Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

An increased frequency of extreme meteorological phenomena and exceptional weather conditions, including exceptionally hot summers and summer heat waves, is often quoted as a symptom of contemporary climate change. Considerable fluctuation of thermal conditions from year to year is particularly characteristic of the temperate geographical zone covering much of the European continent. Nonetheless, the last decade of the twentieth century and the first decade of the twenty-first century stand out with particularly frequent occurrences of hot summer months and even entire summer seasons. Examples of this phenomenon include extremely hot summer seasons of 2003 in Western Europe and of 2010 in the European part of Russia characterised by an enormous extent of both the scale of thermal anomalies and their spatial coverage [1].

Such intensive and persistent heat waves are classified as extreme climatic phenomena and are known to have multiple adverse consequences. Overheating has severely adverse effects on the human health and well-being and may increase the risk of death [2–7]. Droughts that accompany heat spells damage the agriculture and increase the risk of wide spread forest fires. High air temperatures cause the melting of mountain glaciers, which cannot recover during the subsequent winter. Modelling research shows that the tendency for such adverse weather conditions to increase in frequency is likely to continue leading to more frequent, lasting and intense hot spells [8–12]. The exceptional heat waves of 2003 and 2010 constitute particularly strong manifestations of the contemporary warming of the European climate and abound in proofs of the effects mentioned before. Sources of such anomalies are still to be fully understood. In the temperate zone of Europe the direct causes include stationary high-pressure systems, which normally involve tropical advection and, additionally, strong insolation on long summer days due to the lack of cloud cover typical of high-pressure systems. Various scholars pointed to different primary causes of these anomalies, including ocean thermal conditions [13], especially their surface temperature [14–17], but also a growing concentration of greenhouse gases in the atmosphere [18–20].

This chapter focuses primarily on the frequency of exceptionally hot summer months in Central and Eastern Europe, their spatial extent and the thermal characteristic of the months. Most of the study area is relatively flat and only the south-western part features uplands and two large mountain ranges, i.e. the Carpathian Mts. and Sudety Mts., running predominantly along an east–west axis. This morphology contributes to a great variation in the frequency, intensity, duration and timing of hot spells within a relatively small area [21–23].

This chapter builds on the authors’ earlier research on whole summer seasons [1].

2 Data and Method

The input to the study involved monthly average, maximum and minimum air temperatures; the number of days with tmax >25, 30 and 35 °C (defined respectively as summer days, hot and very hot days); and days with tmin >20 °C (tropical nights), as recorded in the summer months (June–August) at 59 weather stations in Central and Eastern Europe in 1951–2010 (Table 2.1). The study area spans 45–60°N and 15–65°E, excluding the Scandinavian Peninsula. The weather records were sourced from an on-line database European Climate Assessment & Dataset (ECA&D).

Table 2.1 List of weather stations included in the study (station names after CLINO) [24]
Full size table

An EHM was defined as a month, in which the average air temperature was higher by at least two standard deviations (t ≥ tav. +2σ) than the long-term average (1951–2010). The same definition had been applied for extremely hot summers [1, 22].

The study area is predominantly lowland. Indeed, 49 of the 59 weather stations are located below 200 m a.s.l., 7 stations are between 200 and 300 m and only 3 stations are above 300 m a.s.l. (L’viv at 323 m, Praha at 365 m and Cluj at 413 m). Astrachan’ lies in a depression at −23 m.

The central and eastern sections of the area, which form the majority, have a moderate continental climate, while a transitional maritime to continental climate prevails over the western part. The average long-term temperature of the warmest month, i.e. July, ranges from ca. 15 °C in the extreme north of the area to 25 °C and more in the southeast. This is evidence of a wide range of thermal characteristics within the study area. In this context it is important to note that the definition of an EHM is relative, which means that in each case it is related to the average long-term temperature in a given month at a given station. The adoption of this method has two specific consequences:

  1. 1.

    The average air temperature of an EHM ranges widely. For example, in July it may even fall short of 20 °C in the north-east, while in the southeast it must be 10 °C higher to meet the criterion (t ≥ tav. +2σ). This means that an EHM may mean very different thermal conditions in different sections of the same climatic zone.

  2. 2.

    Subsequent summer months with similar average temperatures may or may not qualify as EHMs. For example, in 1972 in Moscow, the average temperature of July was 22.4 °C and 20.6 °C in August, but only the cooler August cleared the threshold (t ≥ tav. +2σ).

