Occurrence of Heatwaves in Selected Regions of Poland and Greece and the Characteristics of their Biometeorological Conditions

For many years, thermal conditions have been the subject of numerous scientific publications in the field of climatology due to the rapidly advancing warming observed on both global and regional scales. According to NOAA (2024), 10 of the warmest years globally since the mid-19th century occurred in the 21st century, all recorded in the last decade (2014–2023). Additionally, since June 2023, each subsequent month has been characterised by the highest temperature anomalies since the mid-19th century.

One of the manifestations of progressive climate warming is the increasingly frequent occurrence of extreme air temperatures, including heatwaves. Among the regions of the world, where a systematic increase in the number and duration of heatwaves is observed, is Europe (Russo et al. 2015, Muthers et al. 2017, Tolika 2019, Tomczyk et al. 2020, Shevchenko et al. 2022, Ballester et al. 2023).

The occurrence of heatwaves in Poland has been the subject of numerous publications at the national (Porębska, zdunek 2013, Wibig 2018, 2021, Tomczyk et al. 2019a, Tomczyk, Bednorz 2023), regional (krzyżewska 2014) and local scale (krzyżewska 2015, Półrolniczak et al. 2018, 2024). Among the most intense heatwaves, the years 1959, 1963, 1968, 1992, 1994, 2006, 2010 and 2015 are often mentioned (krzyżewska, Dyer 2018, Wibig 2018, 2021, Tomczyk et al. 2019a, Tomczyk, Bednorz 2023). Research to date indicates that the most bioclimatically burdensome heatwaves in recent decades were observed in 1994 and 2015 (krzyżewska, Dyer 2018, krzyżewska et al. 2019, Tomczyk et al. 2020, Tomczyk, Bednorz 2023). These were also the longest heatwaves at most stations in Poland, lasting more than 2 weeks (Tomczyk, Bednorz 2023). one of the consequences of high air temperatures is an increase in the number of deaths (Kuchcik 2017, 2021). Graczyk et al. (2019, 2022) demonstrated an increase in the number of deaths during selected heatwaves in Poland. According to these authors, the total number of additional deaths during the heatwave observed in 1994 in the 10 largest cities in Poland might have exceeded 1070. On the hottest days in the analysed period, the number of deaths in some cities was more than three times higher than the average for the reference period.

Similarly, in Greece, over the last few decades, including 1987, 1988, 1998, 2000, 2007, 2012, 2021, and 2022, several intense heatwaves were identified (Founda, Giannakopoulos 2009, Theoharatos et al. 2010, Zoumakis et al. 2012, Katavoutas, Founda 2019, Founda et al. 2022, Ballester et al. 2023). A significant portion of research on heatwaves has focused on this phenomenon in the context of the urban heat island effect in cities such as Thessaloniki (Keppas et al. 2021) and Athens (Founda et al. 2015, Founda, Santamouris 2017, Katavoutas, Founda 2019, Giannaros et al. 2023). The consequences of such weather conditions include an increase in the number of deaths (Zoumakis et al. 2012, 2013, Kouis et al. 2019) as well as fire hazards (Founda et al. 2022). Parliari et al. (2023) and Papadopoulos et al. (2024) showed that, in the next decades in Thessaloniki, there will be an increase in mortality related to high air temperature, especially among old people. As indicated by Founda et al. (2022), due to the preparedness of authorities and agencies, increased public awareness, and local social acclimatisation, the 2021 heatwave did not result in numerous excessive heat-related deaths, unlike the heatwave observed in July 1987.

As demonstrated above, heatwaves are becoming an increasingly common weather phenomenon, posing a threat to human health (Ballester et al. 2023) and life, as well as various sectors of the economy. Given the rapidly advancing climate warming, it is pertinent to analyse the occurrence of heatwaves in recent decades. Therefore, this study aimed to characterise the occurrence of heatwaves and the bioclimatic conditions during their occurrence in two different regions of Europe from 1966 to 2022.

This study was based on data on maximum daily air temperature (T_max), as well as air temperature and humidity at 12:00 UTC, from two stations: Poznań (Poland) and Thessaloniki (Greece), spanning the period from 1966 to 2022 (Fig. 1). Data for the Polish station were obtained from the Institute of Meteorology and Water Management–National Research Institute, while data for the Greek station were sourced from the Department of Meteorology and Climatology, School of Geology, Aristotle University of Thessaloniki in Greece.

Fig. 1.

Location of meteorological stations.

