative diagrams of the seasonal occurrences of the circulation types of the two classification systems are presented. When comparing of the automated (objective) and empirical (subjective) classification methods, it was found that the frequency of the empirical anticyclonic (cyclonic) types is much higher (lower) than that of the automated anticyclonic (cyclonic) types both on an annual and seasonal basis. The highest and statistically significant correlations between the circulation types of the two classification systems, as well as those between the cumulated seasonal anticyclonic and cyclonic types occur in winter for both classifications, since the weather-influencing effect of the atmospheric circulation in this season is the most prevalent. Precipitation amounts in Budapest display a decreasing trend in accordance with the decrease in the occurrence of the automated cyclonic types. In contrast, the occurrence of the empirical cyclonic types displays an increasing trend. There occur types in a given classification that are usually accompanied by high ratios of certain types in the other classification.

Analysis of the Synoptic Conditions of the Extreme Rainfall Heights of 2014 in Thessaloniki

The year of 2014 is characterized by particularly high rainfall totals in the area of Thessaloniki on an annual, seasonal, monthly as well as daily basis where several maxima have been exceeded. In the present study, an attempt is made to analyze the synoptic atmospheric conditions during the year of interest. The analysis of the atmospheric circulation conditions showed that depending on the season and the atmospheric level, 500 or 1000 hPa, the intense precipitation conditions of the examined year are due to the increased frequency of some cyclonic types (C, Cne, Cwsw) and to the increased intensity of daily precipitation of the same cyclonic types or to the combination of these factors. It should be highlighted that the mechanisms of the intense precipitation during winter are different from the other seasons.

Physics Parameterizations of Regional Climate Model RegCM4: Sensitivity to Convective Precipitation Schemes

Cumulus convection is one of the most complex subgrid-scale process parameterized in climate models, which has major influence on the determination of the vertical distribution of energy, water and momentum. Numerous convective parameterizations have been developed and applied in global and regional climate models, with the results being very sensitive to the scheme selection. Thus, it is essential to study the sensitivity of climate models to the convective precipitation schemes. The examined regional climate model used in the study is RegCM4. Its spatial resolution is 25 × 25 km and different simulations were performed with changes in cumulus convection scheme and the corresponding closure assumptions. The selected domain of interest is the Mediterranean region. The simulations were forced by ERA-Interim data and cover the time period 1981–1990. The different simulated data were processed and the differences between the simulations were calculated, compared and mapped in order to examine the modifications that occur in precipitation from changes in convective precipitation scheme.

ECHAM5/MPI General Circulation Model Simulations of Teleconnection Indices Over Europe

In this paper, a study of potential future changes of the atmospheric circulation over Europe is presented. Atmospheric circulation is studied through two teleconnection indices, the North Atlantic Oscillation (NAO) Index – which mostly affects the climate of western Europe – and the North Sea-Caspian Pattern (NCP) Index – mainly affecting eastern Mediterranean and the Balkan Peninsula. Firstly, the ECHAM5/MPI (Max Planck Institute) General Circulation Model’s (GCM’s) simulations for the two teleconnection indices are evaluated against NCEP/NCAR reanalysis data for the control run period 1971–2000, both in a temporal and a spatial scale. Secondly, the GCM’s future simulations are studied. The main goals of this study are to validate the ECHAM5 GCM regarding the two teleconnection indices under consideration, and consequently to examine whether the current dominant circulation patterns change or not throughout the twenty-first century.