Individual weather events, for instance, thunderstorms, windstorms, cold surges, and heat waves, can threaten human life, the natural environment, and infrastructure. Examples are large-scale flood events, e.g., in Pakistan in 2022 with more than 1,100 deaths and 33 million people affected (Nanditha et al. 2022) and in 2010 with more than 18 million people affected (Akhtar et al. 2011), forest damage caused by windstorms (Gregow et al. 2017), power outages due to ice storms (Armenakis 2014), and increased mortality due to heat waves (Ballester et al. 2011). A lot of research has been devoted to better understanding the dynamics of such high-impact weather events (e.g., Hoskins and Berrisford 1988; Lackmann 2001; Wernli et al. 2002; Schneidereit et al. 2012) and the World Meteorological Organization is therefore running a major initiative on high-impact weather with the main goal of disaster risk reduction (Jones and Golding 2014). However, for certain economic sectors, the seasonal accumulation of weather events may matter more than a single short-term extreme event. A first example is the reinsurance industry, for which accumulated losses due to a clustering of storms within a season can become critical (e.g., Mailier et al. 2006; Dacre and Pinto 2020). Other examples are seasonal averaged snow conditions and their impact on ski tourism (e.g., Hennessy et al. 2008), the effects of very dry winters for water supply (e.g., Goodrich and Ellis 2008) and of very dry and hot summers for forest vitality (e.g., Hermann et al. 2022). Therefore, extreme seasons like the windiest season (e.g., due to maximum storm clustering) or the driest and wettest season become particularly relevant, and the severity of their impact might increase with their spatial scale, because of a potentially increased exposure of humans, infrastructure, and ecosystems. Despite the important societal implications, systematic research on extreme seasons, their intensity, scale, and involved meteorology, has so far been limited.
Previous research on extreme seasons focused mainly on case studies. A pioneering study by Namias (1978) addressed the causes of the “abnormal winter” 1976/77, which was mainly characterized by very cold conditions over eastern North America and related it to a persistent El Niño and an intensification of extratropical cyclones. Ulbrich et al. (2001) investigated the large-scale conditions in the stormy European winter 1999/2000. The unprecedented hot summers in central Europe in 2003 and in Russia in 2010 triggered novel research on the role of atmosphere–soil interaction in creating persistent flow anomalies (e.g., Schär et al. 2004; Fischer et al. 2007; Seneviratne et al. 2010). In contrast, Dole et al. (2014), studying the extremely warm spring 2012 in the United States, and Davies (2015), investigating the anomalous winter 2013/14 with very cold conditions in the eastern United States and severe wet and stormy conditions over parts of Europe, emphasized the key role of transient weather variability in contrast to external factors. A final example is the cold European winter 2009/10, which was characterized by an unusually zonal North Atlantic jet (Harnik et al. 2014) and an extremely persistent phase of the North Atlantic Oscillation (Cattiaux et al. 2010). Sprenger et al. (2017) showed that this winter had the largest integrated Northern Hemisphere anomalies (in the time period 1980–2014) of jet streams, cyclones, and Rossby wave breakings, highlighting the key role of transient weather systems for anomalous seasons. This indicates the special characteristics of extreme seasons that are potentially influenced by both short-term weather variability (which is largely internal to the atmosphere) and longer-term climate variability that involves multiple components of the climate system. A seamless approach, combining weather and climate expertise is therefore required to make progress in understanding extreme seasons. Moreover, for systematic analyses of extreme seasons (i.e., analyses of a large number of events) an objectively identified set of events is required, which, however, is lacking so far for most variables of interest.
This study applies a recently developed objective approach to identify extreme seasons (Röthlisberger et al. 2021) to global ERA5 reanalysis data for the period 1950–2020. In contrast to the earlier study, which focused on extremely hot summers and cold winters, here six types of extreme seasons will be identified systematically and globally, based on seasonal mean fields of 2-m temperature (T2m), precipitation (P), and 10-m wind gusts (G10m). From both tails of the distributions, we will consider extremely cold and hot, dry and wet, and calm and stormy seasons, respectively. The seasons considered will be March–May (MAM), June–August (JJA), September–November (SON), and December–February (DJF). The identified extreme seasons come in the form of two-dimensional spatial objects that include all connected grid points that fulfill our local return period (LRP) criterion. In total, with our approach, we identified 30,215 extreme season objects, which can be explored by the community with the aid of the web application https://intexseas-explorer.ethz.ch. Our paper describes the identification of extreme season objects, introduces our interactive web page, and demonstrates two examples of research on extreme seasons.
Identification of extreme seasons
The data basis for our catalog of extreme seasons are hourly ERA5 reanalysis fields (Hersbach et al. 2020) from the European Centre for Medium-Range Weather Forecasts (ECMWF) from 1950 to 2020,1 interpolated on a horizontal grid with a resolution of 0.5°. We apply the generic method that was developed by Röthlisberger et al. (2021) to identify extreme seasons in T2m, and we adopt the method to identify extreme seasons also in G10m and P. Subsequently, we briefly describe the main steps of this method and refer to the online supplemental material (https://doi.org/10.1175/BAMS-D-21-0348.2) and Röthlisberger et al. (2021) for details.
The identification of the extreme seasons involves three steps. In a first step we calculate at every grid point the seasonal mean values of T2m, G10m, and P over the meteorological seasons mentioned before. Seasonal anomalies are then derived as deviations from the seasonal climatology over the entire 71-yr period. To avoid signals from the forced climate change trend we remove an estimate of the forced T2m trend, estimated from the multi-model ensemble mean from CMIP6 (Eyring et al. 2016; Brunner et al. 2020) until 2014 from the historical run and extended by the SSP370 scenario. As in Röthlisberger et al. (2021), a 5-yr running mean is applied to smooth the forced trend from CMIP6. For P and G10m no forced trend is considered in this study, for two reasons. First, (climate model based) forced trend estimates are typically more uncertain for these two variables than for temperature (e.g., Deser et al. 2012). Second, the P and G10m forced trends are typically smaller compared to their unforced seasonal variability than for T2M, which implies that considering the forced trend is less crucial for G10m and P than for T2m for the purpose of this study. We therefore decided to keep our methodology simple and easily reproducible, with known caveats (i.e., no forced P and G10m trends considered) rather than introducing potentially large and regionally varying forced trend uncertainties for P and G10m into our extreme season identification scheme.