Poznań is located in western Poland in the region of the Greater Poland Lakeland, which is a large macroregion situated in central Greater Poland and southeastern Kuyavia (Richling et al. 2021). According to the Köppen-Geiger climate classification (kottek et al. 2006), Poznań falls under the Cfb climate type and is characterised as a warm temperate humid climate. Thessaloniki is located in northern Greece, on the Gulf of Thessaloniki (Aegean Sea). It is the second largest city in Greece after Athens. According to the köppen-Geiger climate classification (kottek et al. 2006), Thessaloniki is classified under the Cfa climate type and is characterised as a subtropical humid climate.

Based on the acquired data, the average maximum air temperature was calculated for the summer season and each month within the season. Subsequently, changes in T_max were examined over the multi-year study period, and their statistical significance was determined at the 0.05 level using the Student’s t-test. In this study, the summer season was defined as the period from May to September.

Hot days were identified and analysed for all seasons. The direction and statistical significance of changes in the number of hot days were determined (at the 0.05 level) using the non-parametric Mann–kendall test. A hot day was defined as a day with T_max >90th percentile for that specific day. Following this, based on the identified hot days, heatwaves were defined as sequences of at least three consecutive hot days. Similar definitions were used in other studies (Lhotka, kyselý 2015, Tomczyk, Bednorz 2016). Subsequently, the occurrence of heatwaves was analysed across individual months and decades (the most recent 7-year period), and heatwaves were categorised by their duration.

To characterise biometeorological conditions, Humidex was utilised. This index considers two fundamental meteorological elements: air temperature and relative humidity (RH). It serves as a measure of perceived temperature, especially during hot days (koźmiński, Michalska 2013). One drawback of this index is its exclusion of factors like solar radiation and wind speed, which can intensify or alleviate human thermal sensation. However, in situations where there is insufficient input data to calculate more advanced indicators, this simplicity of Humidex becomes its advantage. Humidex has found adoption in various global regions, e.g. Europe (Scoccimarro et al. 2017), the Czech Republic (Středová et al. 2015) and Greece (Giannopoulou et al. 2014).

The Bioklima 2.6 program (Błażejczyk, Błażejczyk 2006) was employed to calculate the index. Humidex is reported in °C and categorises thermal discomfort into five levels (Table 1).

The index is calculated based on the following formula: Humidex =t + 0.5555 × (ρ − 10) $${\rm{Humidex}}\, = t\, + \,0.5555\, \times \,\left( {{\rm{\rho }}\, - \,10} \right)$$

where: –

t – air temperature in °C,

Humidex calculations used approximations to express vapour pressure of water (ρ, hPa) by RH (RH, %) according to the following formula: ρ = 6112 × 10[(7.5t) / (237.7 + t)] RH / 100 $${\rm{\rho }}\, = \,6112\, \times \,10\left[ {\left( {7.5t} \right)\,/\,\left( {237.7\, + \,t} \right)} \right]\,{\rm{RH}}\,/\,100$$

where: –

t – air temperature in °C,

In this part of the study, the Humidex index was calculated for each day which comprised heatwaves over the multi-year period at 12:00 UTC. This hour is commonly used in biometeorological research because it typically coincides with peak human activity during the day. Subsequently, the average Humidex index value was computed for each heatwave, and the thermal discomfort categories were determined based on the index values during heatwaves.

This study revealed a significant variation in thermal conditions over the examined period. Both stations showed a statistically significant increase in average T_max, although changes were more intense in Poznań. As demonstrated by Ustrnul et al. (2021), the largest changes in maximum air temperature in Poland from 1951 to 2018 were observed in spring and summer, although the changes in summer were slightly smaller (0.40°C/10 years), with the most intense changes occurring in southern Poland. Zoumakis et al. (2012) also observed ongoing summer warming in Thessaloniki. A similar trend was found in Athens, where the rate of change for maximum air temperature was 0.56°C/10 years, although it was lower than the rate for minimum air temperature (Founda et al. 2022). An increase in average T_max was noted in each of the analysed months. The weakest changes were recorded in May at both stations, which were not statistically significant in Poznań. Conversely, the most intense warming occurred in August at both stations as well as in June in Poznań.

The progressive warming has led to an increasing frequency of hot days. This study showed a statistically significant increase in the number of analysed days at both stations, with changes being more intense in Thessaloniki. These findings are consistent with previous studies, indicating an increase in the number of heatwaves and their duration in many regions of Europe (Tomczyk et al. 2019b). Using Athens as an example and considering the occurrence of heatwaves from the beginning of the 20th century to 2020, Founda et al. (2022) demonstrated that the most heatwave cases and the longest durations were observed in the last 30 years (1991–2020). In the coming decades, further increases in the number of hot days in Poland are expected (Tomczyk et al. 2022). The most intense changes are forecasted for the central and southern parts of the country, where by the end of the 21st century, an increase of several days compared to the reference period can be expected. A similar situation will develop in southern Europe. It is estimated that by 2050, the frequency of hot days will also increase, and Thessaloniki will experience up to 20 additional consecutive hot days and nearly an extra month of night-time temperatures >20°C, leading to increased discomfort and health problems for the local population (Giannakopoulos et al. 2011).

