Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland

Szymon PorĘba; Mateusz Taszarek; Zbigniew Ustrnul

doi:10.1175/waf-d-21-0099.1

Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland

PorĘba, Szymon; Taszarek, Mateusz; Ustrnul, Zbigniew 2022-08-05 00:00:00 AUGUST 2022 PO R E ˛ BA ET A L . 1447 Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland a,b c,d a,b SZYMON PORE ˛ BA, MATEUSZ TASZAREK, AND ZBIGNIEW USTRNUL Department of Climatology, Jagiellonian University, Krakow, ´ Poland Institute of Meteorology and Water Management, National Research Institute, Warsaw, Poland Department of Meteorology and Climatology, Adam Mickiewicz University, Poznan, Poland National Severe Storms Laboratory, Norman, Oklahoma (Manuscript received 14 June 2021, in ﬁnal form 24 January 2022) ABSTRACT: The relationship between convective parameters derived from ERA5 and cloud-to-ground (CG) lightning ﬂashes from the PERUN network in Poland was evaluated. All ﬂashes detected between 2002 and 2019 were divided into intensity categories based on a peak 1-min CG lightning ﬂash rate and were collocated with proximal proﬁles from ERA5 to assess their climatological variability. Thunderstorms in Poland are the most frequent in July, between 1400 and 1600 UTC and over the southeastern parts of the country. The highest median of most unstable convective available potential energy (MUCAPE) for CG lightning ﬂash events is from June to August, between 1400 and 1600 UTC (around 900 J kg ), whereas patterns in 0–6-km wind shear [deep-layer shear (DLS)] are reversed, with the highest median values during winter and night (around 25 m s ). The best overlap of MUCAPE and DLS (MUWMAXSHEAR parameter) is 2 22 in July–August, typically between 1400 and 2000 UTC with median values of around 850 m s . Thunderstorms in Poland 21 21 are the most frequent in MUCAPE below 1000 J kg , and DLS between 8 and 15 m s . Along with increasing MUCAPE and DLS, peak CG lightning ﬂash rates increase as well. Compared to DLS, MUCAPE is a more important parameter in forecasting any lightning activity, but when these two are combined together (MUWMAXSHEAR) they are more reliable in distinguishing between thunderstorms producing small and high CG lightning ﬂash rates. Our results also indicate that higher CG lightning ﬂash rates result in thunderstorms more frequently associated with severe weather reports (hail, tornado, wind). SIGNIFICANCE STATEMENT: Each year severe thunderstorms produce considerable material losses and lead to deaths across central Europe; thus, a better understanding of local storm climatologies and their accompanying environ- ments is important for operational forecasters, emergency managers, and risk estimation. In this research we address this issue by analyzing 18 years of lightning intensity data and collocated atmospheric environments. Thunderstorms in Poland are the most frequent in July between 1400 and 1600 UTC and form typically in environments with low atmo- spheric instability and moderate vertical shear of the horizontal wind. The probability for storms producing intense lightning increases when both of these environmental parameters reach higher values. KEYWORDS: Deep convection; Lightning; Severe storms; Thunderstorms; Forecasting 1. Introduction events can be observed based on the records from the European Severe Weather Database (ESWD; Dotzek et al. A cloud-to-ground (CG) lightning ﬂash is on average regis- 2009), which also indicate that the most common convective tered on 150–160 days per year in Poland (Taszarek et al. threat across central Europe is severe wind, followed by exces- 2015). Thunderstorms are most frequent during the summer sive rainfall, large hail, damaging lightning, and tornadoes (i.e., June–August) and over the southeastern part of the (Groenemeijer et al. 2017; Taszarek et al. 2019). country (Bielec-Ba ˛kowska 2003). The same period is also Although thunderstorm intensity can be measured in vari- linked to the peak occurrence of severe weather outbreaks ous ways, it can be generally assumed that the presence of such as derechos and tornadic supercells that are responsible severe weather reports, high CG lightning ﬂash rates, and/or for considerable material losses in Poland (Celinski-Mysław high radar reﬂectivity depends largely on the vertical velocity and Matuszko 2014; Pilorz 2015; Widawski and Pilorz 2018; and size of the convective updraft (Apke et al. 2018). Pilguj et al. 2019; Pore Rba and Ustrnul 2020; Surowiecki and Regional research studies that focus on evaluating thunder- Taszarek 2020). A similar distribution of severe weather storm intensity and its relation to atmospheric environments are highly important to operational forecasters, as these stud- ies provide forecasters with guidance on the metrics and cor- Denotes content that is immediately available upon publica- responding values that are useful in predicting severe tion as open access. convective storms over speciﬁc areas, times of year, and times of day. For example, an increase in the availability of low-level moisture, midtropospheric lapse rates, and the Corresponding author: Szymon PorReba, szymon.poreba@doctoral. uj.edu.pl degree of convective organization (e.g., supercells, squall DOI: 10.1175/WAF-D-21-0099.1 Ó 2022 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses). 1448 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 lines) governed by sufﬁcient vertical wind shear leads to stronger updrafts (Doswell 2001; Markowski and Richardson 2010; Coffer and Parker 2015; Dennis and Kumjian 2017; Lin and Kumjian 2021). Despite growing numbers of studies on thunderstorm environments, there are still some limitations associated with challenging identiﬁcation and recognition of thunderstorm intensity (further described in section 2b). Thus, in this study we address this issue by using the peak 1-min CG lightning ﬂash rate as a proxy to categorize thun- derstorms according to their lightning intensity and then intercompare them with collocated atmospheric environ- ments. Although studies on thunderstorm environments in ´ ˇ central Europe were conducted in the past (Pucik et al. 2015; Westermayer et al. 2017; Taszarek et al. 2020), none of them evaluated speciﬁc CG lightning ﬂash rates, which in our approach constitutes a new, not previously applied in Europe, thunderstorm intensity proxy. Similar approaches have been tested for the tropics and subtropics (Liu et al. 2020). A pre- liminary comparison of CG lightning ﬂash rate intensity cate- FIG. 1. Mean number of ESWD severe weather reports per grid gories deﬁned in this study with 14 427 ESWD large hail, with speciﬁc thunderstorm intensity category. PERUN CG light- severe wind, heavy rain, and tornado reports from the terri- ning data and ESWD reports were collocated on 0.258 grids within tory of Poland indicates an increasing relative number of 1-h intervals. For each thunderstorm intensity category grids (deﬁned based on a peak 1-min CG lightning ﬂash rate), proximal reports along with increasing CG lightning ﬂash rates (Fig. 1). ESWD reports were counted and then divided by the total number By evaluating 18 years (2002–19) of CG lightning ﬂash rate of grids representing each category. data from the PERUN lightning detection network, we aim to better understand the climatological aspects of severe thun- derstorms and their accompanying environments across Lightning Detection Network (NLDN; Orville 1991; Orville Poland. An improved resolution of the ﬁfth generation of and Hufﬁnes 2001; Orville et al. 2002; Zajac and Rutledge ECMWF reanalysis (ERA5) allows us to study lightning envi- 2001; Cummins and Murphy 2009; Holle 2014; Holle et al. ronments in the way that was not possible with prior genera- 2016; Kingﬁeld et al. 2017; Koehler 2020). These studies show tions of reanalyses. We use hourly intervals (contrary to 3–6-h that the highest annual mean CG lightning ﬂash density steps in previous studies) to evaluate the diurnal, annual, and occurs over Florida, reaching 16 CG lightning ﬂashes km spatial aspects of convective environments collocated with yr and over the coast of Gulf of Mexico with 8–12 CG light- 22 21 speciﬁc CG lightning ﬂash rates}an aspect rarely explored ning ﬂashes km yr . In the same region the number of before. An auxiliary goal constituting the background of the thunderstorm days is the highest with more than 110 thunder- research is also to present an updated 18-yr climatology of storm days per year (Koehler 2020). CG lightning ﬂashes for Poland. In Europe, research on lighting climatologies has been con- ducted using the Advanced Technology Demonstration Network (ATDnet), European Cooperation for Lightning 2. Prior research Detection (EUCLID), and ZEUS networks (Anderson and a. Lightning climatologies Klugmann 2014; Poelman et al. 2016; Taszarek et al. 2019, Lightning climatologies are based on data from lightning 2020; Enno et al. 2020). These studies showed that peak light- detectors, which can be ground-based and satellite-based, ning ﬂash rate in Europe is in the afternoon (1400–1600 UTC) global or regional. Measurements of global lightning activity during summer. The highest annual lightning ﬂash rates are in 22 21 are applied with the ground-based lightning sensors systems, the Alps and over northeastern Italy reaching 7–8km yr such as World Wide Lightning Location Network (WWLLN; (Anderson and Klugmann 2014; Enno et al. 2020). The highest annual mean number of thunderstorm days is associated with http://wwlln.net) and Vaisala GLD360 (Said et al. 2011)or satellite-based instruments like Lightning Imaging Sensor mountainous regions and Mediterranean boundary layer mois- (LIS; Christian et al. 1999). Both sources conﬁrm that the ture, especially along the Alps, Dinaric Alps, and Apennines highest frequency of lightning occurs over continents in tropi- with values more than 60 (Taszarek et al. 2019). In regional cal areas (Virts et al. 2013; DiGangi et al. 2021). Global studies for selected countries, the results are consistent with ground-based systems make it possible to monitor lightning the previously mentioned distributions with mean annual peak 22 21 CG lightning ﬂash rates reaching 15 ﬂashes km yr over the over a broader area including higher latitudes, but their detec- border between Italy and Slovenia (Schulz et al. 2005), 9 ﬂashes tion efﬁciency is lower than regional or satellite-based systems 22 21 22 21 (Burgesser 2017; DiGangi et al. 2021). km yr in Italy (Feudale et al. 2013), 7.43 ﬂashes km yr ´ ´ Regionally, a large number of lightning climatologies have in the Czech Republic (Novak and Kyznarova 2011), 22 21 been performed for the United States with the National 3.06 ﬂashes km yr over the Carpathians in Romania AUGUST 2022 PO R E ˛ BA ET A L . 1449 22 21 (Antonescu and Burcea 2010), 1.01 ﬂashes km yr in large hail (Goodman et al. 1988; Macgorman et al. 1989; 22 21 Estonia (Enno 2011), and around 0.9 ﬂashes km yr in Williams et al. 1989, 1999; Wiens et al. 2005; Steiger et al. ¨ ¨ Finland (Makela et al. 2011). However, these numbers should 2007; Deierling and Petersen 2008; Gatlin and Goodman be interpreted with caution when making comparisons between 2010; Schultz et al. 2016). Moreover, severe thunderstorms regions as each lightning detection network features different are often associated with lightning jumps, which are good detection efﬁciency and spatial inhomogeneities (Price 2008). predecessors of severe weather (Perez et al. 1997; Schultz et al. In Poland, research on the temporal and spatial variability 2009, 2011, 2017; Farnell et al. 2017, 2018; Farnell and Rigo of thunderstorms was limited for many years to observations 2020). A prominent example might be strong, long-lived from manned surface synoptic stations (Stopa 1962; Grabowska supercell thunderstorms that feature powerful updrafts speeds 2001; Bielec-Ba˛kowska 2002, 2003, 2013; Kolendowicz 2006; and are well known for producing severe weather (Smith et al. Ustrnul and Czekierda 2009). These studies showed that 2012). These thunderstorms may have lightning ﬂash rates thunderstorms in Poland occur most often over southeastern exceeding 200 per minute (Lang et al. 2004; Markowski and Poland (30–35 thunderstorm days per year), and least fre- Richardson 2010). On the other hand, there are reports of tor- quently along the coast of the Baltic Sea (10–15 days). nadoes and severe convective wind gusts associated with However, such data have limitations (associated with human weaker and shallower convection, which rarely produce any perception) in thunderstorm identiﬁcation (Czernecki et al. lightning (Pacey et al. 2021). In central Europe, such events 2016; Koehler 2020; DiGangi et al. 2021), spatial biases, and typically occur in cold seasons and