In the second step, statistical modeling is applied to the empirical distribution of seasonal mean anomalies. Thereby, as in Röthlisberger et al. (2021), for T2m a Yeo–Johnson power-transformed (Yeo and Johnson 2000) normal (Gaussian) distribution is fitted to the ERA5 values. The same distribution is chosen to model seasonal mean for G10m, whereas for P a gamma distribution has been chosen (e.g., Husak et al. 2007). At the two tails of the distributions we determine those seasons as extreme, whose anomalies have a LRP of at least 40 years. For the exact parameterizations of the statistical models, goodness-of-fit analyses and further details the interested reader is referred to Röthlisberger et al. (2021) and the supplemental material.
In the third step we form coherent spatial objects by connecting neighboring grid points that fulfill the criterion of LRP ≥40 years. We only keep those objects whose area size exceeds 105 km2 (corresponding to roughly 300 km × 300 km assuming a quadratic shape, or to about 6 × 6 grid points). For the entire 71 years, this approach leads to different numbers of extreme season objects for the different types and seasons: from 784 extremely cold season objects in DJF to 1,797 extremely wet season objects in JJAs. For these extreme season objects we then calculate characteristic measures such as their total area, the land area, the area weighted mean anomaly or the spatial integral of the seasonal mean anomalies over all or only the land grid points. The above characteristics can then be used to select, for instance, the largest or the most intense extreme season objects.
Examples of extreme seasons
As a first example, the 10 coldest and driest winter objects in the extratropics are shown in Fig. 1, with the ranking based on the area-integrated anomalies over land, and the additional criterion that the center of mass is located poleward of 20°. The winter of year Y corresponds to the period from 1 December in Y − 1 to the end of February in Y in the Northern Hemisphere (NH), and to from 1 June to 30 August in Y in the Southern Hemisphere (SH).
Extreme season objects of the 10 (a) coldest and (b) driest winters globally, according to the ranking criterion of area-integrated T2m and P anomalies. The year Y in the Northern Hemisphere (NH) DJF refers to the period from 1 December in Y − 1 until the end of February in Y.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Extreme season objects of the 10 (a) coldest and (b) driest winters globally, according to the ranking criterion of area-integrated T2m and P anomalies. The year Y in the Northern Hemisphere (NH) DJF refers to the period from 1 December in Y − 1 until the end of February in Y.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Extreme season objects of the 10 (a) coldest and (b) driest winters globally, according to the ranking criterion of area-integrated T2m and P anomalies. The year Y in the Northern Hemisphere (NH) DJF refers to the period from 1 December in Y − 1 until the end of February in Y.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
The top 10 cold winter objects cover large parts of the NH and partly overlap (Fig. 1a). Most of them occur in the NH midlatitudes, but one in 1992 extends over the Middle East and one in 1954 over Antarctica. They have an average size of 4,650,896 km2, with a maximum size of 11,622,146 km2 for the object extending over large parts of Russia in 1969. Note that the full objects are depicted in Fig. 1 although the intensity used for the ranking is integrated over the object’s land area only. Two top 10 extreme cold winter objects occurred in each of the years 1954, 1969, and 1979. The two objects of 2010 in Russia and 2014 in North America are also among the 10 largest cold winter objects identified by Röthlisberger et al. (2021) in ERA-Interim reanalyses. Among the here identified top 10 cold winters are well-known events that were described in the scientific literature, such as 1954 in southeastern Europe (Twardosz and Kossowska-Cezak 2016), 1969 in western North America (Green 1969), 1979 in the U.S. Midwest (Wagner 1979), 2010 over large parts of the Northern Hemisphere, here identified as extreme over Russia (Wang et al. 2010), and 2014 in the eastern United States and Canada (Trenary et al. 2015; Yu and Zhang 2015; Davies 2015; Wolter et al. 2015). Figure 2 illustrates some of the potentially dramatic consequences of such extremely cold winters: in 1954 the Bosporus was for the last time totally frozen (Fig. 2a), which according to Yavuz et al. (2007) occurred only 12 times in the last 2,000 years, and in 2014 the Niagara Falls were partly frozen (Fig. 2b).
Photos from extreme winters: (a) the frozen Bosporus in February 1954 (position 3 in Fig. 1 from Üster 2000) and (b) eastern North America in January 2014, showing the frozen Niagara Falls (reproduced with permission from Thomson Reuters, Berlin).
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Photos from extreme winters: (a) the frozen Bosporus in February 1954 (position 3 in Fig. 1 from Üster 2000) and (b) eastern North America in January 2014, showing the frozen Niagara Falls (reproduced with permission from Thomson Reuters, Berlin).
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Photos from extreme winters: (a) the frozen Bosporus in February 1954 (position 3 in Fig. 1 from Üster 2000) and (b) eastern North America in January 2014, showing the frozen Niagara Falls (reproduced with permission from Thomson Reuters, Berlin).
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
The top 10 driest winter objects (again ranked by the precipitation anomaly integrated over the land portions of the objects) are shown as another example (Fig. 1b). They occurred mainly in the extratropics, over Europe, western Asia, and in particular along the North American west coast. Two events are documented in the literature: Glantz (1982) mentioned that months including winter 1977 were the driest on record in Yakima valley in the U.S. state of Washington, and the Canadian drought in winter 2001 was reported by Wheaton et al. (2008). It is interesting that overlapping extreme objects of different types can occur simultaneously, as, for instance, the cold and dry extreme season objects in 1969 over western North America and in 1954 over southeastern Europe. Such cooccurrences of extreme seasons can be regarded as a special type of compound events (e.g., Zscheischler et al. 2020; Ridder et al. 2020; Messori et al. 2021), which are often associated with important socioeconomic impacts.
The ERA5 extreme seasons explorer
This new approach to objectively identify extreme seasons has potential for insightful investigations about extreme events at the interface of weather and climate dynamics. Two illustrative example applications will be presented in the next section. However, in order to make this investigation a potential community effort, we decided to develop an openly accessible web page that offers access to extreme season objects identified as outlined in the “Identification of extreme seasons” section above. All extreme season objects of T2m, G10m, and P, identified from 1950 to 2020 in ERA5 are accessible via the interactive web page https://intexseas-explorer.ethz.ch where all objects can be visualized (Fig. 3). The interested user is invited to browse through the variety of extreme season objects by choosing several selection criteria. They include the extreme season type (e.g., cold or hot extremes), the season (e.g., DJF or MAM), the time period (the full 71 years or a subperiod), and the spatial domain (global or any rectangular section). An important choice is then also the criterion used for the ranking as described in step three in the “Identification of extreme seasons” section (e.g., land area affected).
Preview of the extreme seasons explorer web page (https://intexseas-explorer.ethz.ch).
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Preview of the extreme seasons explorer web page (https://intexseas-explorer.ethz.ch).
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Preview of the extreme seasons explorer web page (https://intexseas-explorer.ethz.ch).