Comparing the occurrence of hot days and heatwaves at both stations, it can be observed that in Thessaloniki, single instances or 2-day sequences of hot days were more frequent, resulting in a smaller number of heatwaves and shorter durations compared to Poznań. In this station, the highest number of heatwaves and the longest cumulative duration were recorded in the last 7-year period included in the analysis, while in Thessaloniki, this occurred between 2006 and 2015. In both stations, the longest cumulative duration of heatwaves was observed during the hottest seasons, specifically in Poznań in 2018 and Thessaloniki in 2012. In many regions of Central Europe, the year 2018 stood out compared to the last few decades (Twardosz 2019, Tomczyk, Bednorz 2020) and even compared to the period since the second half of the 19th century (Hoy et al. 2020). In that year, positive air temperature anomalies persisted in Poland in January and from April to December, with the highest anomalies occurring in April (Tomczyk, Bednorz 2020). According to Hoy et al. (2020), the year 2012 was notable in many regions of southeastern Europe for the number of hot days and tropical nights compared to the multi-year period. Founda et al. (2022) demonstrated a fivefold increase in the number of heatwaves and heatwave days in Athens after 1990 compared to their average values in the previous period from 1900 to 1990.

During the study period, 3-day heatwaves were the most frequently recorded, followed by 4-day and longer heatwaves. Both in Poznań and Thessaloniki, the longest heatwave lasted for 11 days. In Poznań, this occurred in 1994, and in Thessaloniki, it occurred in 2010. As shown in numerous studies on Poland, the heatwave observed in 1994, along with the heatwave observed in 2015, are considered examples of extreme heatwaves that have occurred in this part of Europe (Krzyżewska, Dyer 2018, Wibig 2018, Tomczyk et al. 2019a, Tomczyk, Bednorz 2023). The heatwave observed in 1994 was recorded over a significant area of Central Europe, including Germany, the Czech Republic and Ukraine (Shevchenko et al. 2014, 2022, Lhotka, Kyselý 2015). The year 2012 significantly deviated from average conditions in Greece, including Athens, where several heatwave episodes were recorded (Founda, Santamouris 2017).

The studies revealed a significant variation in biometeorological conditions during heatwaves in the analysed stations. In Poznań, only days with slight thermo-humidity discomfort and comfortable were recorded. In contrast, in Thessaloniki, days with slight and then significant thermo-humidity discomfort were most frequently noted, along with days with dangerous discomfort and occasional comfortable days. Previous studies have shown that the western and southwestern regions of Poland experienced the most burdensome biometeorological conditions across the country, both in terms of the number of heatwaves (Krzyżewska et al. 2019, Tomczyk, Bednorz 2023) and the frequency of days with heat stress (Krzyżewska et al. 2019, owczarek 2019, Tomczyk, Owczarek 2020, Tomczyk et al. 2023). Using Poznań as an example, Półrolniczak et al. (2024) demonstrated that biometeorological conditions during heatwaves largely depend on land use. The authors indicated that the most burdensome conditions were recorded in the city centre and in areas with a high proportion of artificial surfaces and a low proportion of natural surfaces, which mitigate thermal conditions.

In the analysed years, a significant variation in thermal conditions was demonstrated in the examined period in both stations. Summer seasons with average T_max below the long-term average were primarily noted until the end of the 20th century, while from the 21st century, seasons with positive anomalies dominated. In Poznań, the highest anomalies were recorded in 2018, and in Thessaloniki they were recorded in 2012.

The progressive warming has led to an increasing frequency of hot days. Similar to T_max, the 21st-century seasons also stood out in terms of the number of hot days compared to the analysed multi-decade period.

During the study period, an increase in the number and duration of heatwaves was observed in both Poznań and Thessaloniki. In both stations, heatwaves were least frequent during the first two decades of the analysed period, and most frequently in the 21st century.

The studies revealed a significant variation in biometeorological conditions during heatwaves in the analysed stations. Heatwaves in Poznań were characterised by less burdensome conditions compared to those in Thessaloniki. In both stations, further climate warming will cause deterioration of biometeorological conditions during heatwaves.