narrow cold-frontal rain- no method to determine storm intensity. The development of bands (NCFR; Gatzen 2011; Surowiecki and Taszarek 2020). the PERUN lightning detection system in the early twenty- For these events, the estimation of thunderstorm intensity ﬁrst century (Łoboda et al. 2009) has allowed for the moni- based on ﬂash rates may be not possible and represents a limi- toring of thunderstorms with a time lag of less than 1 min and tation of using lightning data for identiﬁcation of severe con- enhanced understanding of the spatial and temporal distribu- vective storms. tion of convective storms in Poland. An 11-yr period c. Convective parameters (2002–13) of CG lightning ﬂashes data from the PERUN was analyzed by Taszarek et al. (2015), who conﬁrmed that the Severe convective storms can be predicted and studied with largest number of thunderstorm days is over southeastern convective parameters that, when applied in the NWP mod- Poland (in accordance with estimates from manned surface els, may serve as a proxy of expected thunderstorm intensity. observations), but indicated a peak lightning ﬂash rate over Studies conﬁrmed that thunderstorm severity is driven mainly 22 21 eastern part reaching 3.1 ﬂashes km yr . Taszarek et al. by increasing convective instability and vertical wind shear, and (2015) also concluded that the highest fraction of nocturnal that these metrics can be used to discriminate between severe lightning is over western Poland. and nonsevere convection (Brooks et al. 2003; Thompson et al. 2012; Allen et al. 2011; Pu ´ci ˇ k et al. 2015; Taszarek et al. 2020). b. Thunderstorm intensity However, despite the growing usage of convective indices in The determination of thunderstorm intensity, which can be operational forecasting, they feature certain limitations. Some based on the evaluation of lightning frequency, radar reﬂec- indices, especially composite, merge meteorological elements tivity, or the occurrence of severe weather, is a considerable such as convective instability and shear without physical basis methodological challenge. One of the methods uses severe (Doswell and Schultz 2006). Calculations and especially the weather reports collected from weather observers and media scaling of parameters are in many ways arbitrary and vary sources (Dotzek et al. 2009; Edwards et al. 2013; Elmore et al. depending on the geographical location. Additionally, convec- 2014; Seimon et al. 2016; Krennert et al. 2018). However, the tive parameters demonstrate only possible atmospheric envi- determination of thunderstorm intensity based on these ronments, but do not predict if convection will be initiated. reports is quite arbitrary (Doswell 1985), and depends, inter Due to these factors, one should interpret the values of convec- alia, on the population density and local reporting efﬁciency tive indices with caution (Doswell and Schultz 2006). More- inducing spatial and temporal inhomogeneities (Doswell over, most convective indices were developed in the United 1985; Verbout et al. 2006; Allen and Tippett 2015; Blair et al. States, where convective environments feature higher mois- 2017; Groenemeijer et al. 2017; Edwards et al. 2018; Taszarek ture and convective instability, while European environ- et al. 2019). The ESWD database for the area of Poland con- ments are much drier with steeper lapse rates in the lower sists of more than 22 000 large hail, tornado, and severe wind troposphere (Brooks et al. 2007; Riemann-Campe et al. reports for the period 2008–20. 2009; Taszarek et al. 2020). Additionally, the obtained Teledetection data such as radar, lightning detectors, or sat- results will also depend on the choice of the parcel used for ellites provide more homogenous information and offer an CAPE calculations and application of the virtual temperature attractive alternative for the determination of thunderstorm correction (Doswell and Rasmussen 1994). A better agree- intensity (Punge et al. 2017; Enno et al. 2020; Fluck et al. ment between continents is observed for wind shear. Low- 2021). Well-organized thunderstorms with broad and strong level lapse rates, wind shear, and synoptic-scale lift [which updrafts have larger CG lightning ﬂash rates compared to are included in parameters such as SHERB (severe hazards ordinary cells with weaker updrafts, and large ﬂash rates are in environments with reduced buoyancy) or MOSH (modi- typically associated with the occurrence of tornadoes and ﬁed SHERB)] have overall better skill in discriminating 1450 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 between severe and nonsevere convection in low CAPE and high shear environments that are common in Europe (Hanstrum et al. 2002; Sherburn and Parker 2014; Sherburn ´ ´ et al. 2016; Celinski-Mysław et al. 2020; Rodrıguez and Bech 2021; Pacey et al. 2021). Many factors inﬂuencing thunderstorm development such as vertical wind proﬁle, moisture availability, convective mode, and the regional climatological aspects of atmospheric environments require extensive research aimed at assessing speciﬁc values of convective parameters (e.g., Gensini and Ashley 2011; Allen and Karoly 2014; Anderson-Frey et al. 2016; Chernokulsky et al. 2019; Ingrosso et al. 2020). Such studies demonstrate the pronounced utility of convective indi- ces in forecasting different types of thunderstorms. Thunder- storms in Europe occur most often in CAPE below 500 J kg , and higher convective instability is typically associ- ated with severe weather such as large hail and heavy precipi- tation. Given an unstable environment, wind shear is a good discriminator between nonsevere and extremely severe storms (∼20 m s can be used as a proxy for very large hail FIG. 2. Hypsometric map of Poland based on the Digital Terrain ´ ˇ and signiﬁcant tornadoes; Pucik et al. 2015; Taszarek et al. Model (USGS/SRTM; Farr et al. 2007). Points indicate the location 2020). However, one should be aware that it is difﬁcult to of the PERUN lightning detection sensors (red points indicate sen- compare results between studies using different reanalysis sors added to the PERUN network after 2014). datasets, as each of them represents a different underlying cli- matology of convective environments, and the values of cer- tain parameters may change along with changing resolution. coverage, with the lowest detection accuracy over northwest- Although many of the previously mentioned studies ern Poland (Taszarek et al. 2015). As new sensors have been focused on comparing severe wind, large hail, heavy rainfall, added to the network, the detection accuracy has increased and tornadoes with accompanying ambient environments, since 2014, especially in the southwestern part of the country. only few evaluated different classes of CG lightning ﬂash Each CG lightning ﬂash detection in the database included rates, which is the main focus of this work. Although CG the time, location, and a peak current of the discharge. To lightning ﬂash rates are not typically predicted by operational provide better homogeneity of the data we took into account forecasters, a correlation between speciﬁc CG lightning ﬂash only situations when a ﬂash had a peak current of at least rates and the relative number of ESWD reports can be 15 kA, as lower values are typically linked to intracloud detec- observed (Fig. 1). tions (Wacker and Orville 1999; Cummins and Murphy 2009; Koehler 2020). After this ﬁlter, a total of 8 251 273 CG light- 3. Data and methods ning ﬂashes (limited to the area of Poland) were used in this study. a. Lightning data The total number of CG lightning ﬂashes within a speciﬁc In this study we used all the available CG lightning ﬂashes thunderstorm depends not only on the evolution of updraft detected by the PERUN system on the territory of Poland intensity, but also on the duration and spatial coverage of the between 2002 and 2019 (18 years). PERUN is a Polish light- entire convective system. Thus, the total number of CG light- ning detection network working operationally since 2002. Its ning ﬂashes over a broader area (e.g., associated with meso- name corresponds to the Slavic god of lightning and thunder scale convective systems; Houze 2018) may not represent (Gieysztor 2006). At the beginning, the system consisted of 9 thunderstorm intensity well in comparison to local severe con- sensors, which after 2014 were expanded to 12 (Fig. 2). At the vective cells with a strong updraft (e.g., supercells), which can same time, the sensors began upgrading from SAFIR3000 produce a comparable number of CG lightning ﬂashes but (Surveillance et Alerte Foudre par Interferometrie Radio- over a much smaller area (Wiens et al. 2005; Steiger et al. electrique) to TLS200 (Total Lightning Sensor). As of 2021, 2007; Calhoun et al. 2013). Thus, to avoid biases associated the PERUN system consists of 8 TLS200 and 4 SAFIR3000 with increasing the thunderstorm area but not intensity, we sensors and is supported by the data from sensors in the decided to use a peak of 1-min ﬂash rate per 100 km . neighboring countries. CG lightning ﬂash detection efﬁciency To obtain information about the CG lightning ﬂash rates, estimated in 2006 reached 95% within the area of Poland all detections were gridded to 10 km 3 10 km boxes}a metric (Bodzak 2006; Bodzak et al. 2006). Since the sensors are not commonly used in prior work (e.g., Biron 2009; Enno 2011; distributed evenly throughout the country, only 38.3% of the Sulik 2021). Then, within these boxes, we investigated the country’s coverage has lightning location accuracy ﬁner than peak 1-min ﬂash count. In this way, a rate of CG lightning 1 km. Accuracy ﬁner than 2 km is found for 76.6% of the ﬂashes per 100 km per minute was obtained, and then AUGUST 2022 PO R E ˛ BA ET A L . 1451 TABLE 1. Thunderstorm intensity classes determined on the TABLE 3. Number of unique ERA5 grids in a given basis of 1-min peak CG lightning ﬂashes per 10 km 3 10 km thunderstorm CG lightning ﬂash intensity category over each grid (considered in hour steps) with corresponding percentile month. values, number of collocated ERA5 grids and a mean annual Weak Moderate Intense Extreme number of days given a speciﬁc category. January 493 120 17 3 Weak Moderate Intense Extreme February 351 64 3 4 Peak 1-min CG 12–45–9 .9 March 1669 763 116 25 lightning ﬂash April 11 278 5266 877 266 rate per 100 km May 45 810 32 126 7853 2714 Percentile of entire 49% 50%–84% 85%–94% 95%–100% June 52 946 40 156 12 402 7344 dataset July 73 392 52 917 16 589 8842 Number of ERA5 240 095 173 781 50 646 26 091 August 47 878 33 430 10 672 6066 grids September 12 911 7674 1923 751 Annual mean 161 126 124 55 October 2031 937 162 64 number of days November 635 190 24 7 December 283 138 8 5 subsequently used to determine thunderstorm lightning inten- sity (Table 1). Intensity classes of weak, moderate, intense, For each ERA5 grid, we assigned a thunderstorm category and extreme thunderstorms were deﬁned based on the 50th, based on the highest peak 1-min CG lightning ﬂash rate 75th, 90th, and 95th percentile thresholds of CG lightning within a speciﬁc hour (to match the ERA5 temporal step). ﬂash rate distribution. A limitation of this approach is that That procedure allowed us to merge two datasets with differ- ent resolutions. Weak thunderstorms contributed to the larg- due to the gridding process the thunderstorm intensity data est number of ERA5 grids, while those of extreme intensity refer to the peak ﬂash rates, which is not always associated were the least frequent, representing an annual mean of with one thunderstorm (several thunderstorms could have 161 and 55 days, respectively (Table 1). With the increase in occurred in the temporal and spatial scope of the reanalysis). the thunderstorm intensity category, the number of grids Nocturnal thunderstorms were classiﬁed based on the crite- decreases, especially in winter}in the case of extreme thun- rion of the solar angle below 2128 (computed for the location derstorms in January, only three grids were used (Table 3). and time of CG lightning ﬂash occurrence). This metric is For each reanalysis proﬁle, temperature, humidity, pres- known as the nautical dawn (NOAA/NWS Glossary 2021). sure, geopotential, U and V were interpolated vertically. b. ERA5 reanalysis A mixed-layer (ML) parcel was deﬁned by mixing a layer 0–500 m above ground level (AGL), while a most unstable To investigate thunderstorm environments, we used the ﬁfth (MU) parcel was based on the highest equivalent potential generation of ECMWF atmospheric reanalysis (ERA5; temperature (Q ) in the 0–3 km AGL; both versions used the Hersbach et al. 2020). ERA5 has a horizontal grid spacing of virtual temperature correction (Doswell and Rasmussen 0.258, a 1-h temporal step, and 137 terrain-following