3 General Characteristic of EHM Occurrence in the Summer Season

During the 60-year period, 326 months met the condition t ≥ tav. +2σ (3.1 % of all months in the period) at 59 weather stations. These EHMs occurred in 47 calendar months in 33 years of the study period, as shown in Table 2.2. The table also specifies the number of stations where a given EHM occurred and offers the calendar of extremely hot summer seasons (EHS). Interestingly, for statistical purposes there were several EHSs without a single EHM.

Table 2.2 Calendar of exceptionally hot months (EHM) and summers (EHS) and number of station at which EHM and EHS occurred in Central and Eastern Europe
Full size table

The study revealed that EHMs varied very widely in terms of their geographical extent, ranging from one or two stations (13 and 2 EHMs) to more than a half of all stations (one EHM each at 36 and 34 stations).

The EHM count at all stations ranged from 3 to 10, but at a clear majority (41) it was 4–6. There was no clear spatial pattern, as even neighbouring stations differed greatly in the number of occurrences, e.g. Chernivtsi 3, Kyiv 8. Generally, however, they were less numerous (3–4) in the west of the area (from Lithuania to Poland, the Czech Republic, Slovakia to Romania) and at some stations in Russia within the belt 50–55°N, than in Belarus, north-eastern Ukraine and Russia between the Black and Caspian Seas and on the Caspian coast of Kazakhstan (7–10). They were also surprisingly frequent (8) in Pecora, the farthest station to the north-east.

The likelihood of an EHM in June, July and August is virtually identical (15, 16 and 16 cases), but the geographical extent of EHMs clearly increases as the summer progresses from 87 station-months in June to 98 in July and 141 in August (Table 2.2). This would suggest that the gradually warming ground plays a role as a factor in levels of air temperature.

EHMs were distributed highly unevenly during the study period, and increased in frequency after 1990 (Tables 2.2 and 2.3). Until that year, their occurrence was similar at 5–7 per decade, with an exception of just 2 EHMs during the period 1961–1970. In the last decade of the twentieth century, this number increased to 11 EHMs (including 5 at a single station), which trend continued in the first decade of the twenty-first century with 15 EHMs. These latter EHMs occurred simultaneously over large areas, the record of which was in 2010. Also EHMs in the twenty-first century tended to be more concentrated in the same year at 2–3 each.

Table 2.3 Years, months and occurrences with EHMs in individual decades of the period 1951–2010 in Central and Eastern Europe
Full size table

Individual EHMs spanned different parts of the study area and while they showed no temporal pattern, there was a general increase in their occurrence in European Russia.

4 Location, Extent and Thermal Characteristic of Extremely Hot Months

As has already been mentioned, 15 of the 47 EHMs occurred in areas with just 1 or 2 meteorological stations (Table 2.2). Some of these represent an interesting case of EHMs around the edges of the study area (Berlin, Bucuresti, Astrachan’), which may suggest that the EHMs recorded there could have covered a wider territory that fell outside the study area boundaries. These 15 cases will be omitted from an examination of the detailed thermal characteristics of EHMs that follows, as will 15 other EHMs which occurred at only 3–6 stations.

Indeed, only the 17 remaining EHMs which occurred at more than 6 stations (more than 10 % of all stations) are included in the discussion, as they represent the greatest geographical extent and/or greatest temperature increase. All 47 EHMs, however, are included in Table 2.3 and in a summary of all EHMs (Table 2.4).

Table 2.4 Exceptionally hot months (EHM) in Central and Eastern Europe (1951–2010)
Full size table

5 EHMs During 1951–2000

During the first 50 years of the study period, only six EHMs extended to more than 10 % of the stations (Table 2.2).

June 1964 (Fig. 2.1, Table 2.5) was an EHM in the south-western part of the area covering sections of Poland, Belarus, Slovakia, Hungary, western Ukraine and northern Romania (Fig. 2.1). The EHM centred on L’viv where the temperature reached t ≥ tav.+3σ (the anomaly, or Δt, in L’viv was 4.2 °C, and outside L’viv it ranged from 2.5 to 4.0 °C). At most of the stations this was the hottest June in the study period. Hot days were recorded across the area.

Fig. 2.1
figure 1

Geographic coverage of the exceptionally hot month of June 1964 (1) and August 1972 (2)

Full size image
Table 2.5 Thermal characteristic of the EHM: June 1964, August 1972
Full size table

August 1972 (Fig. 2.1, Table 2.5) was the first EHM of the period, which covered nearly all of the European part of Russia (Δt from 3.0 to 6.5 °C). In its central section the anomaly reached three standard deviations from the long-term average, or more than 5 °C, including 6.1 °C in Voronez and 6.5 °C in Tambov. (There were only 13 cases where the long-term average temperature was exceeded by more than 6.0 °C). At some of the stations this was the hottest August of the study period. Hot days were recorded across the area (e.g. 7 in St. Petersburg compared to an average of once in 3 years and ca. 20 in the south compared with ca. 5 on average). In the southern half of the EHM area there were cases of the very rare tropical nights (e.g. 8 in Kharkiv, 1–2 on average). The summer of 1972 proved an EHS and the month was the hottest in Finland and northern Russia and the second most expansive EHS during the study period [28].