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
The web page provides not only a visualization as shown in Fig. 3, but also allows one to download the selected objects as a list including their characteristics, as well as the masks of the complete set of extreme seasons as spatial objects in the format of netCDF files. Furthermore, the web page provides a literature list of media reports and, if available, scientific publications that are related to the selected extreme season objects. This list cannot be considered “complete” but provides a wealth of valuable information about these historic events and documents the public and scientific interest in and fascination about extreme seasons. As background information, we also provide the raw data our extreme season objects are based on, which are the seasonal anomalies, the parameter of the statistical models, and the local return periods (also in netCDF format).
Applications
Here we present two examples of analyses about the characteristics and dynamics of extreme seasons. For both of them we made a selection of extreme seasons based on the measure of area-integrated anomalies over land. The first is a case study of overlapping extreme season objects, i.e., of a seasonal compound event, and the second one a climatological, i.e., systematic analysis of extratropical cyclone frequency anomalies and intensities during different types of extreme seasons. For both applications, 6-hourly cyclone and anticyclone masks are identified in the ERA5 sea level pressure (SLP) field bounded by the outermost closed contour (Wernli and Schwierz 2006; Sprenger et al. 2017). From these 6-hourly masks we calculate seasonal means and the anomalies of each season with respect to the total seasonal ERA5 climatology.
Case study of a triple-compound extreme winter.
The event we present as a case study corresponds to the only season with an overlap of three globally identified top 10 extreme season objects of different types in the considered time period. This occurred in winter 1953/54 with overlapping dry, cold, and calm top 10 extreme season objects that extend from the Baltic Sea and the Balkans in the west to the Urals and Caspian Sea in the east (Figs. 1 and 4a). The overlap of the three objects occurs along a band near 48°N from Ukraine to western Kazakhstan. Most striking is the extraordinary dryness that made this winter the driest globally if measured by the object-accumulated P anomaly. Precipitation was reduced by up to 50% compared to the climatological seasonal mean in the area of the dry extreme season object and the dryness was noticeable also over large parts of western and northern Europe. Sheffield et al. (2009) called this the worst drought between 1950 and 2000 in Europe according to their severity–area–duration (SAD) analysis.
(a) Precipitation in winter 1953/54 relative to climatology (PDJF53/54/PDJFclim; colors; %), and the three overlapping dry, cold, and calm extreme season objects (contours; see legend), and (b) time series in DJF 1953/54 of, from top to bottom, P (mm day−1), T2m (K), and G10m (m s−1), where these variables are averaged over the respective extreme season object; SLP (hPa) averaged over the region covered by all three objects, and frequencies of cyclones and anticyclones (% of the area covered by all three objects). Black contours show daily means, gray dashed lines the DJF climatological daily mean, and gray shading the range between the 5th and 95th percentiles.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
(a) Precipitation in winter 1953/54 relative to climatology (PDJF53/54/PDJFclim; colors; %), and the three overlapping dry, cold, and calm extreme season objects (contours; see legend), and (b) time series in DJF 1953/54 of, from top to bottom, P (mm day−1), T2m (K), and G10m (m s−1), where these variables are averaged over the respective extreme season object; SLP (hPa) averaged over the region covered by all three objects, and frequencies of cyclones and anticyclones (% of the area covered by all three objects). Black contours show daily means, gray dashed lines the DJF climatological daily mean, and gray shading the range between the 5th and 95th percentiles.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
(a) Precipitation in winter 1953/54 relative to climatology (PDJF53/54/PDJFclim; colors; %), and the three overlapping dry, cold, and calm extreme season objects (contours; see legend), and (b) time series in DJF 1953/54 of, from top to bottom, P (mm day−1), T2m (K), and G10m (m s−1), where these variables are averaged over the respective extreme season object; SLP (hPa) averaged over the region covered by all three objects, and frequencies of cyclones and anticyclones (% of the area covered by all three objects). Black contours show daily means, gray dashed lines the DJF climatological daily mean, and gray shading the range between the 5th and 95th percentiles.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
The calm extreme season object mostly coincides with the dry object (Fig. 4a). The cold extreme season object, however, occurred farther south and only overlaps with its northern part with the other two. In its southern part the coldness was so extreme that the Bosporus froze completely, which has not happened since. The photo in Fig. 2a was taken at the end of February 1954 (Yavuz et al. 2007). In a ranking of cold winters in Europe based on observational data, winter 1953/54 appears as the largest negative temperature anomaly and the second largest cold area in the time period 1951–2010 (Twardosz and Kossowska-Cezak, 2016). Interestingly, a sharp P gradient divides the cold extreme season object, and its southern part even experienced more precipitation than average (Fig. 4a). For instance, a heavy snowfall event in Romania was documented in early February 1954 (see weblinks and literature list on our interactive web page).
Time series of P, T2m, and G10m reveal that two longer periods determine the unusualness of this extreme winter, the first from 1 December 1953 to 2 January 1954 and the second from 21 January to 28 February 1954 (Fig. 4b). The daily object-averaged values of P and G10m were mostly below average and some are near the 5% climatologically driest and calmest days, respectively. At the same time, the object-averaged value of T2m was almost consistently below 0°C, and 19 days of this season even belong to the climatologically coldest 5% for this region in ERA5 (Fig. 4b). The winter of 1953/54 led to extremely long-lasting local cold spells in southeastern Europe: for instance, Unkašević and Tošić (2015) reported the second longest cold spell (20 days) in Serbia between 1949 and 2012. Lhotka and Kyselý (2015) identified the winter 1953/54 among the three coldest in central Europe in terms of cold spell duration and spatial extent. The two exceptional periods at the beginning and end of the winter are characterized by higher than normal SLP. In mid-December, SLP values were almost 20 hPa larger than climatology. The phases with high SLP coincide with a lack of cyclones and the prevalence of anticyclones in the area of the three extreme season objects (Fig. 4b). Between these two periods, i.e., in the first half of January, cyclones and anticyclones alternated, leading to more normal weather conditions in terms of P and G10m while T2m remained mainly below climatology.
Maps of seasonal mean anomalies with respect to the 1950–2021 climatology help further exploring of the synoptic situation in this extreme winter. The cold T2m anomaly covers a large area beyond the triple-compound object and even most of Europe and northern Africa with values of up to −7 K inside the cold extreme season object (Fig. 5a). A large and intense positive SLP anomaly is located with its core over the eastern Baltic Sea, indicating cold air advection toward the region of interest from the northeast. In contrast, a weak negative SLP anomaly occurs over the western Mediterranean, affecting the western rim of the cold season object.