hybrid- 1994). For the computations of 0–3 km AGL storm-relative sigma model levels (Table 2), which made it possible to explore helicity (SRH03), the internal dynamics method was applied convective environments and their corresponding climatologies for right-moving supercells (Bunkers et al. 2000). Vertical in a way that was not possible with prior reanalyses, especially wind shear was computed as a magnitude of vector difference considering diurnal cycles. The 1-h time step allowed us to assign between the surface and a speciﬁc height. The indices used in lighting ﬂash events to speciﬁc convective environments with the study are associated with the concept of the ingredient- higher precision compared to previous reanalyses. In this case, based forecasting (Johns and Doswell 1992; Doswell et al. the maximum time difference between the CG lightning ﬂash 1996) focusing on the assessment of humidity characteristics, and the corresponding ERA5 grid was less than 30 min instead convective instability, and a vertical proﬁle of the wind, of 3 h for reanalyses consisting of 6-h steps (e.g., ERA-Interim). which governs the convective mode and storm organization For each grid from the previously described CG lightning (Weisman and Klemp 1982; Thompson et al. 2012). In addi- ﬂash rate database, we assigned a proximal grid from ERA5. tion, we also used composite indices, which turned out to be useful in assessing thunderstorm intensity across central TABLE 2. Characteristics of ERA5 reanalysis domain used in Europe: WMAXSHEAR (a square root of 2 times CAPE this study. multiplied by 0–6-km wind shear; Taszarek et al. 2020)dedi- cated to forecasting severe storms, SCP (supercell composite Horizontal grid spacing 0.258 3 0.258 Vertical levels (sigma) 137 hybrid-sigma levels parameter; Thompson et al. 2003) with an updated formula Temporal resolution Hourly from Gropp and Davenport (2018), and HSI (hail size index; Timeframe 2002–19 Czernecki et al. 2019) dedicated for forecasting large hail. As Total unique grid points 500 195 supercells and large hail are typically associated with strong Latitude extent 48.758–55.258N updrafts that also favor intense lightning, we believe these Longitude extent 13.758–24.508E metrics are worth comparing with other convective parameters 1452 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 and among CG lightning ﬂash ratecategories. A completelist TABLE 4. List of convective parameters used in this study (see the appendix for additional metrics). of variables used in this study is presented in Table 4 with additional metrics provided in the appendix. Parameter Abbreviation Unit c. ESWD reports 21 Most unstable (0–3 km AGL) MUCAPE J kg convective available potential A total of 14 427 ESWD reports from the area of Poland energy were used for comparison with lightning intensity categories Most unstable (0–3 km AGL) MUCIN J kg (Fig. 1). Only ESWD reports with Q1 and Q2 quality level clas- convective inhibition ses (meaning reports from reliable sources or conﬁrmed by sci- Most unstable (0–3 km AGL) MUMIXR g kg entiﬁcstudy; Dotzek et al. 2009), time lag of less than 1 h, and mixing ratio spatial proximity of 0.258 (from the detected speciﬁcCG light- 800–500-hPa temperature lapse LR85 K km rate ning density) were considered. For the higher lightning intensity 0–1 km AGL bulk wind shear LLS m s categories, the number of available cases decreases. We calcu- 0–6 km AGL bulk wind shear DLS m s late a mean number of ESWD reports that are within the afore- 2 22 0–3 km AGL storm-relative SRH03 m s mentioned proximity of the grid with speciﬁc CG lightning ﬂash helicity (right-moving rate. In the end, we divide the number of ESWD reports supercells) assigned to a speciﬁc lightning intensity category by the number 2 22 Product of MUCAPE and DLS MUWMAXSHEAR m s of grids for this speciﬁc category. (Taszarek et al. 2020) Supercell composite parameter SCP } (Thompson et al. 2003; Gropp 4. Results and Davenport 2018) a. Climatological aspects of CG lightning flashes Hail size index (Czernecki et al. HSI } in Poland 2019) Over the period 2002–19, we identiﬁed a total number of 2926 days with at least two CG lightning ﬂashes, which repre- probability for CG lightning ﬂashes around 1400–1500 UTC, and sented a mean of 161 thunderstorm days per year. The highest the lowest around 0700–0800 UTC. This pattern is evidently visi- annual mean number of hours with CG lightning ﬂashes ble during the summer and spring, whereas from October to occurred over southeastern and central Poland, reaching March it is less clear (Figs. 4b and 5a). Thespringfeatures the around 80 h (Fig. 3a), consistent with results from Earth highest fraction (12%) of CG lightning ﬂashes occurring in the Networks Global Lightning Detection Network (ENGLN; afternoon (1200–1500 UTC), while during the summer the peak DiGangi et al. 2021). The regions with the most frequent CG fraction occurs in the same hours, but with lower values of lightning ﬂashes included mainly higher elevation areas such around 8%–9%. At night, CG lightning ﬂashes are considerably as the Carpathian Mountains, the Lublin Upland, and the less frequent, with the lowest fractions in spring. During winter Krakow–Cze Rstochowa Upland, as well as the lowland areas and autumn, the fraction of nocturnal CG lightning ﬂashes is across central Poland (geographical regions of Poland are slightly higher than during summer and spring. provided in Fig. 2). Conversely, the lowest frequency of thun- Considering the interannual variability (Fig. 4c), the highest derstorm hours occurred across northwestern Poland, with an overall number of CG lightning ﬂashes occur every year dur- annual mean of less than 20 h. The highest mean annual CG ing the summer (in 2017 almost 1 000 000) and the lowest in 22 21 lightning ﬂash rate reaching 3.0 CG km yr was recorded the winter (in 2008 only 10). Although the spring typically has over the central part of the country and in the region of the a much higher number of CG lightning ﬂashes compared to Krakow–Cze Rstochowa Upland (Fig. 3b). the autumn, in some years, the number of detections can be The annual cycle of CG lightning ﬂashes in Poland is con- comparable (e.g., in 2006, the autumn had a higher number of sistent with typical climatological distributions of temperature detections compared to spring). Slight increases in the num- and precipitation in this part of Europe (Lorenc 2005). Thun- ber of annual CG lightning ﬂashes are probably due to derstorms are the most frequent in mid-July, but the time- improvements in the PERUN system over time. Strong frame with an enhanced mean number of CG lightning ﬂashes year-to-year variability is also observed for the annual frac- starts in mid-April with more than 7000, and ends in late tion of nocturnal CG lightning ﬂashes with values from 7% in September with 14 700 (Figs. 4a and 5a). During that period, 2006 to as much as 22% in 2009 (Fig. 4d). However, no signiﬁ- the annual mean number of CG lightning ﬂashes is greatest, cant long-term trend can be deﬁned. especially in July, reaching 155 474 (Fig. 4a). June has a The highest daily numberofCG lightning ﬂashes was slightly higher number compared to August with around recorded on 10 August 2017, reaching 141 628 (Table 5), which 15 000 more ﬂashes. Thunderstorms are clearly less frequent is 57 981 more detections than the second ranked 28 June during the autumn, winter, and early spring. From October to 2017. On 10 August 2017, the coverage of thunderstorms was March, the mean number of CG lightning ﬂashes typically also the largest of all analyzed cases and reached 55 791 km does not exceed 1300 (Fig. 4a). Considering the fractional dis- (17.8% of the area of Poland). Considering other days, the tribution of CG lightning ﬂashes in each hour, convection in area covered by grids with CG lightning ﬂashes ranged typi- Poland has a well-deﬁned diurnal cycle with the highest cally from 10.7% to 12.3% of the area of Poland (Table 5). AUGUST 2022 PO R E ˛ BA ET A L . 1453 FIG.3. (a) Mean annual number of hours with at least one CG lightning ﬂash, and (b) mean annual number of CG lightning ﬂashes per 1 km . Calculations are based on data from the PERUN lightning detection system for 2002–19 (10 km 3 10 km grid). Spatial distribution of median of (c) MUCAPE, (d) MUCIN, (e) DLS, and (f) 90th percentile of MUWMAXSHEAR. Convective parameters are derived from ERA5 reanalysis for a period 2002–19 (0.258 3 0.258 grid). Please note that only situations with detected CG lightning ﬂashes are considered. 1454 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 FIG. 4. (a) Annual mean number of CG lightning ﬂashes per month, (b) hourly fraction of CG lightning ﬂashes per meteorological season, (c) year-to-year variability of CG lightning ﬂashes per meteorological season, and (d) fraction of nocturnal CG lightning ﬂashes over the years. Calculations were based on the data from the PERUN lightning detection system for the years 2002–19. Please note the logarithmic scale in (c). Nocturnal lightning in (d) is deter- mined based on the solar angle (below 2128). Among the 10 days with the highest daily number of CG light- The components of atmospheric convective instability: ning ﬂashes, four cases occurred both in June and July, with MUMIXR (mixing ratio of the most unstable parcel) and two in August. The most intense thunderstorms in terms of LR85 (800–500-hPa temperature lapse rate) are characterized 1-min peak CG lightning ﬂash rate occurred on 21 June 2013, by differing patterns in temporal distributions. Peak values of 18 June 2013, and 10 August 2017, reaching 108, 80, and MUMIXR reach 12 g kg in July and early August, typically 2 21 74 CG lightning ﬂashes per 100 km per minute, respectively around 1800 UTC (Fig. 5b), while LR85 of 7 K km is (Table 6). observed between 1000 and 1600 UTC during spring from March to May (Fig. 5c). MUMIXR has a better relationship b. Annual and diurnal variability of CG lightning flash with the diurnal cycle of CG lightning ﬂashes but is delayed environments by about 3 h. The best overlap of MUMIXR and LR85 is in In this section we present the climatological temporal and late July and early August between 1200 and 1700 UTC, with spatial distribution of ERA5 convective environments for the resulting median most unstable convective available thunderstorms in Poland. The statistics presented in this sec- potential energy (MUCAPE) reaching around 850 J kg tion refer only to situations when a CG lightning ﬂash was (Fig. 5d). Conversely, when no insolation is available during detected, leading to a smaller sample size during winter and the night, MUCAPE has its minimum. Only for the period larger in summer (Table 3). Thus, results should be inter- from early July to late August does nocturnal MUCAPE have preted with caution as nonelectriﬁed convection is not a median of around 400 J kg . During the spring and considered. autumn, peak values of MUCAPE occur around noon, driven AUGUST 2022 PO R E ˛ BA ET A L . 1455 FIG. 5. (a) Annual and daily variability of grids with at least one CG lightning ﬂash, (b) median of MUMIXR, (c) LR85, (d) MUCAPE, (e) DLS, and (f) 90th percentile of MUWMAXSHEAR. Convective parameters are derived from ERA5 reanalysis for a period 2002–19. Vertical axis has hourly resolution while horizontal weekly. Labels indi- cate the ﬁrst week of each month. Please note that only situations with detected CG lightning ﬂashes are considered. by the surface heating and development of steep lapse rates. In comparison to MUCAPE, the distribution of deep-layer Climatological patterns in MUCAPE are well correlated with shear (DLS) features a reversed pattern (Fig. 5e). From April peak CG lightning ﬂash activity and have a well-deﬁned diur- to October, the median of DLS generally does not exceed nal cycle during the summer in opposition to a poor diurnal 10 m s , with the lowest values in the afternoon hours cycle during winter when thunderstorms in Poland are typi- (1200–1600 UTC). However, shear notably increases during cally rare (Figs. 5a,d). nighttime hours by 5–7m s , which is particularly visible in 1456 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 TABLE 5. Top 10 days with the highest number of CG lightning ﬂashes (left), and the highest spatial coverage of 10 km 3 10 km grids with at least one detection (right). Top 10 days with the highest number of Top 10 days with the highest spatial coverage of grids CG lightning flashes with at least one CG lightning flash Area with flashes (km ) Date No. of flashes Date (percent of country coverage) 10 Aug 2017 141 628 10 Aug 2017 55 791 (17.8%) 28 Jun 2017 83 647 25 Jul 2015 38 596 (12.3%) 21 Jun 2013 78 116 03 Jul 2012 38 014 (12.2%) 9 Aug 2013 74 625 15 Aug 2008 36 668 (11.7%) 26 Jun 2006 74 153 26 Jun 2006 36 482 (11.7%) 25 Jul 2015 72 361 19 Jul 2015 35 977 (11.5%) 13 Jun 2019 66 454 13 Jun 2019 34 520 (11.0%) 3 Jul 2012 64 938 28 Jun 2017 34 239 (11.0%) 15 Aug 2008 64 233 31 Jul 2005 33 434 (10.7%) 19 Jul 2015 63 967 29 Jun 2017 33 339 (10.7%) the summer months. In the period with lesser convective Taszarek 2020). Noteworthy also are the increased values of 2 22 occurrence from October to April, DLS has a typically less MUWMAXSHEAR (around 500 m s of 90th percentile) deﬁned diurnal cycle, but much higher overall values during spring and autumn, which are mainly driven by large (a median of 20–30 m s ). DLS and rather marginal MUCAPE. These conditions are so- A combination of thermodynamic convective instability and called high-shear low-CAPE (HSLC) environments, and are vertical wind shear is a useful environmental proxy for assessing often associated with an enhanced potential of squall lines and ´ ˇ thunderstorm severity (Brooks et al. 2003; Pucik et al. 2015; low-topped supercells capable of producing damaging winds Taszarek et al. 2020). MUWMAXSHEAR, which combines- and tornadoes (Sherburn and Parker 2014; Sherburn et al. 2016; both MUCAPE and DLS, indicates peak severe thunderstorm Anderson-Frey et al. 2019; Mathias et al. 2019; Gatzen et al. potential from June to August, with the 90th percentile exceed- 2020). 