June 1989 (Table 2.5). The EHM covered the entire north-eastern part of the European Russia (Δt from 4.2 to 5.5 °C). At a majority of the stations it was the warmest EHM of the study period. Hot days were observed at all stations.

August 1992 (Fig. 2.2, Table 2.5) The EHM covered a south-western part of the area, including the larger part of Poland in the south. In most of the area the average temperature exceeded the long-term average by more than 3 °C (Δt from 2.9 °C in Poznan to 5.0 °C in Wien; in Wien, Budapest, Debrecen and Krakow the anomaly exceeded three standard deviations). At nearly all stations (excluding those located at the edge of the area) this was the hottest August of the period. The maximum temperatures were very high, hot days were recorded at all stations and there were also some very hot days. Krakow had the highest frequency of very hot days (tmax>35 °C; 13 cf an average of 1 day every 2 years). These exceptionally high temperatures in Krakow could be explained both by the foehn effect and a vast forest fire at Kuźnia Raciborska 120 km to the west [25]. In the south of the area there were many instances of tropical nights (tmin>20 °C = ca. 20, cf an average of ca. 4).

Fig. 2.2
figure 2

Geographic coverage of the exceptionally hot month of August 1992 (1) and June 1999 (2)

Full size image

In Poland, Slovakia, Czech Republic and Hungary the summer of 1992 was classified as an EHS. It was also the hottest summer of the period in southern Poland and Krakow’s only EHS [28].

June 1998 (Table 2.5) was the first EHM to cover the southeast of the study area starting from the latitude of Samara to the Caspian Sea. It was the hottest June in this territory in the study period (Δt from 3.5 to 4.7 °C). Both the maximum and minimum temperatures were very high and all stations recorded both hot and very hot days (e.g. 20 hot and 9 very hot days at Orenburg, cf 9 and 2 on average; 26 and 12 at Atyrau, cf 17 and 5 on average) and tropical nights (e.g. 8 in Orenburg, cf 1 on average and 19 at Atyrau, cf 8 on average).

June 1999 (Fig. 2.2, Table 2.5) was one of the most extensive EHMs, ranging from Finland to north-western Russia, the Baltic states (but not Poland), Belarus, and central Ukraine to the Black Sea. Across its territory it was the hottest June in the study period (Δt from 2.4 to 3.0 °C in the south, to 4.3–5.0 °C in the centre to 3.3–3.9 °C in the far north). The maximum temperatures were particularly high with hot days at all stations, especially in Kyiv (11, cf. 2 on average), but no very hot days.

5.1 EHMs During 2000–2010

There were nearly twice as many EHMs (11) during the last decade of the study period, which were observed at more than 10 % of the stations than during the previous 50 years (Table 2.2).

July 2001 (Fig. 2.3, Table 2.6). The EHM covered an area stretching from Estonia, Latvia, eastern Lithuania, Belarus and the Ukraine without its western part and the western edge of Russia. In the centre of the area, the average temperature was more than 4 °C higher than the long-term average (Δt across the area from 3.1 to 4.9 °C). Hot days were recorded at all the stations, very hot days in the south (e.g. 16 hot and 4 very hot days in Kharkiv, on average every 4 and 5 years) and also tropical nights at nearly all stations, especially on the Black Sea coast (13 in Kharkiv, 26 in Odesa; cf averages of 2 and 7 respectively).

Fig. 2.3
figure 3

Geographic coverage of the EHM of July 2001 (1) and August 2007 (2)

Full size image
Table 2.6 Thermal characteristics of the EHM: July 2001, July and August 2002
Full size table

July 2002 (Table 2.6). The EHM covered more or less the same area as the year before, apart from the far north, but including the whole of the Ukraine (Δt from 3.0 to 3.9 °C). Just as in the previous year, hot days were recorded across the board, but very hot days only in the far south. Tropical nights occurred everywhere.

August 2002 (Table 2.6). This was the second consecutive EHM that year, but this time only at the north-western end of the study area stretching from Finland to the northern half of Poland (Δt from 2.7 to 4.0 °C). At certain stations (Jyvaskyla, Riga, Kaliningrad, Poznan) this was the hottest August of the study period. There were isolated hot days.