Synoptic situation in the extraordinary winter 1953/54 shown by seasonal-mean anomalies relative to the 1950–2021 climatology: (a) T2m (colors; K), and SLP (black contours, negative values dashed; hPa); (b) anticyclone frequency (colors; %), and PV at 315 K (red contours, negative values dashed; PVU, where 1 PVU = 10−6 K kg−1 m2 s−1); (c) cyclone frequency (colors; %), and wind speed at 300 hPa (blue contours, negative values dashed; m s−1); (d) Hovmöller diagram of the meridional wind at 300 hPa (colors; m s−1), T2m anomaly (orange; −5 K), and SLP ≥ 1,025 hPa (hatched) averaged between 40° and 60°N. The pale red, purple, and green contours in (a)–(c) outline the extreme season objects as in Fig. 4.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Synoptic situation in the extraordinary winter 1953/54 shown by seasonal-mean anomalies relative to the 1950–2021 climatology: (a) T2m (colors; K), and SLP (black contours, negative values dashed; hPa); (b) anticyclone frequency (colors; %), and PV at 315 K (red contours, negative values dashed; PVU, where 1 PVU = 10−6 K kg−1 m2 s−1); (c) cyclone frequency (colors; %), and wind speed at 300 hPa (blue contours, negative values dashed; m s−1); (d) Hovmöller diagram of the meridional wind at 300 hPa (colors; m s−1), T2m anomaly (orange; −5 K), and SLP ≥ 1,025 hPa (hatched) averaged between 40° and 60°N. The pale red, purple, and green contours in (a)–(c) outline the extreme season objects as in Fig. 4.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Synoptic situation in the extraordinary winter 1953/54 shown by seasonal-mean anomalies relative to the 1950–2021 climatology: (a) T2m (colors; K), and SLP (black contours, negative values dashed; hPa); (b) anticyclone frequency (colors; %), and PV at 315 K (red contours, negative values dashed; PVU, where 1 PVU = 10−6 K kg−1 m2 s−1); (c) cyclone frequency (colors; %), and wind speed at 300 hPa (blue contours, negative values dashed; m s−1); (d) Hovmöller diagram of the meridional wind at 300 hPa (colors; m s−1), T2m anomaly (orange; −5 K), and SLP ≥ 1,025 hPa (hatched) averaged between 40° and 60°N. The pale red, purple, and green contours in (a)–(c) outline the extreme season objects as in Fig. 4.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
At upper levels a positive anomaly of potential vorticity (PV) at 315 K extends from the Mediterranean eastward to beyond the Caspian Sea and forms a meridionally oriented PV anomaly dipole with a negative PV anomaly north of Scandinavia (Fig. 5b). These PV anomalies indicate a blocked situation over northern Europe (e.g., Kautz et al. 2022) and frequent or persistent cutoff lows over southern Europe. Such meridionally oriented PV anomaly dipoles correspond to potentially long-lasting stationary flow situations as the mutual interaction of the dipole opposes the mean westerly flow. The extreme season objects are located essentially in the center of this upper-level dipole, in a region where the frequency of surface anticyclones is enhanced by more than 20% compared to climatology (Fig. 5b). Consistently, the frequency of cyclones is reduced over northern Europe in this winter season (Fig. 5c). However, a strongly positive cyclone frequency anomaly is apparent over Italy, potentially contributing to the wet conditions in the western part of the cold objects (e.g., the snowfall event in Romania mentioned above).
Considering seasonal-mean flow anomalies farther upstream over the North Atlantic indicates, qualitatively consistent with the SLP anomalies (Fig. 5a), a considerable poleward shift of the 300-hPa jet and strongly reduced winds in the region of the extreme season objects (Fig. 5c). This poleward jet shift goes along with a band of reduced cyclone activity from Newfoundland to Scandinavia and strongly enhanced cyclone frequencies of up to 30% in the Greenland and Barents Seas. The subtropical jet over northern Africa is particularly strong in DJF 1953/54, likely strengthened by frequent Rossby wave breaking into the Mediterranean (Martius et al. 2010) as indicated by the band of increased 315-K PV over the Mediterranean north of the subtropical jet, as discussed above (Fig. 5b).
In line with the seasonal-mean conditions discussed so far is that a large number of temporally organized transient Rossby wave packets occurred during this winter. The Hovmöller diagram of 300-hPa meridional wind (Fig. 5d) reveals a recurrent pattern with ridges over the North American west coast and the central North Atlantic and a trough over the North American east coast. Such recurrent Rossby waves are known to contribute to a quasi-stationary flow pattern and potentially to the persistence of cold, hot, wet, and dry spells (Röthlisberger et al. 2019; Ali et al. 2021). Here they act as precursors to the extremely cold, dry, and calm conditions over eastern Europe. In the region of the extreme season objects from 10° to 60°E, the meridional wind is comparatively weak throughout the season, consistent with the upper-level PV dipole mentioned above.
Finally, we briefly discuss the role of the North Atlantic Oscillation (NAO) and the Siberian high in this winter. The cold temperatures over large parts of Europe accompanied by the blocked flow situation resemble situations usually associated with a negative NAO index (e.g., Rogers 1990; Hurrell 1995; Visbeck et al. 2001; Hurrell et al. 2003; Pinto and Raible 2012; Hanna and Cropper 2017). However, in DJF 1953/54 the NAO index has small positive monthly values from 0.1 to 0.57 indicating a slightly enhanced surface pressure difference between Iceland and the Azores (in agreement with Fig. 5a). However, the positive seasonal-mean SLP anomaly over the Baltic Sea goes along with occasional westward extensions of the Siberian high, as indicated by the hatched regions in Fig. 5d. Especially in the middle of December and in February high pressure reaches across Europe and partially overlaps with the cold anomaly (Fig. 5d). The surface anticyclone almost continually occurs near 100°E, which is considered to be the center of the Siberian high (Sahsamanoglou et al. 1991; Panagiotopoulos et al. 2005), with long-lasting westward extensions into the region of the triple-compound extreme season. Makorgiannis et al. (1981) investigated a sample of 20 individual days between 1968 and 1972 when the Siberian high affected southeastern Europe and required that (i) surface isobars enclose the Siberian high, (ii) a midlevel cyclonic circulation extends over southeastern Europe, and (iii) cold advection occurs in the same area. These conditions are also fulfilled on average in the 1953/54 extreme winter. Cold surges in Europe related to the Siberian high received surprisingly little attention compared to European winter cold extremes associated with a negative NAO. Among these rare case studies are the cold European winter 1962/63 (Hirschi and Sinha 2007), the anomalously cold temperatures in January 2012 in Greece (Tolika et al. 2014), and the cold January 2017 over the Balkan Peninsula (Anagnostopoulou et al. 2017). These studies suggested that the far westward extended Siberian high can help maintain the blocked situation over Europe.