2 22 ing 800 m s during this period (Fig. 5f). The highest values c. Spatial variability of CG lightning flash environments typically occur between 1600 and 2000 UTC in late July and early August. Although DLS is relatively low in July (median Spatially, median MUCAPE (only for CG lightning ﬂash 10–14 m s ), the highMUCAPE inthat periodisthe main events) increases from the northwest toward the southeast driver of the overall MUWMAXSHEAR value. A similar (Fig. 3c), and thus the lowest values are along the Baltic Sea effect occurs in the diurnal cycle. Although MUCAPE drops coast (450 J kg ) and the highest in the Bieszczady Moun- during nighttime, MUWMAXSHEAR is still high, as it is com- tains over the southeast (700 J kg ). This pattern is consis- pensated by increasing DLS. This reﬂects an enhanced noctur- tent with the frequency of thunderstorm days based on nal potential for severe thunderstorms, that as a result of SYNOP reports (Bielec-Ba ˛kowska 2003; Czernecki et al. increasing shear, often evolve into large and well-organized 2016). mesoscale convective systems (MCS) with the potential to pro- The median of the MUCIN (convective inhibition) has a duce severe wind and heavy rain (Geerts et al. 2017; Reif and spatial pattern reversed to MUCAPE with the strongest Bluestein 2017; Haberlie and Ashley 2019; Surowiecki and inhibition over the western and northwestern parts of the country reaching 230 J kg and weakest over the south (210 J kg ), indicating the area with the lowest resistance to TABLE 6. Time and location of top 10 grid boxes (10 km 3 10 km) convection initiation (Fig. 3d). We hypothesize that the with the highest CG lightning ﬂash rates. enhanced inhibition occurring along the coast of the Baltic Sea is associated with low sea surface temperatures that lead Peak 1-min CG to temperature inversions in typically marginal MUCAPE sit- lightning flash rate 8 8 Date and time Lon ( E) Lat ( N) per 100 km uations. The regions with the strongest inhibition over west- ern Poland experience the lowest average annual rainfall in 1627 UTC 21 Jun 2013 21.000 50.732 108 Poland (Lorenc 2005), which also seems to be consistent with 1312 UTC 21 Jun 2013 20.409 52.619 88 the reduced convective frequency over that area (Fig. 3a). 1545 UTC 18 Jun 2013 21.708 50.730 80 0349 UTC 10 Aug 2017 18.083 52.044 74 Similarly to MUCIN, the spatial patterns in DLS are also 2118 UTC 12 Aug 2019 21.427 50.911 72 reversed with respect to MUCAPE, with the lowest values 2144 UTC 12 Aug 2019 21.855 50.998 70 over southeastern Poland (∼10 m s ) and the highest over 0545 UTC 10 Aug 2017 19.352 53.238 70 the northwest (∼14 m s ; Fig. 3e). This pattern indicates 1628 UTC 21 Jun 2013 21.000 50.732 70 more frequent HSLC environments over northwestern 1601 UTC 18 Jun 2013 21.849 50.639 70 Poland and in turn more common low-shear, high-CAPE 0219 UTC 14 Jul 2016 22.693 50.540 67 (LSHC) situations over the southeast. However, noteworthy AUGUST 2022 PO R E ˛ BA ET A L . 1457 FIG. 6. Number of situations with speciﬁc MUCAPE and DLS values for weak, moderate, intense, and extreme thunderstorms (divided based on a peak 1-min CG lightning ﬂash rate, as in Table 1). Constant values of MUWMAX- SHEAR (a product of MUCAPE and DLS; Taszarek et al. 2020) are indicated by dashed lines. White points indicate the median value for each category. Convective parameters are derived from ERA5 reanalysis for a period 2002–19. Please note that only situations with detected CG lightning ﬂashes are considered. is the northeastern part of the country, where a combination slightly differing thermodynamic convective instability (a dif- 21 21 of climatologically enhanced MUCAPE (∼600 J kg ) and ference of 129 J kg in the medians of MUCAPE). One can DLS (∼13 m s ) indicates increased conditional potential also see a dependence that thunderstorms developing in an for severe thunderstorms in this region, despite their limited environment with increased MUCAPE require lower DLS frequency compared to the southeast (Fig. 3a). This is con- and conversely, a higher DLS is typically accompanied by ﬁrmed by the spatial distribution of the 90th percentile of weaker convective instability. This pattern is well depicted by MUWMAXSHEAR, which in this region reaches even the constant values of MUWMAXSHEAR on the scatter- 2 22 850 m s } the highest values among the entire country (Fig. 3f). plots (dashed lines on Fig. 6). Thunderstorms with both high CAPE and high DLS are very rare in Poland. For intense and d. Variability of convective environments in relation to extreme thunderstorms (Figs. 6c,d), a shift in the number of CG lightning flash rates cases toward higher MUCAPE and a gradual increase in The two-dimensional distributions of MUCAPE and DLS DLS can be recognized. In the extreme thunderstorm for different CG lightning ﬂash rate intensity categories category (Fig. 6d), a median of 834 J kg MUCAPE and (weak, moderate, intense, extreme; Table 1, Fig. 6) indicate 12.7 m s DLS is notably higher compared to other classes. that the conditional probability for larger CG lightning ﬂash However, the largest increase of 153 J kg in the medians of rates increases as convective instability and vertical wind MUCAPE is between the moderate and intense thunder- shear increase. The median for MUWMAXSHEAR increases storm categories. 2 22 for each category with 330, 375, 441, and 519 m s , respec- In summary, although the probability for higher CG light- tively (Fig. 6). This result is broadly consistent with Liu et al. ning ﬂash rates increases with both DLS and MUCAPE, the (2020). increase in the latter is generally more important. A median 2 22 Both weak (Fig. 6a) and moderate (Fig. 6b) thunderstorm MUWMAXSHEAR of 441 m s for the intense thunder- 2 22 categories feature similar kinematic environments but a storm category is similar to the value of 450 m s from 1458 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 FIG. 7. Heidke skill score for MUCAPE, DLS, and MUWMAXSHEAR calculated by comparing the weak thunder- storm intensity category with at least moderate (orange), at least intense (red), and extreme (magenta). Taszarek et al. (2019), which was considered as a discrimina- of HSS indicate that predicting CG lightning ﬂash rate tor between severe and nonsevere thunderstorms across cen- with convective parameters is a very challenging task, and tral Europe (conditional on successful convective initiation). its application in operational forecasting is limited. How- On the other hand, a different pattern of DLS and CAPE ever, forecasting any type of convective hazards across dependency was found in Europe for the relative frequency of Europe is a difﬁcult task (Brooks et al. 2011). lightning (Westermayer et al. 2017). High relative frequency The increase of MUMIXR and MUCAPE corresponds 21 21 was found in low (,10 m s )and high (.20 m s )DLS with increasing conditional probability for a higher CG light- environments, while CAPE values below approximately ning ﬂash rate category. The annual and daily distributions of 200–400 J kg demonstrated low relative frequency. This sug- these indices are similar (Figs. 8a,c), reaching the highest val- 21 21 21 gests that both enhanced CAPE and DLS are necessary for ues (medians from 10 g kg and 500 J kg to 12 g kg and storms producing high CG lightning ﬂash rates, but only a 1000 J kg for weak to extreme thunderstorms) in summer small amount of CAPE is needed for occurrence of any light- (June–August), except for the 90th percentile of MUCAPE ning (given successful convective initiation). However, as for extreme thunderstorm category, which is notably highest thunderstorms with high CG lightning ﬂash rates may not (∼2700 J kg ) in June (compared to July for MUMIXR). always be associated with severe weather and vice versa, cau- Diurnally, for both indices, the highest values are in the tion needs to be taken when interpreting this result. afternoon (Figs. 8b,d). MUMIXR median oscillates around To provide additional statistical tests for MUCAPE, DLS 10–13 g kg during both day and night (Fig. 8b). The highest and MUWMAXSHEAR, we calculated the Heidke skill median in MUCAPE for weak and moderate thunderstorms score (HSS). Peak HSS values were found for MUCAPE of occurs between 1000 and 1400 UTC, for strong between 1200 21 21 1000 J kg (HSS of 0.13), DLS of 12.5 m s (HSS of 0.05), and 1400 UTC, and for extreme between 1400 and 1800 UTC. 2 22 and MUWMAXSHEAR of 600 m s (HSS of 0.16) for This pattern indicates that storms developing in a high extreme versus weak CG lightning ﬂash rate category MUCAPE environment during late afternoon are more likely (Fig. 7). A shift of peak HSS values among speciﬁcCG to become severe, which is possibly linked to increasing verti- lightning ﬂash rate categories was observed with increas- cal wind shear toward late evening and night. ing values of MUCAPE and MUWMAXSHEAR, while Analysis of MUCIN demonstrated that environments for for DLS, differences were marginal. The biggest differ- severe and extreme thunderstorms are associated not only with ences among categories and the highest HSS were higher CAPE and shear but also with stronger inhibition observed for MUWMAXSHEAR, which seems to better (Figs. 8e,f). The strongest inhibition (the most negative values discriminate among CG lightning ﬂash rate categories of MUCIN) occurs in the summer, but in contrast to compared to MUCAPE or DLS alone. Overall low values MUCAPE, value distributions are relatively even from June to AUGUST 2022 PO R E ˛ BA ET A L . 1459 FIG. 8. Box-and-whisker plots presenting the (left) annual and (right) diurnal distributions of thunderstorm intensity classes (weak, moderate, intense, extreme; as in Table 1) for (a),(b) MUMIXR, (c),(d) MUCAPE, and (e),(f) MUCIN. Convective parameters are derived from ERA5 reanalysis for a period 2002–19. Please note that only situa- tions with detected CG lightning ﬂashes are considered. The horizontal line inside the box demonstrates the median, the extent of the box is from the 25th to 75th percentile, and whiskers represent the 10th and 90th percentiles. 21 21 21 September (median for all categories from 25to 215 J kg ). 