Significant changes in the regional climate of the Aegean during 1961–2002

Abstract

Significant changes in the regional climate of coastal and island stations in and bordering the Aegean during 1961-2002 are identified. Here we start with a very large number of parameters describing the surface temperature climate measured at 9 Greek island and coastal stations, and highlight those parameters which show regionally significant trends. Statistical significance of trends is assessed over multiple sliding time-windows of data. For each parameter and each time-window, a regional summary is produced using two quantities: the statistical significance of the trend in the regionally averaged parameter, and the number of stations which individually show significant trends in that parameter. The relationship between these two quantities provides a simple way of highlighting those climate parameters for which the region shows a trend, which is spatially relatively homogeneous, and those for which the region may need to be divided into sub-regions. Some very detailed characteristics of trends in daily surface temperatures are summarized in a concise way. Notably, the established cooling during the 1960s is shown to be especially strong for November and also December. Some trends were also shown to be significant only in terms of temperatures of unseasonably warm or cool days. For example, cooling in May temperatures in the late 70s/early 80s occurred most clearly for the low-temperature range-appearing as a decrease in the temperature reached by unseasonably cold days. This behaviour was inhomogeneous across the region, with significant cooling at only a subset of the stations. Such results could be used by subsequent analysis to focus on particular parts of the data, rather than using averaging to reduce the data.

Recent trend analysis of mean air temperature in Greece based on homogenized data

“…Moreover, it is observed that all stations show a negative trend in November with an average cooling rate of −0.02°C/year. This result is in agreement with Good et al (2008) who pointed out that the significant cooling in the Aegean region was shown to be mostly in November and (less so) in December. As far as December is concerned, the cooling appears much clearly in the north, in the central, and in the eastern Aegean region (rate from −0.02 to −0.032°C/ year).…”

“…A continuing trend toward warmer summers from the middle of 1980s has been identified (Akylas et al 2005), while 2003 has been recorded up to 2004 as the hottest summer for Athens but up to date followed an even warmer summer of 2007. In a more recent study, Good et al (2008) demonstrated that clear trend signals in the Aegean are restricted to specific months of the year and to certain sub-regions. Furthermore, the Mediterranean cooling between 1950 and 1970 in Greece was found to be strongest in November and (less) in December.…”

“…In accordance to the previous studies, Klein Tank et al (2002) concluded that air temperature series of Greece are different than those obtained in other European countries and do not show statistically significant trends during 1976-1999. Thus, due to its unique geography and remarkable range of microclimates, Greece has been the subject of several studies dealing with temperature variation and trends (Karapiperis 1954;Arseni-Papadimitriou 1973;Giles and Flocas 1984;Katsoulis 1987;Arseni-Papadimitriou and Maheras 1991;Metaxas et al 1991;Bartzokas and Metaxas 1995;Proedrou et al 1997;Makrogiannis et al 1998;Retalis et al 1998;Sakellariou et al 2000;Feidas et al 2004;Founda et al 2004;Akylas et al 2005;Good et al 2008;Philandras et al 2008;Founda 2011). Karapiperis (1954) studied the temperature evolution in Athens for the period from 1852 to 1952 taking into account the observations from the National Observatory of Athens (NOA) station.…”

Evaluation of a regional climate model using in situ temperature observations over the Balkan Peninsula

“…This implies that the model's extreme low minimum temperatures occur over individual days, which do not constitute spells. Across the northeastern region (Romania), the model overestimates the maximum and underestimates the minimum winter temperature, resulting in an expansion of the diurnal temperature range. Exceptionally low temperatures and an overestimation of the number of cold spells is found at a few stations across the southern Balkan Peninsula and might be associated with an uncommon prevalence of anticyclonic circulation at the surface, which favours decreases in minimum temperatures (Good et al, 2008). …”

Spatio-temporal analysis of warming in Bangladesh using recent observed temperature data and GIS

“…Temperature is one of the most important parameters of climate and the trend of temperature (maximum, minimum and mean) is used as an indicator of climate warming or cooling. Numerous studies were carried out over the world and mentioned the trends of temperature of the different parts of the globe (Easterling et al 1997;Domonkos and Tar 2003;Franke et al 2004;Founda et al 2004;Begert et al 2005;IPCC 2007;Good et al 2008;Brázdil et al Abstract This study focused on the annual and seasonal warming at local scale by analysing the trends, anomalies, change points and shifting of isotherm in temperature from 34 meteorological stations distributed over Bangladesh, spanning 40 years from the year 1971-2010. For trends, a linear regression using least square model was applied.…”