In summary, the extreme winter 1953/54 in eastern Europe and western Russia was characterized by a poleward shift of the North Atlantic jet and cyclone activity, and therefore reduced cyclone frequencies, strongly increased anticyclone frequencies, and strongly enhanced SLP values over northern and eastern Europe also due to repeated westward extensions of the Siberian high. These exceptionally persistent surface anomalies were most likely imposed by a pattern of recurrent Rossby waves and the formation of a prominent meridionally oriented PV dipole at upper levels. More generally, this case study illustrates, on the one hand, the value of studying weather systems (here cyclones, anticyclones, and Rossby waves) in order to understand extreme events on the longer, seasonal time scale, and on the other hand, the potential complexity of processes at this interface of weather and climate dynamics.
Importance of cyclones for extreme seasons.
The second example application addresses the question whether certain types of extreme seasons are systematically associated with more or fewer cyclones occurring in the region of the extreme season objects, and/or with more intense or weaker cyclones. A plausible hypothesis may be that cyclone frequencies are (strongly) enhanced during extremely stormy and wet seasons, and another hypothesis is that cyclones are more intense during stormy seasons, but not necessarily also during extreme wet seasons—hypotheses that can be examined quantitatively with the extreme season objects presented in this study. To investigate this systematically, cyclone frequency anomalies are averaged within all top 50 extreme season objects of the different types (i.e., wet and dry, hot and cold, stormy and calm extreme seasons in DJF, MAM, JJA, and SON, respectively) poleward of 20° latitude. Figure 6a shows the resulting cyclone frequency anomalies in the form of box-and-whisker plots for each extreme season type, and the types are sorted according to their median cyclone frequency anomaly. The top 10 seasons of each type are highlighted with red circles, and bold blue circles indicate the three extreme season objects in winter 1953/54 discussed above.
Cyclone statistics for the top 50 extreme seasons of different types ranked by the spatial integral of the seasonal mean values over all object’s land grid points, in the form of box-and-whiskers diagrams: (a) object-average cyclone frequency anomalies, where the top 10 extreme seasons are marked by red circles, and blue circles mark the extreme seasons in DJF 1953/54 discussed in the previous section, and (b) the distribution of minimum SLP (hPa) of all cyclones that overlap with the extreme season object during the considered season. The dashed horizontal lines mark the highest and lowest median SLP.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Cyclone statistics for the top 50 extreme seasons of different types ranked by the spatial integral of the seasonal mean values over all object’s land grid points, in the form of box-and-whiskers diagrams: (a) object-average cyclone frequency anomalies, where the top 10 extreme seasons are marked by red circles, and blue circles mark the extreme seasons in DJF 1953/54 discussed in the previous section, and (b) the distribution of minimum SLP (hPa) of all cyclones that overlap with the extreme season object during the considered season. The dashed horizontal lines mark the highest and lowest median SLP.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Cyclone statistics for the top 50 extreme seasons of different types ranked by the spatial integral of the seasonal mean values over all object’s land grid points, in the form of box-and-whiskers diagrams: (a) object-average cyclone frequency anomalies, where the top 10 extreme seasons are marked by red circles, and blue circles mark the extreme seasons in DJF 1953/54 discussed in the previous section, and (b) the distribution of minimum SLP (hPa) of all cyclones that overlap with the extreme season object during the considered season. The dashed horizontal lines mark the highest and lowest median SLP.
Citation: Bulletin of the American Meteorological Society 104, 3; 10.1175/BAMS-D-21-0348.1
Clearly, the dry and wet extreme season objects are associated with the largest cyclone frequency anomalies on both ends of the scale, with less than normal cyclones for dry and more for wet objects. This indicates that cyclones are important ingredients of extreme wet and dry seasons, at least in most seasons. The dry extreme season over eastern Europe in winter 1953/54 fits nicely with the statistics with a strongly negative cyclone frequency anomaly, also shown in Fig. 5c. Exceptions are dry summers (with comparatively many cyclones, though still near or below average) and wet springs (with comparatively few cyclones). The pattern becomes less robust when only considering the top 10 extreme seasons. For instance, most of the driest winters have very few cyclones; however, most of the driest springs have zero to even slightly positive cyclone frequency anomalies. This indicates that the generally intuitive pattern in dry and wet seasons has very high case-to-case variability, which motivates further research on the exact role of weather systems in the occurrence of extreme seasons and its regional variability.
To investigate whether extreme seasons are also associated systematically with more or less intense cyclones, we show in Fig. 6b the statistics of minimum SLP of all cyclones that overlap with an extreme season object. This evaluation is done with hourly time resolution; i.e., every hour it is checked whether a cyclone mask (Wernli and Schwierz 2006) overlaps with the extreme season object, and if yes, then the minimum SLP value of this cyclone enters the statistics. For extreme dry seasons we find the least intense cyclones with a SLP minimum of 1,010 hPa in the median, which supports the notion of fewer and less intense cyclones for this category. In contrast, the cyclone-rich wet extreme seasons appear in the central portion along the cyclone intensity scale. In the midlatitudes, heavy or long-lasting precipitation is not necessarily associated with particularly intense cyclones as, e.g., in the case study described Grams et al. (2014), and in addition, several objects of this category occur in subtropical regions (not shown) where cyclones are less intense in terms of SLP.
Also according to expectations, but with on average weaker cyclone frequency anomalies are the extreme seasons related to G10m. Calm extreme seasons are associated with negative cyclone anomalies (that are weaker than those for dry seasons, except for summer), and stormy extreme seasons tend to be associated with a positive cyclone frequency anomaly, but here the median anomalies are surprisingly weak. On the cyclone intensity scale, the four stormy extreme seasons take the top four positions, with the lowest median SLP below 990 hPa in winter, which reveals that the intensity of cyclones matters more than their frequency for this category of extreme seasons. Extreme calm seasons are associated with few but also surprisingly deep cyclones, indicating that for calm seasons cyclone frequency is most relevant.
Finally, in this top 50 extreme season analysis, extremely warm and cold seasons do neither show systematic cyclone frequency anomalies nor intensities, but the results of this global-scale analysis could be region dependent. Only extremely hot winters stand out with quite intense cyclones. This may indicate the relevance of warm air advection by comparatively deep cyclone in particular in winter when baroclinicity is highest. Thus, according to our results, the role of cyclones for temperature extremes on the seasonal time scale is more subtle than for wind or precipitation and motivates further research on this topic. The purpose of this brief analysis in this paper is twofold: On the one hand it illustrates the kind of systematic analyses on extreme seasons that is enabled by the presented extreme season catalog. On the other hand it verifies prior expectations about the role of cyclones for extreme seasons while at the same time documenting their large case-to-case variability in this role.