250 J kg in 5 J kg steps (Table 7). The results indicate that Diurnally, the weakest inhibition (highest values) occurs during for each thunderstorm category, probability for the occurrence strong surface heating between 1000 and 1600 UTC, while is highest for MUCIN values above 210 J kg (weaker inhibi- outside this period a median of MUCIN varies from 25to tion) and lowest below 240 J kg . While in higher MUCIN 220 J kg (Fig. 8f). Convective inhibition is typically stronger (weaker inhibition), conditional probability increases along in environments favorable for higher CG lightning ﬂash rates. with decreasing thunderstorm intensity, an opposite relation- This is conﬁrmed by calculating the conditional probability for ship can be observed for lower MUCIN (stronger inhibition). each thunderstorm category and MUCIN intervals from 0 to Delayed convection initiation during the day driven by 1460 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 TABLE 7. Conditional probability distribution for MUCIN intervals and thunderstorm category. MUCIN intervals (J kg ) [0; 210] (210; 220] (220; 230] (230; 240] (240; 250] Weak 0.6734 0.1637 0.0811 0.0493 0.0325 Moderate 0.6386 0.1760 0.0924 0.0551 0.0378 Intense 0.5887 0.1922 0.1096 0.0652 0.0442 Extreme 0.5441 0.2080 0.1181 0.0773 0.0525 stronger inhibition subsequently promotes stronger moisture lowest frequency of thunderstorms and lowest convective pooling, higher convective instability, better storm organization instability. It is also worth highlighting that, contrary to pre- (shear is higher toward evening hours; Fig. 9b), and a more viously analyzed variables (LR85, MUCAPE, MUCIN, sudden development of convective updrafts promoting higher and LLS), MUWMAXSHEAR has no clearly deﬁned CG lightning ﬂash rates. Moreover, environments favorable for diurnal cycle. This is because of the distribution of its extreme thunderstorms are often associated with advection of components}convective instability, which has a clear diur- very warm air with steep lapse rates and an elevated mixed nal cycle with the maximum in the afternoon, and shear, layer, a situation commonly referred to as a “loaded gun” pro- which has an inverted diurnal cycle with minimum values in ﬁle. These environments in Europe are often associated with the afternoon. The combination of these two results in a “Spanish plume” synoptic setups (Carlson and Ludlam 1968; ﬂattened diurnal distribution. VanDelden2001; Mathias et al. 2017). On the other hand, While we are aware that SCP and HSI are parameters stronger convective inhibition during extreme thunderstorms is designed to forecast supercells and large hail, the strength of also pronounced at night (medians around 220 J kg ), which the updraft in the supercells and conditions favorable for may correspond to well organized, elevated thunderstorms large hail formation are also supportive for lightning genera- (Colman 1990; Grant 1995). These thunderstorms initiate tion. Thus, we evaluate these parameters in our study as well above a stable layer without being inﬂuenced by strong mixed and note that their increasing values are consistent with more layer or surface-based convective inhibition. frequent lightning. The highest values of both SCP and HSI As previously discussed, months with the highest thermody- are observed between May and August. While values of SCP namic convective instability feature the lowest values of kine- during other months rarely exceed 0, the HSI timeframe with matic parameters (Fig. 9). DLS, 0–1-km wind shear [low-level elevated values is from April to October. A rapid increase in wind shear (LLS)], and SRH03 from April to September has a the HSI in April is maintained at a similar level until August 21 21 2 22 similar range of values (∼11 m s , ∼6m s ,and ∼70 m s , with the 90th percentile of every storm category exceeding a respectively) among thunderstorm intensity classes (Figs. 9a,c,e). value of 2 (Fig. 10e), which highlights an elevated risk of large Only for the extreme thunderstorms, this relationship is better hail in this period. For the SCP, 90th percentile for intense pronounced with overall slightly higher values of all three and extreme thunderstorms during the summer exceeds a metrics. The diurnal distributions of DLS, LLS, and SRH03 value of 2 (Fig. 10c), which is a good discriminator between (Figs. 9b,d,f) are characterized by the lowest values in the hours supercell and nonsupercell thunderstorms based on a calibra- with the highest thermodynamic convective instability (from tion derived from environments across the United States 1200 to 1600 UTC), which is especially evident for LLS that has (Thompson et al. 2003, 2007; Gropp and Davenport 2018). a stronger diurnal amplitude compared to DLS and SRH03. At This may lead to the conclusion that conditions supportive for night and in the morning, the extreme thunderstorm category supercell development in Poland are relatively uncommon features higher median values compared to other categories, and occur typically between May and August. Diurnally, SCP 21 21 2 22 reaching 17 m s ,7 m s ,and 140 m s for DLS, LLS, and has no clear peak, but its minima (with the 90th percentile SHR03, respectively. Low differences between thunderstorm below 0.7) can be explicitly depicted between 0000 and 0200 categories, demonstrates limited utility in forecasting speciﬁc UTC (Fig. 10d). In contrast, HSI has a clear diurnal cycle ﬂash rates using only kinematic parameters. (Fig. 10f) with its peak of ∼1.2 (medians) between 1200 and Composite indices combining multiple parameters into one 1400 UTC (hours with the highest lapse rates). For both product aimed at forecasting speciﬁc phenomena offers an parameters, higher values are characteristic in more intense attractive approach for forecasters, and thus we also evalu- thunderstorm categories, but for SCP, the difference between ated them in our study. MUWMAXSHEAR reaches the high- categories is even higher. The distribution of the 90th percen- 2 22 est median during summer months (300–600 m s ) with a 90th tile for HSI demonstrates rather similar values (3.0–3.4 during 2 22 percentile of extreme thunderstorms exceeding 1250 m s summer) for each thunderstorm category, while for SCP, (Fig. 10a). A notion of increasing MUWMAXSHEAR with extreme thunderstorms have clearly higher values compared increasing CG lightning ﬂash rates is valid from April to Octo- to other categories. Based on that, we can conclude that ber, while from November to March, such a dependence is not supercells in Poland can occur at any time of the aforemen- present. Diurnally (Fig. 10b), the biggest difference between the tioned period and that CG lightning ﬂash rates increase with weak and extreme category is between 0400 and 0800 UTC increasing chances for supercells (i.e., values of SCP). This (morning hours of local time in Poland)}the period with the result is unsurprising, as powerful updrafts in supercell AUGUST 2022 PO R E ˛ BA ET A L . 1461 FIG.9.As in Fig. 7, but for (a),(b) DLS, (c),(d) LLS, and (e),(f) SRH03. thunderstorms may reach vertical velocities of even 50 m s of this study. In addition to the parameters evaluated in this (Lehmiller et al. 2001) and drive generation of very large CG study, in the appendix, we provide median, 10th, and 90th per- lightning ﬂash rates. centiles for speciﬁc CG lightning ﬂash rate categories. In the distributions presented, considerable overlap occurred between thunderstorm categories. For convective instability e. Importance of relative humidity on and composite parameters, this overlap was less pronounced, convection initiation suggesting that CG lightning ﬂash rates are more sensitive to The last analyzed meteorological aspect is the relative convective instability parameters than shear. Improving the analysis by the relation of convective parameters with the radar humidity at the isobaric levels of 850, 700, and 500 hPa, and a data would probably reduce this overlap, but is out of the scope mean in the 0–4 km AGL layer (Fig. 11). Relative humidity, 1462 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 FIG.10. As in Fig. 7, but for (a),(b) MUWMAXSHEAR, (c),(d) SCP, and (e),(f) HSI. although it does not inﬂuence thunderstorm intensity, is a Among the previously mentioned pressure levels, the high- crucial factor for the development of any thunderstorms est values of the interquartile range and the median of relative (Westermayer et al. 2017). The lower the relative humidity is, humidity occurred at 850 hPa (approximately 1.5 km MSL; the stronger the dry air entrainment can be, which may Fig. 11b). At this level, 3/4 of all analyzed lightning cases had strongly limit the development of convective updrafts. On the a relative humidity between 70% and 85%. Equally high val- other hand, relatively dry mid tropospheric air but in environ- ues of the interquartile range occurred also in a mean 0–4-km ments with strong synoptic-scale forcing and sufﬁcient humid- relative humidity (Fig. 11a), stretching from 66% to 80%, ity at low levels, can result in releasing potential instability while at the level of 700 hPa (Fig. 11c) this spread varied from and successful convection development. This effect is demon- 65% to 83%. The relative humidity at the level of 500 hPa strated on a rather ﬂat distribution of relative humidity at 500 turned out to be less important as thunderstorms occurred hPa compared to lower levels. in a wide spectrum of values and the median was 56% AUGUST 2022 PO R E ˛ BA ET A L . 1463 FIG. 11. Histograms of relative humidity at 850-, 700-, and 500-hPa isobaric levels and a mean layer of 0–4km AGL for all CG lightning ﬂashes in Poland. Yellow lines indicate the interquartile range, i.e., position of the 25th percentile (Q1) and 75th percentile (Q3), and the red line indicates the median (Q2). Relative humidity values are derived from ERA5 reanalysis for a period 2002–19. (Fig. 11d). A mean 0–4-km relative humidity turned out to be lower resolution of prior reanalyses. We also evaluated if the most consistent in terms of value distribution, and had a convective parameters can serve as predictors of speciﬁc well-deﬁned peak at 75%. These results are consistent with CG lightning ﬂash rate intensity categories (an aspect Westermayer et al. (2017), who also highlighted the impor- rarely studied in prior research) and investigated their spa- tance of relative humidity with respect to convection initiation tiotemporal climatological variability. We proposed a new across Europe. That study concluded that the probability for deﬁnition for thunderstorm intensity categories by using a thunderstorm development drops signiﬁcantly when the rel- 1-min peak CG lightning ﬂash rates. The proposed method- ative humidity is below 50%, which is very similar to the dis- ology features some limitations: severe, shallow convection tributions provided in our work, but was for a longer research may have poor lightning activity, the choice of timeframe period. and gridtodeﬁne ﬂash rates is arbitrary, and intracloud ﬂashes, which are important in lightning jump calculations (Farnell et al. 2018; Farnell and Rigo 2020), were omitted 5. Concluding remarks in our study (due to low quality of this data). Synthesis of In this study we used over 8 million CG lightning ﬂashes PERUN lightning data with ERA5 reanalysis yielded detected across Poland over the course of 18 years (2002–19) numerous ﬁndings, among which the most important are and intercompared speciﬁc CG lightning ﬂash rates with listed below: accompanying convective environments from ERA5. The hourly steps of ERA5 and a large sample size of the PERUN 1) A mean of 161 thunderstorm days occurs each year in lightning database provided the opportunity to explore the Poland, most frequently over the southeast. However, the diurnal cycles of thunderstorm environments and their rela- highest frequency of CG lightning ﬂashes is in eastern 22 21 tion to CG lightning ﬂash rates, which was not possible with a Poland with up to 3 ﬂashes km yr . 1464 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 2) July is the month with the largest number of CG lightning differences can be found if Poland is compared to regions of the ﬂashes (around 150 000 per year), reaching peak fre- Mediterranean or northwestern Europe that are more heavily quency typically between 1400 and 1600 UTC. Nocturnal inﬂuenced by their proximity to the sea surface and cyclonic CG lightning ﬂashes share an annual fraction of 4%–24% ´ activity (Tudurı and Ramis 1997; Cohuet et al. 2011; Holley for the speciﬁc years. et al. 2014). Mediterranean storms also typically feature much 3) The greatest convective instability occurs from June to higher instability with a peak of the season shifted toward late August between 1400 and 1600 UTC, with the highest low- summer and autumn (Riemann-Campe et al. 2009). Comparing level moisture at 1800 UTC. Thunderstorms occurring from our results to previous research, we noted that while enhanced March to May feature the highest midtropospheric lapse CAPE and DLS are typically necessary for storms producing rates. Patterns in wind shear are reversed to MUCAPE and high CG lightning ﬂash rates, only a small amount of CAPE have the highest values during the winter and at night. The (∼200 J kg ) is needed for the occurrence of any lightning, most conducive conditions for convection initiation in given successful convective initiation (Westermayer et al. 2017). Poland are between 1000 and 1400 UTC as evidenced by We also found that median MUCAPE and MUWMAX- the weakest convective inhibition and steepest lapse rates. SHEAR values for the extreme lightning category during sum- 4) The best overlap of convective instability and wind shear is mer correspond to large hail environments in central Europe in July and August, typically between 1400 and 2000 UTC (Pu ´cik ˇ et al. 2015; Taszarek et al. 2020). On the other hand, a (based on the 90th percentile of MUWMAXSHEAR). median DLS of 12.5 m s for the extreme lightning category 