Heatwave climatology in Ukraine: current (1946-2020) and projected (2020-2100)

Climatological and meteorological aspects of heatwaves (HWs) have been extensively studied in various parts of the world, as these extreme weather events have significant harmful effects on both humans and the environment. Several studies have examined HW climatology in Ukraine for specific historical periods using observational air temperature data. However, these results were obtained based on a relatively small number of meteorological stations.

In our study, we calculated HW climatology in Ukraine for both the current historical (1946–2020) and the projected (2020–2100) periods. For the historical period, we utilized the observation-based gridded dataset ClimUAd, which was recently developed for Ukraine. These gridded data are based on a comprehensive collection of instrumental meteorological measurements collected at 178 stations during 1946-2020 across the country. The dataset provides gridded daily time series of four essential climate variables: minimum, mean, and maximum surface air temperature, as well as atmospheric precipitation. The spatial resolution of ClimUAd is 0.1° × 0.1° in both longitude and latitude, enabling the analysis of HW climatology with fine spatial detail across Ukraine.

For the projected period, we applied a mini-statistical ensemble of climate simulations obtained with seven global climate models (GCMs) from the Coupled Model Intercomparison Project Phase 6. GCMs were selected for the ensemble based on two criteria: the use of the Gregorian calendar and a computational grid resolution no courser than 1.5^o×1.5^o. Our analysis incorporated climate projections obtained under two Shared Socioeconomic Pathways (SSP) scenarios: SSP2-4.5 and SSP5-8.5. Prior to calculating HW metrics, all climate projections of surface air temperature were bias-corrected using the quantile delta mapping method. In the bias-correction procedure, ClimUAd data were used along with historical climate simulations for the period 1985-2014.

For both the historical and projected periods, HWs were identified using daily maximum air temperature (TX) data. To detect HWs, we define this extreme weather phenomenon as an event when TX exceeds the 90-th percentile, calculated based on the WMO standard reference period of 1961-1990, for at least three consecutive days, allowing for a one-day gap. This approach is frequently used and widely recommended in studies as the most suitable for HW analysis with pure climatological purposes. The applied definition enables the identification of HWs throughout an entire year. To quantify HW peculiarities, we calculated four HW metrics on a yearly scale: HW number, HW frequency, HW duration, and HW amplitude. All HW calculations were performed using the heatwaveR package.

Our findings revealed a significant increase in all HW metrics during the historical period, with the most pronounced changes observed in the western part of Ukraine. In the projected period, the HW metrics continue to increase at a similar rate for both considered SSPs until approximately the mid-century. However, in the latter part of XXI, changes in HW climatology under SSP5-8.5 differ considerably from those under SSP2-4.5. The SSP5-8.5 scenario indicates that more than half of days in a year at the end of XXI could qualify as HWs.

This work has received funding through the MSCA4Ukraine project, funded by the European Union