Summary
Research on extreme seasons, i.e., meteorological extreme events on the seasonal time scale, is relevant because their impact is potentially different than that of the more commonly studied, shorter time-scale extremes. Extreme seasons are thus an emerging research topic at the weather–climate interface. However, previous studies have almost exclusively investigated single case studies in specific regions of interest (e.g., a particularly cold winter in North America) and a more general and objective event definition as the basis for more systematic analyses of numerous such events is lacking so far. In this study, we identified extreme seasons systematically, based on statistical modeling, as contiguous regions of extreme seasonal-mean values of near-surface temperature, wind, or precipitation. We provide a fully open-access, global catalog of our extreme seasons based on ERA5, which can be accessed through the extreme seasons explorer https://intexseas-explorer.ethz.ch. On this web portal, extreme seasons of different types and in different regions can be visualized. This application was developed with the intention to enable and facilitate more research on extreme seasons by the broader community. This study also shows examples which document different kinds of analyses (a detailed case study and a systematic investigation) that are made possible by the extreme season catalog. Given the scientific background of the authors, these example analyses focused on the role of weather systems for extreme seasons to occur; however, a wide palette of further analyses regarding the causes and impacts of extreme seasons is conceivable. This palette includes, for instance, ecosystem dynamics during extreme seasons, health impacts of extreme seasons, quantification of the relevance of extreme seasons for the (re)insurance business or energy markets, and the linkage between the occurrence of extreme seasons and large-scale modes of atmospheric variability. We therefore invite peers to participate in extreme season research and hope that this study will serve as a useful starting point for improving the general understanding of the physical processes behind and socioeconomic consequences of extreme seasons.
For DJF we use December 1950–February 2021.
Acknowledgments.
We thank Russ Schumacher and two anonymous reviewers for their constructive feedback. We are most grateful to Urs Beyerle and Mathias Hauser for there help with setting up the extreme seasons web page at our institute and to all colleagues in the INTEXseas project for fruitful discussions. This research has been supported by the European Research Council, Horizon 2020 research and innovation programme (INTEXseas, Grant 787652).
Data availability statement.
All extreme season objects identified with our approach in ERA5 are available in netCDF format via the extreme seasons explorer web page.
References
Akhtar, S., and Coauthors, 2011: The South Asiatic monsoon and flood hazards in the Indus River basin, Pakistan. J. Basic Appl. Sci., 7, 101–115.
Ali, S. M., O. Martius, and M. Röthlisberger, 2021: Recurrent Rossby wave packets modulate the persistence of dry and wet spells across the globe. Geophys. Res. Lett., 48, e2020GL091452, https://doi.org/10.1029/2020GL091452.
Anagnostopoulou, C., K. Tolika, G. Lazoglou, and P. Maheras, 2017: The exceptionally cold January of 2017 over the Balkan Peninsula: A climatological and synoptic analysis. Atmosphere, 8, 252, https://doi.org/10.3390/atmos8120252.
Armenakis, N., 2014: Urban impacts of ice storms: Toronto December 2013. Nat. Hazards, 74, 1291–1298, https://doi.org/10.1007/s11069-014-1211-7.
Ballester, J., J.-M. Robine, F. R. Herrmann, and X. Rodó, 2011: Long-term projections and acclimatization scenarios of temperature-related mortality in Europe. Nat. Commun., 2, 358, https://doi.org/10.1038/ncomms1360.
Brunner, L., M. Hauser, R. Lorenz, and U. Beyerle, 2020: The ETH Zurich CMIP6 next generation archive: Technical documentation. ETH Zurich Tech. Doc., 10 pp., https://doi.org/10.5281/zenodo.3734128.
Cattiaux, J., R. Vautard, C. Cassou, P. Yiou, V. Masson-Delmotte, and F. Codron, 2010: Winter 2010 in Europe: A cold extreme in a warming climate. Geophys. Res. Lett., 37, L20704, https://doi.org/10.1029/2010GL044613.
Dacre, H. F., and J. G. Pinto, 2020: Serial clustering of extratropical cyclones: A review of where, when and why it occurs. npj Climate Atmos. Sci., 3, 48, https://doi.org/10.1038/s41612-020-00152-9.
Davies, H. C., 2015: Weather chains during the 2013/2014 winter and their significance for seasonal prediction. Nat. Geosci., 8, 833–837, https://doi.org/10.1038/ngeo2561.
Deser, C., A. Phillips, V. Bourdette, and H. Teng, 2012: Uncertainty in climate change projections: The role of internal variability. Climate Dyn., 38, 527–546, https://doi.org/10.1007/s00382-010-0977-x.
Dole, R., and Coauthors, 2014: The making of an extreme event: Putting the pieces together. Bull. Amer. Meteor. Soc., 95, 427–440, https://doi.org/10.1175/BAMS-D-12-00069.1.
Eyring, V., S. Bony, G. A. Meehl, C. A. Senior, B. Stevens, R. J. Stouffer, and K. E. Taylor, 2016: Overview of the Coupled Model Intercomparison Project phase 6 (CMIP6) experimental design and organization. Geosci. Model Dev., 9, 1937–1958, https://doi.org/10.5194/gmd-9-1937-2016.
Fischer, E. M., S. I. Seneviratne, P. L. Vidale, D. Lüthi, and C. Schär, 2007: Soil moisture–atmosphere interactions during the 2003 European summer heat wave. J. Climate, 20, 5081–5099, https://doi.org/10.1175/JCLI4288.1.
Glantz, M. H., 1982: Consequences and responsibilities in drought forecasting: The case of Yakima, 1977. Water Resour. Res., 18, 3–13, https://doi.org/10.1029/WR018i001p00003.
Goodrich, G. B., and A. W. Ellis, 2008: Climatic controls and hydrologic impacts of a recent extreme seasonal precipitation reversal in Arizona. J. Appl. Meteor. Climatol., 47, 498–508, https://doi.org/10.1175/2007JAMC1627.1.
Grams, C. M., H. Binder, S. Pfahl, N. Piaget, and H. Wernli, 2014: Atmospheric processes triggering the central European floods in June 2013. Nat. Hazards Earth Syst. Sci., 14, 1691–1702, https://doi.org/10.5194/nhess-14-1691-2014.
Green, N., 1969: The winter of 1968–69 in the lower Fraser valley of British Columbia. Atmosphere, 7, 63–65, https://doi.org/10.1080/00046973.1969.9676569.
Gregow, A., A. Laaksonen, and M. E. Alper, 2017: Increasing large scale windstorm damage in western, central and northern European forests, 1951–2010. Sci. Rep., 7, 46397, https://doi.org/10.1038/srep46397.