2 22 However, values exceeding 500 m s can occur from was much lower than 20 m s typically required for signiﬁcant March to November and at any time of the day. tornadoes and very large hail. This indicates that while there is 5) Thunderstorms in Poland are the most frequent in MUCAPE some relationship between the relative number of ESWD 21 21 below 1000 J kg and DLS between 8 and 15 m s .When reports and CG lightning ﬂash rates, the use of lightning data greater convective instability is available, DLS is typically as a proxy for convective hazards features certain limitations. lower (warm-season thunderstorms). Conversely, in highly As we showed in this study, predicting CG lightning ﬂash rates sheared environments, convective instability is typically mar- with convective parameters is a very challenging task and its ginal (cold-season thunderstorms). Situations with both high application in operational forecasting is limited. However, it is MUCAPE and DLS are rare in Poland. LSHC environments worth highlighting that forecasting convective hazards such as are the most frequent over southeastern Poland, while HSLC large hail, tornadoes, and severe wind has always been a very are the most common over the northwest. challenging task for operational forecasters across Europe 6) Along with increasing MUCAPE and DLS, peak CG light- (Brooks et al. 2011). The use of CG lightning ﬂash rates for ning ﬂash rates increase as well. However, a combination of studying storm intensity offers a much larger sample size and these two (MUWMAXSHEAR) demonstrated the highest spatial homogeneity compared to severe weather reports that (but still limited) skill compared to MUCAPE or DLS alone. are biased by population density. With improvements in 7) The vast majority of thunderstorms had a low to midlevel ground-based lightning detection networks, we expect similar relative humidity higher than 60%. studies to be developed in future for other regions of the world, 8) Proposed thunderstorm intensity proxy based on CG thus contributing to better understanding of local thunderstorm lightning ﬂash rates demonstrated a correlation with the climatologies and their accompanying environments. number of ESWD severe weather reports. Acknowledgments. This research was funded by the Insti- The results listed above are broadly consistent with prior tute of Meteorology and Water Management - Polish studies concerning the climatological aspects of lightning Research Institute (project S7), the Priority Research Area (Anderson and Klugmann 2014; Poelman et al. 2016; Taszarek Anthropocene under the program “Excellence Initiative} et al. 2019; Enno et al. 2020), convective environments accom- Research University” at the Jagiellonian University in panying central European thunderstorms (Pu ´cik ˇ et al. 2015; Krakow, ´ and supported by a grant from the Polish National Taszarek et al. 2020; Walawender et al. 2017; Westermayer Science Centre (project 2020/39/D/ST10/00768). The reanal- et al. 2017), and the comparison of satellite observations of ysis computations were performed in Poznan Supercomput- thunderstorms with ERA-Interim environments over tropical ing and Networking Center (project 448). and subtropical regions (Liu et al. 2020). Given a complex European orography and coastline that lead to high variability in local thunderstorm climatologies, Data availability statement. ERA5 data (temperature, spe- focusing on smaller areas (e.g., area of Poland) can provide ciﬁc humidity, geopotential, pressure, U, and V) were down- more precise information with respect to what is happening loaded from the European Centre for Medium-Range on the regional scale. This is conﬁrmed by comparison to Weather Forecasts (ECMWF), Copernicus Climate Change Service (C3S) at Climate Data Store (https://cds.climate. prior research that showed maximum thunderstorm activity in copernicus.eu/). PERUN lightning data was provided by the Poland being shifted by more than a month compared to Polish Institute of Meteorology and Water Management - neighboring parts of Europe (Taszarek et al. 2020). Concern- ing spatiotemporal distributions of convective environments, National Research Institute and due to the proprietary nature the area of Poland seems to not differ substantially from the of the data, cannot be made openly available. Contact rafal. area of eastern and central Europe. However, signiﬁcant lewandowski@imgw.pl for usage information. AUGUST 2022 PO R E ˛ BA ET A L . 1465 TABLE A1. The 10th, 50th, and 90th percentiles for selected convective parameters associated with weak (wea), moderate (mod), intense (int), and extreme (ext) CG lightning ﬂash rate categories. EL = equilibrium level; MUEL = MU equilibrium level; WBZ = wet bulb zero; STP = signiﬁcant tornado parameter; CPTP = cloud physics thunder parameter. 10th percentile 50th percentile 90th percentile Wea Mod Int Ext Wea Mod Int Ext Wea Mod Int Ext MUCAPE (J kg ) 62 116 190 248 448 572 724 834 1334 1475 1611 1714 MLCAPE (J kg ) 0 1 5 5 223 298 396 455 925 1048 1180 1265 8 8 MUCAPE from 0 to 220 C(J kg ) 25 54 92 124 205 244 289 320 447 481 515 540 MUCIN (J kg ) 243 248 256 263 26 27 29 211 0000 MLCIN (J kg ) 284 299 2122 2152 212 216 222 229 0000 MULCL (m AGL) 438 430 432 442 975 975 1000 1050 1875 1850 1850 1925 MLLCL (m AGL) 450 450 470 470 875 875 875 875 1575 1550 1525 1500 MULFC (m AGL) 740 850 975 1075 1600 1642 1750 1850 2550 2575 2625 2725 MLLFC (m AGL) 0 270 390 340 1725 1800 1900 2025 2725 2775 2875 3000 MLEL (m AGL) 0 3900 2875 3150 7000 7900 8700 9140 10 700 11 000 11 263 11 400 MUEL (m AGL) 4550 5592 6700 7600 8500 9300 10 100 10 600 11 400 11 564 11 800 11 900 MLEL temperature (8C) 9.5 1.2 21.7 22.0 227.9 233.2 237.7 239.5 2521. 253.6 254.9 255.5 MUEL temperature (8C) 212.7 217.9 224.0 228.4 239.1 243.8 247.8 250.0 256.2 257.3 258.2 258.9 MUMIXR (g kg ) 7.0 7.8 8.7 9.3 10.2 10.8 11.5 12.0 13.3 13.6 14.1 14.4 MLMIXR (g kg ) 6.9 7.6 8.6 9.4 10.1 10.6 11.3 11.8 12.8 13.1 13.4 13.7 0–1-km lapse rate (K km ) 2.7 2.4 2.0 1.4 7.0 6.7 6.3 5.8 9.9 9.8 9.5 9.2 0–3-km lapse rate (K km ) 5.2 5.2 5.1 4.9 6.7 6.6 6.6 6.4 8.1 8.0 7.9 7.8 3–6-km lapse rate (K km ) 5.6 5.7 5.6 5.6 6.2 6.2 6.1 6.1 7.0 6.9 6.8 6.7 2-m dewpoint (8C) 9.0 10.5 12.3 13.6 14.7 15.5 16.4 17.2 18.5 18.9 19.2 19.5 2-m temperature (8C) 14.4 15.6 16.9 17.7 20.0 20.7 21.5 21.9 25.8 26.1 26.5 26.8 850-hPa dewpoint (8C) 1.6 3.2 5.2 6.6 8.1 8.9 10.0 10.8 12.1 12.5 13.0 13.6 850-hPa temperature (8C) 5.7 7.5 9.5 11.0 12.2 13.0 14.1 15.0 16.4 16.8 17.4 17.9 700-hPa dewpoint (8C) 29.9 27.8 25.9 24.8 22.8 22.2 21.3 20.5 1.4 1.7 2.2 2.8 700-hPa temperature (8C) 24.5 22.9 21.1 0.3 1.4 2.1 3.2 4.0 5.2 5.5 5.9 6.3 500-hPa dewpoint (8C) 233.3 231.3 228.8 226.7 222.4 221.4 219.9 218.6 214.9 214.6 214.1 213.5 500-hPa temperature (8C) 221.7 219.9 217.8 216.2 214.2 213.5 212.5 211.8 210.7 210.5 210.2 210.0 Height of WBZ (m AGL) 1850 2100 2375 2575 2800 2900 3100 3250 3400 3450 3550 3600 Effective shear (m s ) 2.41 2.8 3.4 3.9 7.19 7.7 8.5 9.6 14.4 14.9 15.9 17.0 0–1-km wind shear (m s ) 1.7 1.7 1.8 2.1 5.1 5.1 5.4 5.8 10.7 10.5 10.8 11.2 0–3-km wind shear (m s ) 3.1 3.3 3.6 4.2 8.2 8.5 9.1 10.1 11.1 15.8 16.3 16.9 0–6-km wind shear (m s _ 4.1 4.1 4.7 5.4 11.1 11.1 11.6 12.7 21.3 20.5 20.8 21.7 1–3-km mean wind (m s ) 3.3 3.2 3.4 3.8 8.2 8.2 8.4 8.8 14.7 14.3 14.4 14.6 2 22 0–1-km SRH (m s ) 22.5 20.8 0.6 2.6 33.4 35.3 38.5 45.0 117.2 118.9 124.5 139.8 2 22 0–3-km SRH (m s ) 15.7 17.7 20.9 26.9 65.8 69.1 75.7 86.1 167.2 172.5 184.1 204.3 STP (Thompson et al. 2007) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 SCP (Thompson et al. 2007) 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 1.0 1.4 2.0 2.8 HSI (Czernecki et al. 2019) 0.0 0.0 0.0 0.1 0.6 0.7 0.8 0.9 1.7 1.8 1.9 2.1 CPTP (Bright et al. 2004) 0.0 0.0 0.0 1.5 18.9 33.9 53.5 68.2 123.7 138.5 153.0 163.2 Precipitable water (mm) 18.7 21.1 24.2 26.9 29.5 31.4 33.9 36.0 39.0 39.9 41.1 42.7 2 22 MUWMAXSHEAR (m s ) 81.2 108.5 142.9 183.2 292.1 331.1 398.4 479.9 676.1 730.3 822.6 915.7 2 22 MLWMAXSHEAR (m s ) 0.2 14.1 29.6 32.9 188.9 216.6 264.1 309.0 518.6 566.0 642.4 712.5 APPENDIX variability and ENSO inﬂuence. Int. J. Climatol., 34,81–97, https://doi.org/10.1002/joc.3667. Statistical Distribution of Convective Indices in }}, and M. K. Tippett, 2015: The characteristics of United States hail reports: 1955–2014. Electron. J. Severe Storms Thunderstorm Intensity Classes Meteor., 10 (3), 1–31. This appendix contains statistical distribution (10th, 50th, }}, D. J. Karoly, and G. Mills, 2011: A severe thunderstorm cli- and 90th percentiles) of all convective indices that were used matology for Australia and associated thunderstorm environ- in this study in function of thunderstorm intensity Table A1. ments. Aust. Meteor. Oceanogr. J., 61, 143–158, https://doi. org/10.22499/2.6103.001. Anderson, G., and D. Klugmann, 2014: A European lightning REFERENCES density analysis using 5 years of ATDnet data. Nat. Hazards Earth Syst. Sci., 14,815–829, https://doi.org/10.5194/nhess-14- Allen, J. T., and D. J. Karoly, 2014: A climatology of Australian 815-2014. severe thunderstorm environments 1979–2011: Inter-annual 1466 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 Anderson-Frey, A. K.,Y.P.Richardson, A. R. Dean,R.L. a new hodograph technique. Wea. Forecasting, 15,61–79, Thompson, and B. T. Smith, 2016: Investigation of near- https://doi.org/10.1175/1520-0434(2000)015,0061:PSMUAN. storm environments for tornado events and warnings. Wea. 2.0.CO;2. Forecasting, 31, 1771–1790, https://doi.org/10.1175/WAF-D- Burgesser, R. E., 2017: Assessment of the World Wide Lightning 16-0046.1. Location Network (WWLLN) detection efﬁciency by com- }}, }}, }}, }},and }}, 2019: Characteristics of tornado parison to the Lightning Imaging Sensor (LIS). Quart. J. events and warnings in the southeastern United States. Wea. Roy. Meteor. Soc., 143, 2809–2817, https://doi.org/10.1002/qj. Forecasting, 34, 1017–1034, https://doi.org/10.1175/WAF-D- 3129. 18-0211.1. Calhoun, K. M., D. R. MacGorman, C. L. Ziegler, and M. I. Big- Antonescu, B., and S. Burcea, 2010: A cloud-to-ground lightning gerstaff, 2013: Evolution of lightning activity and storm climatology for Romania. Mon. Wea. Rev., 138, 579–591, charge relative to dual-doppler analysis of a high-precipita- https://doi.org/10.1175/2009MWR2975.1. tion supercell storm. Mon. Wea. Rev., 141, 2199–2223, https:// Apke,J.M., J. R. Mecikalski, K. Bedka,E. W. McCaulJr.,C.R. doi.org/10.1175/MWR-D-12-00258.1. Homeyer, and C. P. Jewett, 2018: Relationships between Carlson, T. N., and F. H. Ludlam, 1968: Conditions for the occur- deep convection updraft characteristics and satellite-based rence of severe local storms. Tellus, 20, 203–226, https://doi. super rapid scan mesoscale atmospheric motion vector–der- org/10.3402/tellusa.v20i2.10002. ived ﬂow. Mon. Wea. Rev., 146, 3461–3480, https://doi.org/10. Celinski-Mysław, D., and D. Matuszko, 2014: An analysis of 1175/MWR-D-18-0119.1. selected cases of derecho in Poland. Atmos. Res., 149, 263– Bielec-Ba˛kowska, Z., 2002: Zro ´zni ˙ cowanie przestrzenne i 281, https://doi.org/10.1016/j.atmosres.2014.06.016. zmiennos ´c ´ wieloletnia wystRepowania burz w Polsce }}, A. Palarz, and M. Taszarek, 2020: Climatology and atmo- (1949–1998). Wydawnictwo Uniwersytetu Sla˛skiego, 159 pp. spheric conditions associated with cool season bow echo }}, 2003: Long-term variability of thunderstorms occurrence in storms in Poland. Atmos. Res., 240, 104944, https://doi.org/10. Poland in the 20th century. Atmos. Res., 67–68,35–52, https:// 1016/j.atmosres.2020.104944. Chernokulsky, A. V., M. V. Kurgansky, and I. I. Mokhov, 2019: doi.org/10.1016/S0169-8095(03)00082-6. }}, 2013: Burze i grady w Polsce (Thunderstorms and hails in On characteristic reanalysis-based values of convective insta- Poland). Prz. Geogr., 132,99–132, https://doi.org/10.4467/ bility indices for Northern Eurasia tornadoes. IOP Conf. Ser. 20833113PG.13.005.1095. Earth Environ. Sci., 231, 012012, https://doi.org/10.1088/1755- Biron, D., 2009: LAMPINET-Lightning detection in Italy. Light- 1315/231/1/012012. ning: Principles, Instruments and Applications: Review of Christian, H. J., and Coauthors, 1999: The lightning imaging sen- Modern Lightning Research, H. D. Betz, U. Schumann, and sor. 11th Int. Conf. on Atmospheric Electricity, H.J.Christian, E. Laroche, Eds., Springer, 141–159. Ed., NASA Conf. Publ. NASA/CP-1999-209261, 746–749. Blair, S. F., and Coauthors, 2017: High-resolution hail observa- Coffer, B. E., and M. D. Parker, 2015: Impacts of increasing low- tions: Implications for NWS warning operations. Wea. level shear on supercells during the early evening transition. Forecasting, 32, 1101–1119, https://doi.org/10.1175/WAF-D- Mon. Wea. Rev., 143, 1945–1969, https://doi.org/10.1175/ 16-0203.1. MWR-D-14-00328.1. Bodzak, P., 2006: Detekcja i lokalizacja wyładowan ´ atmosferycz- Cohuet, J. B., R. Romero, V. Homar, V. Ducrocq, and C. Ramis, 2011: Initiation of a severe thunderstorm over the Mediterra- nych (Lightning Detection and Localization). Instytut Mete- orologii i Gospodarki Wodnej, 135 pp. nean Sea. Atmos. Res., 100,603–620, https://doi.org/10.1016/j. }}, Z. Dziewit, and G. Wojciech, 2006: The ratio of cloud to atmosres.2010.11.002. cloud-to-ground ﬂashes in summer seasons for Poland terri- Colman, B. R., 1990: Thunderstorms above frontal surfaces in tory on the basis of data from PERUN system. 