In addition the entire summer season of 2002 was classified as an EHS at a few stations in the western part of the study area.

In 2003, all summer the months qualified as EHMs, but only in June and August were more than 10 % of the stations involved (Table 2.2). In these months, the EHM area covered a south-western section of the study area (Δt from 2.3 to 3.3 °C), which was on the periphery of an EHS observed throughout Western Europe. There the exceptional heat wave contributed to higher death rates and to an increased rate of melting of Alpine glaciers noted in several studies [e.g. 4, 18, 19, 23, 26–29].

June 2003 (Table 2.6) was an EHM in Western Europe, which only covered the south-western edge of the study area from Praha and Krakow to Beograd (Δt from 2.8 to 4.6 °C). At nearly all of the stations it was the hottest June of the study period. In Wien and Zagreb the average temperature climbed to t ≥ tav.+3σ. Hot days were recorded at all stations, but there were only sporadic cases of very hot days.

August 2003 (Table 2.6) was the second case, 49 years after July 1954, of a dual-area EHM: one in the foreland of the Ural Mountains to the east of the area (Δt from 4.2 to 5.1 °C, the hottest August of the study period) and the other in the far south-west stretching from Berlin to Beograd (Δt from 2.7 to 4.6 °C). Again, like in June, this latter area was on the periphery of a powerful EHM that covered Western Europe and featured the greatest temperature increase rate during the study period. The single largest temperature anomaly was recorded in the south-western part of the Massif Central, France (Gordon weather station) [12]. In the western part of the study area the summer of 2003 also qualified as an EHS.

July 2006 (Table 2.6) was an EHM in the west and stretched from Poland to western Hungary and Croatia and, just as in 2003, it was part of an EHM centred on Western Europe. Throughout the area affected it was the hottest July in the study period (Δt from 2.8° in Zagreb to 5.2 °C in Poznan). Hot days occurred very frequently at all stations (e.g. 18 each in Warszawa and Krakow, cf 3–4 on average), while very hot days were sporadic.

The summer of 2007 again involved three EHMs, even if July only covered 5 % of the stations. In the south-western part of the area the season qualified as EHS.

June 2007 (Table 2.6). The EHM covered a small area in the south-west to the south of the line Vienna-Odessa and was part of a larger EHM in the Balkan Peninsula. In the east of the affected area, Bucuresti to Odesa, this was the hottest June during the study period (Δt from 2.7 to 3.7 °C, except Sulina 2.2 °C). Throughout the area hot days and tropical nights were commonly recorded, but very hot days were only recorded sporadically.

August 2007 (Fig. 2.3, Table 2.6) covered the largest area to that date, including a larger south-eastern part of European Russia, parts of Estonia, Latvia and Belarus, eastern Ukraine and western Kazakhstan. The long-term average was exceeded by ca. 3 °C in the north, by 3.5–4.5 °C in the south and by ca. 5 °C in the centre (Δt from 2.8 to 5.3 °C). Hot days were recorded throughout the area, from 2 to 3 in the north (cf an average of once every few years) to 30–31 on the Caspian coast (cf an average of 18–20), where very hot days peaked in frequency (19–22, cf an average of 3–6). Tropical nights were also observed throughout the area with the record numbers on the Black Sea and Caspian coasts (Odesa 17, cf an average of 7; Atyrau 25, cf an average of 10).

The most intense heat wave of 2007 was recorded in the Balkan Peninsula [30].

All three summer months of 2010 qualified as EHMs, but for the first time their coverage was similar, especially in July and August (Fig. 2.4). The latter two were also the EHMs with the largest territorial extent during the study period. Both had a similar coverage to August 2007, including a western part of the European Russia, Belarus, the Ukraine and western Kazakhstan. In an earlier study the authors [28] found that the average temperature at certain Russian stations exceeded the long-term averages by up to 4σ.

Fig. 2.4
figure 4

Geographic coverage of the EHM of July (1) and August (2) 2010

Full size image

June 2010 (Table 2.7). The EHM was split into two relatively small areas: one from Kyiv to Voronez along the border between Russia and the Ukraine (Δt from 3.4 to 4.0 °C) and the other in the southeast from Astrachan’ to Aktobe (Δt from 3.1 to 4.9 °C). At most of these stations this was the hottest June during the study period. Tropical nights were recorded at all stations and very hot days at nearly all of them.