Hanna, E., and T. E. Cropper, 2017: North Atlantic Oscillation. Oxford Research Encyclopedia of Climate Science, Oxford University Press, https://doi.org/10.1093/acrefore/9780190228620.013.22.
Harnik, N., E. Galanti, O. Martius, and O. Adam, 2014: The anomalous merging of the African and North Atlantic jet streams during the Northern Hemisphere winter of 2010. J. Climate, 27, 7319–7334, https://doi.org/10.1175/JCLI-D-13-00531.1.
Hennessy, K. J., P. H. Whetton, K. Walsh, I. N. Smith, J. M. Bathols, M. Hutchinson, and J. Sharples, 2008: Climate change effects on snow conditions in mainland Australia and adaptation at ski resorts through snowmaking. Climate Res., 35, 255–270, https://doi.org/10.3354/cr00706.
Hermann, M., M. Röthlisberger, A. Gessler, A. Rigling, C. Senf, T. Wohlgemuth, and H. Wernli, 2022: Analysis of multi-seasonal meteorological storylines leading to reduced forest greenness in Europe in 2000–2020. Biogeosciences, https://doi.org/10.5194/egusphere-2022-425.
Hersbach, H., and Coauthors, 2020: The ERA5 global reanalysis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi.org/10.1002/qj.3803.
Hirschi, J. J.-M., and B. Sinha, 2007: Negative NAO and cold Eurasian winters: How exceptional was the winter of 1962/1963? Weather, 62, 43–48, https://doi.org/10.1002/wea.34.
Hoskins, B. J., and P. Berrisford, 1988: A potential vorticity perspective of the storm of 15–16 October 1987. Weather, 43, 122–129, https://doi.org/10.1002/j.1477-8696.1988.tb03890.x.
Hurrell, J. W., 1995: Transient eddy forcing of the rotational flow during northern winter. J. Atmos. Sci., 52, 2286–2301, https://doi.org/10.1175/1520-0469(1995)052<2286:TEFOTR>2.0.CO;2.
Hurrell, J. W., Y. Kushnir, G. Ottersen, and M. Visbeck, 2003: An overview of the North Atlantic Oscillation. The North Atlantic Oscillation: Climatic Significance and Environmental Impact, Geophys. Monogr., Vol. 134, Amer. Geophys. Union, 1–35, https://doi.org/10.1029/134GM01.
Husak, G. J., J. Michaelsen, and C. Funk, 2007: Use of the gamma distribution to represent monthly rainfall in Africa for drought monitoring applications. Int. J. Climatol., 27, 935–944, https://doi.org/10.1002/joc.1441.
Jones, S. C., and B. Golding, 2014: HIWeather: A research activity on high impact weather within the World Weather Research Programme. WWRP Rep., 87 pp.
Kautz, L.-A., O. Martius, S. Pfahl, J. G. Pinto, A. M. Ramos, P. M. Sousa, and T. Woollings, 2022: Atmospheric blocking and weather extremes over the Euro-Atlantic sector—A review. Wea. Climate Dyn., 3, 305–336, https://doi.org/10.5194/wcd-3-305-2022.
Lackmann, G. M., 2001: Analysis of a surprise western New York snowstorm. Wea. Forecasting, 16, 99–116, https://doi.org/10.1175/1520-0434(2001)016<0099:AOASWN>2.0.CO;2.
Lhotka, O., and J. Kyselý, 2015: Characterizing joint effects of spatial extent, temperature magnitude and duration of heat waves and cold spells over central Europe. Int. J. Climatol., 35, 1232–1244, https://doi.org/10.1002/joc.4050.
Mailier, P. J., D. B. Stephenson, C. A. T. Ferro, and K. I. Hodges, 2006: Serial clustering of extratropical cyclones. Mon. Wea. Rev., 134, 2224–2240, https://doi.org/10.1175/MWR3160.1.
Makorgiannis, T. J., B. D. Giles, and A. A. Flocas, 1981: The problem of the extension of the Siberian anticyclone towards southeast Europe, and its relation to atmospheric circulation anomalies over the Northern Hemisphere. Arch. Meteor. Geophys. Bioclimatol., 30A, 185–196, https://doi.org/10.1007/BF02257842.
Martius, O., C. Schwierz, and H. C. Davies, 2010: Tropopause-level waveguides. J. Atmos. Sci., 67, 866–879, https://doi.org/10.1175/2009JAS2995.1.
Messori, G., and Coauthors, 2021: Compound climate events and extremes in the midlatitudes: Dynamics, simulation, and statistical characterization. Bull. Amer. Meteor. Soc., 102, E774–E781, https://doi.org/10.1175/BAMS-D-20-0289.1.
Namias, J., 1978: Multiple causes of the North American abnormal winter 1976–77. Mon. Wea. Rev., 106, 279–295, https://doi.org/10.1175/1520-0493(1978)106<0279:MCOTNA>2.0.CO;2.
Nanditha, J. S., and Coauthors, 2022: The Pakistan flood of August 2022: Causes and implications. ESS Open Archive, accessed 26 October 2022, https://doi.org/10.1002/essoar.10512560.1.
Panagiotopoulos, F., M. Shahgedanova, A. Hannachi, and D. B. Stephenson, 2005: Observed trends and teleconnections of the Siberian high: A recently declining center of action. J. Climate, 18, 1411–1422, https://doi.org/10.1175/JCLI3352.1.
Pinto, J. G., and C. C. Raible, 2012: Past and recent changes in the North Atlantic Oscillation. Wiley Interdiscip. Rev.: Climate Change, 3, 79–90, https://doi.org/10.1002/wcc.150.
Ridder, N. N., and Coauthors, 2020: Global hotspots for the occurrence of compound events. Nat. Commun., 11, 5956, https://doi.org/10.1038/s41467-020-19639-3.
Rogers, J. C., 1990: Patterns of low-frequency monthly sea level pressure variability (1899–1986) and associated wave cyclone frequencies. J. Climate, 3, 1364–1379, https://doi.org/10.1175/1520-0442(1990)003<1364:POLFMS>2.0.CO;2.
Röthlisberger, M., L. Frossard, L. F. Bosart, D. Keyser, and O. Martius, 2019: Recurrent synoptic-scale Rossby wave patterns and their effect on the persistence of cold and hot spells. J. Climate, 32, 3207–3226, https://doi.org/10.1175/JCLI-D-18-0664.1.
Röthlisberger, M., M. Hermann, C. Frei, F. Lehner, E. M. Fischer, R. Knutti, and H. Wernli, 2021: A new framework for identifying and investigating seasonal climate extremes. J. Climate, 34, 7761–7782, https://doi.org/10.1175/JCLI-D-20-0953.1.