19th Int. environments without positive CAPE. Part II: Organization Lightning Detection Conf. (ILDC)/1st Int. Lightning Meteo- and instability mechanisms. Mon. Wea. Rev., 118, 1123–1144, rology Conf. (ILMC), Tucson, AR, Vaisala, https://my. https://doi.org/10.1175/1520-0493(1990)118,1123:TAFSIE.2. vaisala.net/en/events/ildcilmc/archive/Pages/ILDC-2006-Archive. 0.CO;2. aspx. Cummins, K. L., and M. J. Murphy, 2009: An overview of light- Bright, D. R., M. S. Wandishin, R. E. Jewell, and S. J. Weiss, ning locating systems: History, techniques, and data uses, with an in-depth look at the U.S. NLDN. IEEE Trans. 2004: A physically based parameter for lightning prediction Electromagn. Compat., 51, 499–518, https://doi.org/10.1109/ and its calibration in ensemble forecasts. 22nd Conf. on Severe Local Storms, Hyannis, MA, Amer. Meteor. Soc., 4.3., TEMC.2009.2023450. http://ams.confex.com/ams/pdfpapers/84173.pdf. Czernecki, B., M. Taszarek, L. Kolendowicz, and J. Konarski, Brooks, H. E., J. W. Lee, and J. P. Craven, 2003: The spatial dis- 2016: Relationship between human observations of thunder- tribution of severe thunderstorm and tornado environments storms and the PERUN lightning detection network in from global reanalysis data. Atmos. Res., 67–68,73–94, Poland. Atmos. Res., 167,118–128, https://doi.org/10.1016/j. https://doi.org/10.1016/S0169-8095(03)00045-0. atmosres.2015.08.003. }}, A. R. Anderson, K. Riemann, I. Ebbers, and H. Flachs, }}, }},M.Marosz, M. Po ´ łrolniczak, L. Kolendowicz, A. Wys- 2007: Climatological aspects of convective parameters from zogrodzki, and J. Szturc, 2019: Application of machine learn- the NCAR/NCEP reanalysis. Atmos. Res., 83, 294–305, ing to large hail prediction - The importance of radar reﬂec- tivity, lightning occurrence and convective parameters https://doi.org/10.1016/j.atmosres.2005.08.005. }}, and Coauthors, 2011: Evaluation of European Storm Fore- derived from ERA5. Atmos. Res., 227, 249–262, https://doi. cast Experiment (ESTOFEX) forecasts. Atmos. Res., 100, org/10.1016/j.atmosres.2019.05.010. 538–546, https://doi.org/10.1016/j.atmosres.2010.09.004. Deierling, W., and W. A. Petersen, 2008: Total lightning activity Bunkers, M. J., B. A. Klimowski, J. W. Zeitler, R. L. Thompson, as an indicator of updraft characteristics. J. Geophys. Res., and M. L. Weisman, 2000: Predicting supercell motion using 113, D16210, https://doi.org/10.1029/2007JD009598. AUGUST 2022 PO R E ˛ BA ET A L . 1467 Dennis, E. J., and M. R. Kumjian, 2017: The impact of vertical Farr, T. G., and Coauthors, 2007: The Shuttle Radar Topography wind shear on hail growth in simulated supercells. J. Atmos. Mission. Rev. Geophys., 45, RG2004, https://doi.org/10.1029/ 2005RG000183. Sci., 74, 641–663, https://doi.org/10.1175/JAS-D-16-0066.1. DiGangi, E. A., M. Stock, and J. Lapierre, 2021: Thunder hours: Feudale, L., A. Manzato, and S. Micheletti, 2013: A cloud-to ground lightning climatology for north-eastern Italy. Adv. Sci. How old methods offer new insights into thunderstorm clima- Res., 10,77–84, https://doi.org/10.5194/asr-10-77-2013. tology. Bull. Amer. Meteor. Soc., 103, E548–E569, https://doi. Fluck, E., M. Kunz, P. Geissbuehler, and S. P. Ritz, 2021: Radar- org/10.1175/BAMS-D-20-0198.1. based assessment of hail frequency in Europe. Nat. Hazards Doswell, C. A., 1985: The operational meteorology of convective Earth Syst. Sci., 21,683–701, https://doi.org/10.5194/nhess-21- weather: Volume II: Storm scale analysis. NOAA Tech. 683-2021. Memo. ERL ESG-15, 240 pp., https://repository.library.noaa. Gatlin, P. N., and S. J. Goodman, 2010: A total lightning trending gov/view/noaa/11215. algorithm to identify severe thunderstorms. J. Atmos. Oceanic }}, 2001: Severe convective storms}An overview. Severe Con- Technol., 27,3–22, https://doi.org/10.1175/2009JTECHA1286. vective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 1–26, https://doi.org/10.1175/0065-9401-28.50.1. Gatzen, C., 2011: A 10-year climatology of cold-season narrow }}, and E. N. Rasmussen, 1994: The effect of neglecting the vir- cold-frontal rainbands in Germany. Atmos. Res., 100, 366– tual temperature correction on CAPE calculations. Wea. Fore- 370, https://doi.org/10.1016/j.atmosres.2010.09.018. casting, 9, 625–629, https://doi.org/10.1175/1520-0434(1994)009 }}, A. H. Fink, D. M. Schultz, and J. G. Pinto, 2020: An 18- ,0625:TEONTV.2.0.CO;2. year climatology of derechos in Germany. Nat. Hazards }}, and D. M. Schultz, 2006: On the use of indices and parame- Earth Syst. Sci., 20,1335–1351, https://doi.org/10.5194/nhess- ters in forecasting severe storms. Electron. J. Severe Storms 20-1335-2020. Meteor., 1 (3), 1–22. Geerts, B., and Coauthors, 2017: The 2015 Plains Elevated Con- }}, H. E. Brooks, and R. A. Maddox, 1996: Flash ﬂood fore- vection at Night ﬁeld project. Bull. Amer. Meteor. Soc., 98, casting: An ingredients-based methodology. Wea. Forecasting, 767–786, https://doi.org/10.1175/BAMS-D-15-00257.1. 11,560–581, https://doi.org/10.1175/1520-0434(1996)011,0560: Gensini, V. A., and W. S. Ashley, 2011: Climatology of potentially FFFAIB.2.0.CO;2. severe convective environments from North American Dotzek, N., P. Groenemeijer, B. Feuerstein, and A. M. Holzer, regional reanalysis. Electron. J. Severe Storms Meteor., 6 (8), 2009: Overview of ESSL’s severe convective storms research 1–40. using the European Severe Weather Database ESWD. Gieysztor, A., 2006: Mitologia Słowian. Wydawnictwo Uniwersy- Atmos. Res., 93, 575–586, https://doi.org/10.1016/j.atmosres. tetu Warszawskiego, 407 pp. 2008.10.020. Goodman, S. J., D. E. Buechler, P. D. Wright, and W. D. Rust, Edwards, R., J. G. LaDue, J. T. Ferree, K. Scharfenberg, 1988: Lightning and precipitation history of a microburst-pro- C. Maier, and W. L. Coulbourne, 2013: Tornado intensity esti- ducing storm. Geophys. Res. Lett., 15, 1185–1188, https://doi. mation: Past, present, and future. Bull. Amer. Meteor. Soc., org/10.1029/GL015i011p01185. 94,641–653, https://doi.org/10.1175/BAMS-D-11-00006.1. ´ ´ ´ Grabowska, K., 2001: Zmiennoscaktywnosci burzowej w }}, J. T. Allen, and G. W. Carbin, 2018: Reliability and climato- niekto ´rych regionach geograﬁcznych Polski. Prace i Studia logical impacts of convective wind estimations. J. Appl. Geograﬁczne UW, 29, 181–190. Meteor. Climatol., 57, 1825–1845, https://doi.org/10.1175/ Grant, B. N., 1995: Elevated cold-sector severe thunderstorms: A JAMC-D-17-0306.1. preliminary study. Natl. Wea. Dig., 19,25–31. Elmore, K. L., Z. L. Flamig, V. Lakshmanan, B. T. Kaney, Groenemeijer, P., and Coauthors, 2017: Severe convective storms V. Farmer, H. D. Reeves, and L. P. Rothfusz, 2014: MPING: in Europe: Ten years of research at the European Severe Crowd-sourcing weather reports for research. Bull. Amer. Storms Laboratory. Bull. Amer. Meteor. Soc., 98, 2641–2651, Meteor. Soc., 95, 1335–1342, https://doi.org/10.1175/BAMS-D- https://doi.org/10.1175/BAMS-D-16-0067.1. 13-00014.1. Gropp, M. E., and C. E. Davenport, 2018: The impact of the noc- Enno, S. E., 2011: A climatology of cloud-to-ground lightning turnal transition on the lifetime and evolution of supercell over Estonia, 2005–2009. Atmos. Res., 100,310–317, https:// thunderstorms in the Great Plains. Wea. Forecasting, 33, doi.org/10.1016/j.atmosres.2010.08.024. 1045–1061, https://doi.org/10.1175/WAF-D-17-0150.1. }}, J. Sugier, R. Alber, and M. Seltzer, 2020: Lightning ﬂash Haberlie, A. M., and W. S. Ashley, 2019: A radar-based climatol- density in Europe based on 10 years of ATDnet data. Atmos. ogy of mesoscale convective systems in the United States. J. Res., 235, 104769, https://doi.org/10.1016/j.atmosres.2019. Climate, 32,1591–1606, https://doi.org/10.1175/JCLI-D-18- 0559.1. Farnell, C., and T. Rigo, 2020: The lightning jump algorithm for Hanstrum, B. N., G. A. Mills, A. Watson, J. P. Monteverdi, and nowcasting convective rainfall in Catalonia. Atmosphere, 11, C. A. Doswell, 2002: The cool-season tornadoes of California 397, https://doi.org/10.3390/atmos11040397. and Southern Australia. Wea. Forecasting, 17,705–722, }}, }}, and N. Pineda, 2017: Lightning jump as a nowcast pre- https://doi.org/10.1175/1520-0434(2002)017,0705: dictor: Application to severe weather events in Catalonia. TCSTOC.2.0.CO;2. Atmos. Res., 183, 130–141, https://doi.org/10.1016/j.atmosres. Hersbach, H. B., and Coauthors, 2020: The ERA5 global reanaly- 2016.08.021. sis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi. }}, }},and }}, 2018: Exploring radar and lightning varia- org/10.1002/qj.3803. bles associated with the Lightning Jump. Can we predict the Holle, R. L., 2014: Diurnal variations of NLDN-reported cloud- size of the hail? Atmos. Res., 202,175–186, https://doi.org/10. to-ground lightning in the United States. Mon. Wea. Rev., 1016/j.atmosres.2017.11.019. 142,1037–1052, https://doi.org/10.1175/MWR-D-13-00121.1. 1468 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 }}, K. L. Cummins, and W. A. Brooks, 2016: Seasonal, 221–251, https://doi.org/10.1175/1520-0469(1989)046,0221:LRRTTS monthly, and weekly distributions of NLDN and GLD360 .2.0.CO;2. ¨ ¨ cloud-to-ground lightning. Mon. Wea. Rev., 144, 2855–2870, Makela, A., P. Rossi, and D. M. Schultz, 2011: The daily cloud-to- https://doi.org/10.1175/MWR-D-16-0051.1. ground lightning ﬂash density in the contiguous United States Holley, D. M., S. R. Dorling, C. J. Steele, and N. Earl, 2014: A cli- and Finland. Mon. Wea. Rev., 139, 1323–1337, https://doi.org/ matology of convective available potential energy in Great 10.1175/2010MWR3517.1. Britain. Int. J. Climatol., 34, 3811–3824, https://doi.org/10. Markowski, P., and Y. Richardson, 2010: Mesoscale Meteorology 1002/joc.3976. in Midlatitudes. John Wiley & Sons, Ltd., 407 pp. Houze, R. A., Jr., 2018: 100 years of research on mesoscale con- Mathias, L., V. Ermert, F. D. Kelemen, P Ludwig, and J. G. Pinto, 2017: Synoptic analysis and hindcast of an intense bow echo in vective systems. A Century of Progress in Atmospheric and Related Sciences: Celebrating the American Meteorological western Europe: The 9 June 2014 storm. Wea. Forecasting, 32, Society Centennial, Meteor. Monogr.,No. 59, Amer.Meteor. 1121–1141, https://doi.org/10.1175/WAF-D-16-0192.1. }}, P. Ludwig, and J. G. Pinto, 2019: Synoptic-scale conditions Soc., https://doi.org/10.1175/AMSMONOGRAPHS-D-18- and convection-resolving hindcast experiments of a cold-sea- 0001.1. Ingrosso, R., P. Lionello, M. M. Miglietta, and G. A. Salvadori, son derecho on 3 January 2014 in western Europe. Nat. Haz- 2020: Statistical investigation of mesoscale precursors of sig- ards Earth Syst. Sci., 19, 1023–1040, https://doi.org/10.5194/ niﬁcant tornadoes: The Italian case study. Atmosphere, 11, nhess-19-1023-2019. NOAA/NWS, 2021: Nautical dawn. NOAA/NWS Glossary, 301, https://doi.org/10.3390/atmos11030301. Johns, R. H., and C. A. Doswell, 1992: Severe local storms fore- https://w1.weather.gov/glossary/index.php?word=nautical+ casting. Wea. Forecasting, 7,588–612, https://doi.org/10.1175/ dawn. Novak ´ , P., and H. Kyznarova ´, 2011: Climatology of lightning in 1520-0434(1992)007,0588:SLSF.2.0.CO;2. Kingﬁeld, D. M., K. M. Calhoun, and K. M. de Beurs, 2017: the Czech Republic. Atmos. Res., 100, 318–333, https://doi. Antenna structures and cloud-to-ground lightning location: org/10.1016/j.atmosres.2010.08.022. 