Table 2.7 Thermal characteristics of the EHM in 2010
Full size table

July 2010 (Fig. 2.4, Table 2.7). The EHM covered an area of record size stretching from the Baltic coast to the eastern part of the Black Sea and the Caspian Sea. At all stations, except Berlin and Poznan, it was the hottest July during the study period. Everywhere the average long-term temperature was exceeded by more than 3.5 °C and in the central part of the area affected by more than 5 °C, including by more than 6 °C at 8 stations (record Δt values: 6.9 °C at Kursk and 6.8 °C in Moscow and Tambov). At 14 stations from Jyvaskyla to Samara the average monthly temperature equalled t ≥ tav.+3σ. Hot days were observed throughout the area; from 5 at Jyvaskyla (an average of once in 3 years), 14 at St. Petersburg (on average 1) to 23–25 at Kursk and Voronez (on average 3–6) and 31 (on all days of the month) on the Caspian Sea (on average 21–24). In the central and south-eastern part of the area there were also very hot days (e.g. 7–9 at Kursk and Voronez, on average once in a few years; 24 at Atyrau, on average 9). Tropical nights were also recorded at all stations and at most of them at record levels, including the only occurrence during the study period at Jyvaskyla, 16 each in St. Petersburg and Moscow (less than one on average) and 26–27 at Astrachan’ and Atyrau (on average 13–14).

August 2010 (Fig. 2.4, Table 2.7) This EHM by-and-large overlapped with the territory of the July EHM, but included a shift to the east into the southern Ural Mts., at the cost of northern and western coverage (Δt from 2.7 to 7.0 °C). In its central section it was the hottest August during the study period. Here, in the central part of European Russia and the neighbouring part of Ukraine, the average monthly temperature reached t ≥ tav.+3σ, which in absolute terms typically meant 5 °C. In the town of Kursk, which recorded the greatest anomaly at t ≥ tav.+3.9σ, or 7.0 °C, as well as in Kiev, Kharkiv and Voronez, this was the third consecutive EHM, a sole event of the type during the study period. All the stations within the EHM area recorded hot days, ranging from 5 in St. Petersburg (on average once in 3 years), to 16 in Moscow (on average 1), 20 at Kursk (on average 2) to 27 at Astrachan’ (on average 18). Very hot days were also recorded by most stations where they typically reached the highest frequency during the study period. Moscow with its 7 very hot days (after 9 such days in July) stood out particularly, as they included the sole case of a maximum temperature above 35 °C during the study period. On the Caspian coast the incidence of very hot days, common in the area, was 3–5 times greater than the average. Tropical nights were noted almost at all stations, including 19 at Kursk (on average less than 1) and 8 in Moscow (on average once in 5 years), while in the far south of the territory affected they were 2–3 times more frequent than on average.

The final year of the study period was unique within its decade with two consecutive EHMs with the greatest spatial scale and greatest temperature increase. This long spell of exceptionally high temperatures caused a considerable deterioration in living conditions and highly adverse economic effects as it contributed to the development of vast wildfires, which, in turn, caused the levels of air pollution in Moscow to rise by 2–3 times [31]. The wildfires may have also made a considerable contribution to the scale of the temperature increase [1]. This proposition is supported by a study on an EHS of 1992 in Poland [25].

6 Conclusions

During the period 1951–1960, 47 exceptionally hot summer months (EHMs) of varying territorial extent occurred in Central and Eastern Europe. In total 22 EHMs covered areas represented by 1–3 out of the total 59 stations (5 %), 30 EHM were recorded by 6 stations (10 %) and only 7 EHMs occurred in areas with more than 15 stations (more than 25 %).

The EHMs with small station coverage must be seen in the context of the boundaries of the study area, beyond which many EHMs might indeed have continued.

A vast majority of the EHMs occurred in a single area and there were only three EHMs, which had two separate locations during the study period (July 1954, August 2003 and August 2006).

EHMs did not display any specific spatial or temporal patterns of occurrence, although there were temporary trends when they concentrated in a similar area, including 4 EHMs in the far north-west (1981–1991), 5 EHMs in the far south-west (twenty-first century), and a series of EHMs in European Russia (2007 and 2010). During the study period, the frequency of EHMs increased after 1990 and further still after 2000.

These latest EHMs grew in size and frequency to 2 or even 3 in the same year, sometimes in the same area. The persistence of exceptionally hot spells increased while the deviation of the average temperature from the long-term average was often greater than ever before. This constitutes an unquestionable example of the contemporary global warming.

The most intense and vast EHMs in the area included: August 1972 (Russia), August 1992 (the south-west), June 1999 (broad belt from Finland to the Black Sea), August 2007 (Russia) and July and August 2010 (Russia). The persistence of exceptionally hot spells increased while the deviation of the average temperature from the long-term average was often greater than ever before. This constitutes an unquestionable example of the contemporary global warming.