Sahsamanoglou, H. S., T. J. Makrogiannis, and P. P. Kallimopoulos, 1991: Some aspects of the basic characteristics of the Siberian anticyclone. Int. J. Climatol., 11, 827–839, https://doi.org/10.1002/joc.3370110803.
Schär, C., P. L. Vidale, D. Lüthi, C. Frei, C. Häberli, M. A. Liniger, and C. Appenzeller, 2004: The role of increasing temperature variability in European summer heatwaves. Nature, 427, 332–336, https://doi.org/10.1038/nature02300.
Schneidereit, A., S. Schubert, P. Vargin, F. Lunkeit, X. Zhu, D. H. W. Peters, and K. Fraedrich, 2012: Large-scale flow and the long-lasting blocking high over Russia: Summer 2010. Mon. Wea. Rev., 140, 2967–2981, https://doi.org/10.1175/MWR-D-11-00249.1.
Seneviratne, S. I., T. Corti, E. L. Davin, M. Hirschi, E. B. Jaeger, I. Lehner, B. Orlowsky, and A. J. Teuling, 2010: Investigating soil moisture–climate interactions in a changing climate: A review. Earth-Sci. Rev., 99, 125–161, https://doi.org/10.1016/j.earscirev.2010.02.004.
Sheffield, J., K. M. Andreadis, E. F. Wood, and D. P. Lettenmaier, 2009: Global and continental drought in the second half of the twentieth century: Severity–area–duration analysis and temporal variability of large-scale events. J. Climate, 22, 1962–1981, https://doi.org/10.1175/2008JCLI2722.1.
Sprenger, M., and Coauthors, 2017: Global climatologies of Eulerian and Lagrangian flow features based on ERA-Interim. Bull. Amer. Meteor. Soc., 98, 1739–1748, https://doi.org/10.1175/BAMS-D-15-00299.1.
Tolika, K., P. Maheras, I. Pytharoulis, and C. Anagnostopoulou, 2014: The anomalous low and high temperatures of 2012 over Greece—An explanation from a meteorological and climatological perspective. Nat. Hazards Earth Syst. Sci., 14, 501–507, https://doi.org/10.5194/nhess-14-501-2014.
Trenary, L., T. DelSole, M. K. Tippett, and B. Doty, 2015: Was the cold eastern US winter of 2014 due to increased variability [in “Explaining Extreme Events of 2014 from a Climate Perspective”]? Bull. Amer. Meteor. Soc., 96 (12), S15–S19, https://doi.org/10.1175/BAMS-D-15-00138.1.
Twardosz, R., and U. Kossowska-Cezak, 2016: Exceptionally cold and mild winters in Europe (1951–2010). Theor. Appl. Climatol., 125, 399–411, https://doi.org/10.1007/s00704-015-1524-9.
Ulbrich, U., A. H. Fink, M. Klawa, and J. G. Pinto, 2001: Three extreme storms over Europe in December 1999. Weather, 56, 70–80, https://doi.org/10.1002/j.1477-8696.2001.tb06540.x.
Unkašević, M., and I. Tošić, 2015: Seasonal analysis of cold and heat waves in Serbia during the period 1949–2012. Theor. Appl. Climatol., 120, 29–40, https://doi.org/10.1007/s00704-014-1154-7.
Üster, C., 2000: Once Upon a Time: Istanbul from Empire to Republic. Türkiye İş Bankası Yayınları, 247 pp.
Visbeck, M. H., J. W. Hurrell, L. Polvani, and H. M. Cullen, 2001: The North Atlantic Oscillation: Past, present, and future. Proc. Natl. Acad. Sci. USA, 98, 12 876–12 877, https://doi.org/10.1073/pnas.231391598.
Wagner, A. J., 1979: Weather and circulation of January 1979: Widespread record cold with heavy snowfall in the Midwest. Mon. Wea. Rev., 107, 499–506, https://doi.org/10.1175/1520-0493(1979)107<0499:WACOJ>2.0.CO;2.
Wang, C., H. Liu, and S.-K. Lee, 2010: The record-breaking cold temperatures during the winter of 2009/2010 in the Northern Hemisphere. Atmos. Sci. Lett., 11, 161–168, https://doi.org/10.1002/asl.278.
Wernli, H., and C. Schwierz, 2006: Surface cyclones in the ERA-40 dataset (1958–2001). Part I: Novel identification method and global climatology. J. Atmos. Sci., 63, 2486–2507, https://doi.org/10.1175/JAS3766.1.
Wernli, H., S. Dirren, M. A. Lininger, and M. Zillig, 2002: Dynamical aspects of the life cycle of the winter storm ‘Lothar’ (24–26 December 1999). Quart. J. Roy. Meteor. Soc., 128, 405–429, https://doi.org/10.1256/003590002321042036.
Wheaton, E., S. Kulshreshtha, V. Wittrock, and G. Koshida, 2008: Dry times: Hard lessons from the Canadian drought of 2001 and 2002. Can. Geogr., 52, 241–262, https://doi.org/10.1111/j.1541-0064.2008.00211.x.
Wolter, K., J. K. Eischeid, X.-W. Quan, T. N. Chase, M. Hoerling, R. M. Dole, G. J. Van Oldenborgh, and J. E. Walsh, 2015: How unusual was the cold winter of 2013/14 in the upper Midwest [in “Explaining Extreme Events of 2014 from a Climate Perspective”]? Bull. Amer. Meteor. Soc., 96 (12), S10–S14, https://doi.org/10.1175/BAMS-D-15-00126.1.
Yavuz, V., N. Akcar, and C. Schlüchter, 2007: The frozen Bosphorus and its paleoclimatic implications based on a summary of the historical data. The Black Sea Flood Question: Changes in Coastline, Climate, and Human Settlement, V. Yanko-Hombach et al., Eds., Springer Netherlands, 633–649, https://doi.org/10.1007/978-1-4020-5302-3_26.
Yeo, I.-K., and R. A. Johnson, 2000: A new family of power transformations to improve normality or symmetry. Biometrika, 87, 954–959, https://doi.org/10.1093/biomet/87.4.954.
Yu, B., and X. Zhang, 2015: A physical analysis of the severe 2013/2014 cold winter in North America. J. Geophys. Res. Atmos., 120, 10 149–10 165, https://doi.org/10.1002/2015JD023116.
Zscheischler, J., and Coauthors, 2020: A typology of compound weather and climate events. Nat. Rev. Earth, 1, 333–347, https://doi.org/10.1038/s43017-020-0060-z.