1995–2015. Geophys. Res. Lett., 44, 5203–5212, https://doi.org/ Orville, R. E., 1991: Lightning ground ﬂash density in the contigu- ous United States}1989. Mon. Wea. Rev., 119, 573–577, 10.1002/2017GL073449. Koehler, T. L., 2020: Cloud-to-ground lightning ﬂash density and https://doi.org/10.1175/1520-0493(1991)119,0573:LGFDIT.2.0. thunderstorm day distributions over the contiguous United CO;2. States derived from NLDN measurements: 1993–2018. Mon. }}, and G. R. Hufﬁnes, 2001: Cloud-to-ground lightning in the United States: NLDN results in the ﬁrst decade, 1989–98. Wea. Rev., 148, 313–332, https://doi.org/10.1175/MWR-D-19- 0211.1. Mon. Wea. Rev., 129,1179–1193, https://doi.org/10.1175/1520- Kolendowicz, L., 2006: The inﬂuence of synoptic situations on the 0493(2001)129,1179:CTGLIT.2.0.CO;2. }}, }},W.R.Burrows,R.L.Holle, and K. L. Cummins, occurrence of days with thunderstorms during a year in the 2002: The North American Lightning Detection Network territory of Poland. Int. J. Climatol., 26, 1803–1820, https:// doi.org/10.1002/joc.1348. (NALDN)}First results: 1998–2000. Mon. Wea. Rev., 130, Krennert, T., G. Pistotnik, R. Kaltenberger, and C. Csekits, 2018: 2098–2109, https://doi.org/10.1175/1520-0493(2002)130,2098: TNALDN.2.0.CO;2. Crowdsourcing of weather observations at national meteoro- Pacey, G. P., D. M. Schultz, and L. Garcia-Carreras, 2021: Severe logical and hydrological services in Europe. Adv. Sci. Res., 15,71–76, https://doi.org/10.5194/asr-15-71-2018. convective windstorms in Europe: Climatology, preconvective Lang, T. J., and Coauthors, 2004: The Severe Thunderstorm Elec- environments, and convective mode. Wea. Forecasting, 36, 237–252, https://doi.org/10.1175/WAF-D-20-0075.1. triﬁcation and Precipitation Study. Bull. Amer. Meteor. Soc., Perez, A. H., L. J. Wicker, and R. E. Orville, 1997: Characteristics 85, 1107–1126, https://doi.org/10.1175/BAMS-85-8-1107. Lehmiller, G. S., H. B. Bluestein, P. J. Neiman, F. M. Ralph, and of cloud-to-ground lightning associated with violent torna- W. F. Feltz, 2001: Wind structure in a supercell thunderstorm does. Wea. Forecasting, 12,428–437, https://doi.org/10.1175/ 1520-0434(1997)012,0428:COCTGL.2.0.CO;2. as measured by a UHF wind proﬁler. Mon. Wea. Rev., 129, Pilguj, N., M. Taszarek, Ł. Pajurek, and M. Kryza, 2019: High-res- 1968–1986, https://doi.org/10.1175/1520-0493(2001)129,1968: WSIAST.2.0.CO;2. olution simulation of an isolated tornadic supercell in Poland Lin, Y., and M. R. Kumjian, 2021: Inﬂuences of CAPE on hail on 20 June 2016. Atmos. Res., 218, 145–159, https://doi.org/10. 1016/j.atmosres.2018.11.017. production in simulated supercell storms. J. Atmos. Sci., 79, Pilorz, W., 2015: Very large hail occurrence in Poland from 2007 179–204, https://doi.org/10.1175/JAS-D-21-0054. Liu, N., C. Liu, B. Chen, and E. Zipser, 2020: What are the favor- to 2015. Contemp. Trends. Geosci., 4,45–55, https://doi.org/ able large-scale environments for the highest-ﬂash-rate thun- 10.1515/ctg-2015-0005. Poelman, D. R., W. Schulz, G. Diendorfer, and M. Bernardi, 2016: derstorms on Earth? J. Atmos. Sci., 77, 1583–1612, https://doi. The European lightning location system EUCLID}Part 2: org/10.1175/JAS-D-19-0235.1. Łoboda, M., H. D. Betz, P. Baranski, J. Wiszniowski, and Z. Observations. Nat. Hazards Earth Syst, 16,607–616, https:// Dziewit, 2009: New lightning detection networks in doi.org/10.5194/nhess-16-607-2016. Pore Rba, S., and Z. Ustrnul, 2020: Forecasting experiences associ- Poland}LINET and LLDN. Open Atmos. Sci. J., 3,29–38, ated with supercells over South-Western Poland on July 7, https://doi.org/10.2174/1874282300903010029. Lorenc, H., 2005: Atlas klimatu Polski. Instytut Meteorologii i 2017. Atmos. Res., 232, 104681, https://doi.org/10.1016/j. Gospodarki Wodnej, 116 pp. atmosres.2019.104681. Macgorman, D. R., D.W. Burgess, V.Mazur,W. D.Rust, Price, C., 2008: Lightning sensors for observing, tracking and now- W. L. Taylor, and B. C. Johnson, 1989: Lightning rates relative casting severe weather. Sensors, 8,157–170, https://doi.org/10. to tornadic storm evolution on 22 May 1981. J. Atmos. Sci., 46, 3390/s8010157. AUGUST 2022 PO R E ˛ BA ET A L . 1469 Pu ´cik ˇ T., P. Groenemeijer, D. Ryv ´ a, and M. Kolar, ´ 2015: Proxim- Smith, B. T., R. L. Thompson, J. S. Grams, C. Broyles, and H. E. ity soundings of severe and nonsevere thunderstorms in cen- Brooks, 2012: Convective modes for signiﬁcant severe thun- tral Europe. Mon. Wea. Rev., 143, 4805–4821, https://doi.org/ derstorms in the contiguous United States. Part I: Storm clas- 10.1175/MWR-D-15-0104.1. siﬁcation and climatology. Wea. Forecasting, 27, 1114–1135, Punge, H. J., K. M. Bedka, M. Kunz, and A. Reinbold, 2017: Hail https://doi.org/10.1175/WAF-D-11-00115.1. frequency estimation across Europe based on a combination Steiger, S. M., R. E. Orville, and L. D. Carey, 2007: Total light- of overshooting top detections and the ERA-INTERIM ning signatures of thunderstorm intensity over north Texas. reanalysis. Atmos. Res., 198,34–43, https://doi.org/10.1016/j. Part I: Supercells. Mon. Wea. Rev., 135, 3281–3302, https:// atmosres.2017.07.025. doi.org/10.1175/MWR3472.1. Reif, D. W., and H. B. Bluestein, 2017: A 20-year climatology of Stopa, M., 1962: Burze w Polsce. Prace Instytutu GeograﬁiPAN, nocturnal convection initiation over the central and southern 34,109–179. Great Plains during the warm season. Mon. Wea. Rev., 145, Sulik, S., 2021: Formation factors of the most electrically active thunderstorm days over Poland (2002–2020). Atmos. Res., 34, 1615–1639, https://doi.org/10.1175/MWR-D-16-0340.1. Riemann-Campe, K., K. Fraedrich, and F. Lunkeit, 2009: Global 100386, https://doi.org/10.1016/j.wace.2021.100386. climatology of convective available potential energy (CAPE) Surowiecki, A., and M. Taszarek, 2020: A 10-year radar-based cli- and convective inhibition (CIN) in ERA-40 reanalysis. matology of mesoscale convective system archetypes and der- Atmos. Res., 93, 534–545, https://doi.org/10.1016/j.atmosres. echos in Poland. Mon. Wea. Rev., 148,3471–3488, https://doi. 2008.09.037. org/10.1175/MWR-D-19-0412.1. Rodr´ıguez, O., and J. Bech, 2021: Tornadic environments in the Taszarek, M., B. Czernecki, and A. Kozioł, 2015: A cloud-to- IberianPeninsula andthe Balearic Islands basedonERA5 ground lightning climatology for Poland. Mon. Wea. Rev., reanalysis. Int. J. Climatol., 41,E1959–E1979, https://doi.org/ 143,4285–4304, https://doi.org/10.1175/MWR-D-15-0206.1. 10.1002/joc.6825. }}, and Coauthors, 2019: A climatology of thunderstorms across Said, R.K., M. J. Murphy, N.W.S. Demetriades,K.L. Cummins, Europe from a synthesis of multiple data sources. J. Climate, and U. S. Inan, 2011: Methodology and performance esti- 32, 1813–1837, https://doi.org/10.1175/JCLI-D-18-0372.1. mates of the GLD360 lightning detection network. Proc. 14th }}, J.T.Allen, T. Pu ´c ˇik, K. Hoogewind, and H. E. Brooks, Int. Conf. on Atmospheric Electricity, Rio de Janeiro, Brazil, 2020: Severe convective storms across Europe and the United ICAE, 243 pp., http://www.icae2011.net.br/upload/resumos. States. Part II: ERA5 environments associated with lightning, pdf. large hail, severe wind and tornadoes. J. Climate, 33, 10 263– Schultz, C. J., W. A. Petersen, and L. D. Carey, 2009: Preliminary 10 286, https://doi.org/10.1175/JCLI-D-20-0346.1. development and evaluation of lightning jump algorithms Thompson, R. L., R. Edwards, J. A. Hart, K. L. Elmore, and P. for the real-time detection of severe weather. J. Appl. Markowski, 2003: Close proximity soundings within supercell Meteor. Climatol., 48,2543–2563, https://doi.org/10.1175/ environments obtained from the Rapid Update Cycle. 2009JAMC2237.1. Wea. Forecasting, 18,1243–1261, https://doi.org/10.1175/1520- }}, }},and }}, 2011: Lightning and severe weather: A com- 0434(2003)018,1243:CPSWSE.2.0.CO;2. parison between total and cloud-to-ground lightning trends. }}, C. M. Mead, and R. Edwards, 2007: Effective storm-relative Wea. Forecasting, 26, 744–755, https://doi.org/10.1175/WAF- helicity and bulk shear in supercell thunderstorm environ- D-10-05026.1. ments. Wea. Forecasting, 22, 102–115, https://doi.org/10.1175/ }}, L. D. Carey, E. V. Schultz, and R. J. Blakeslee, 2017: Kine- WAF969.1. matic and microphysical signiﬁcance of lightning jumps versus }}, B. T. Smith, J. S. Grams, A. R. Dean, and C. Broyles, 2012: non-jump increases in total ﬂash rate. Wea. Forecasting, 32, Convective modes for signiﬁcant severe thunderstorms in the 275–288, https://doi.org/10.1175/WAF-D-15-0175.1. contiguous United States. Part II: Supercell and QLCS tor- Schultz, E. V.,C.J. Schultz,L. D.Carey,D.J.Cecil,and M. Bate- nado environments. Wea. Forecasting, 27,1136–1154, https:// man, 2016: Automated Storm tracking and the lightning doi.org/10.1175/WAF-D-11-00116.1. jump algorithm using GOES-R Geostationary Lightning Tudur´ı, E., and C. Ramis, 1997: The environments of signiﬁcant Mapper (GLM) proxy data. J. Operat. Meteor., 4,92–107, convective events in the western Mediterranean. Wea. Forecast- https://doi.org/10.15191/nwajom.2016.0407. ing, 12, 294–306, https://doi.org/10.1175/1520-0434(1997)012,0294: Schulz, W., K. Cummins, G. Diendorfer, and M. Dorninger, 2005: TEOSCE.2.0.CO;2. Cloud-to-ground lightning in Austria: A 10-year study using Ustrnul, Z., and D. Czekierda, 2009: Atlas ekstremalnych zjawisk data from a lightning location system. J. Geophys. Res., 110, meteorologicznych oraz sytuacji synoptycznych w Polsce. D09101, https://doi.org/10.1029/2004JD005332. Instytut Meteorologii i Gospodarki Wodnej, 182 pp. Seimon, A., J. T. Allen, T. A. Seimon, S. J. Talbot, and D. K. Van Delden, A., 2001: The synoptic setting of thunderstorms in Hoadley, 2016: Crowdsourcing the El Reno 2013 tornado: A western Europe. Atmos. Res., 56,89–110, https://doi.org/10. new approach for collation and display of storm chaser imag- 1016/S0169-8095(00)00092-2. ery for scientiﬁc applications. Bull. Amer. Meteor. Soc., 97, Verbout, S. M., H. E. Brooks, L. M. Leslie, and D. M. Schultz, 2069–2084, https://doi.org/10.1175/BAMS-D-15-00174.1. 2006: Evolution of the U.S. tornado database: 1954–2003. Sherburn, K. D., and M. D. Parker, 2014: Climatology and ingre- Wea. Forecasting, 21,86–93, https://doi.org/10.1175/WAF910. dients of signiﬁcant severe convection in high-shear, low- 1. CAPE environments. Wea. Forecasting, 29, 854–877, https:// Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H. Holzworth, doi.org/10.1175/WAF-D-13-00041.1. 2013: Highlights of a new ground-based, hourly global light- }}, }}, J. R. King, and G. M. Lackmann, 2016: Composite ning climatology. Bull. Amer. Meteor. Soc., 94, 1381–1391, environments of severe and nonsevere high-shear, low-CAPE https://doi.org/10.1175/BAMS-D-12-00082.1. convective events. Wea. Forecasting, 31,1899–1927, https:// Wacker, R. S., and R. E. Orville, 1999: Changes in measured doi.org/10.1175/WAF-D-16-0086.1. lightning ﬂash count and return stroke peak current after the 1470 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 1994 U.S. National Lightning Detection Network upgrade: 2. extremely severe and widespread geohazard in Poland. Envi- Theory. J. Geophys. Res., 104, 2159–2162, https://doi.org/10. ron. Socio-Econ. Stud., 6,10–16, https://doi.org/10.2478/ 1029/1998JD200059. environ-2018-0002. Walawender, E., R. Kielar, and Z. Ustrnul, 2017: Use of RegCM Wiens, K. C., S. A. Rutledge, and S. A. Tessendorf, 2005: The 29 gridded dataset for thunderstorm favorable conditions analy- June 2000 supercell observed during STEPS. Part II: Light- sis over Poland}Climatological approach. Theor. Appl. Cli- ning and charge structure. J. Atmos. Sci., 62, 4151–4177, matol., 127,229–240, https://doi.org/10.1007/s00704-015-1620- https://doi.org/10.1175/JAS3615.1. x. Williams, E., and Coauthors, 1999: The behavior of total lightning Weisman, M. L., and J. B. Klemp, 1982: The dependence of activity in severe Florida thunderstorms. Atmos. Res., 51, numerically simulated convective storms on vertical wind 245–265, https://doi.org/10.1016/S0169-8095(99)00011-3. shear and buoyancy. Mon. Wea. Rev., 110, 504–520, https:// Williams, E. R., M. E. Weber, and R. E. Orville, 1989: The rela- doi.org/10.1175/1520-0493(1982)110,0504:TDONSC.2.0. tionship between lightning type and convective state of thun- CO;2. derclouds. J. Geophys. Res., 94, 13 213–13220, https://doi.org/ Westermayer, A. T., P. Groenemeijer, G. Pistotnik, R. Sausen, 10.1029/JD094iD11p13213. and E. Faust, 2017: Identiﬁcation of favorable environments Zajac, B. A., and S. A. Rutledge, 2001: Cloud-to-ground lightning for thunderstorms in reanalysis data. Meteor. Z., 26,59–70, activity in the contiguous United States from 1995 to 1999. https://doi.org/10.1127/metz/2016/0754. Mon. Wea. Rev., 129, 999–1019, https://doi.org/10.1175/1520- Widawski, A., and W. Pilorz, 2018: The mesoscale convective sys- tems with bow echo radar signatures as an example of 0493(2001)129,0999:CTGLAI.2.0.CO;2. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Weather and Forecasting American Meteorological Society http://www.deepdyve.com/lp/american-meteorological-society/diurnal-and-seasonal-variability-of-era5-convective-parameters-in-B9KO4NveSa

Loading next page...

References (140)

P. Bodzak (2006)
135
B. R. Colman (1990)
1123
, 118
N. Liu (2020)
1583
, 77
R. Ingrosso (2020)
301
, 11
Z. Bielec-Bąkowska (2002)
159
T. L. Koehler (2020)
313
, 148
K. S. Virts (2013)
1381
, 94
K. D. Sherburn (2014)
854
, 29
B. Antonescu (2010)
579
, 138
B. A. Zajac (2001)
999
, 129
T. J. Lang (2004)
1107
, 85
H. B. Hersbach (2020)
1999
, 146
C. Farnell (2018)
175
, 202
E. V. Schultz (2016)
92
, 4
L. Kolendowicz (2006)
1803
, 26
B. N. Hanstrum (2002)
705
, 17
R. E. Orville (2002)
2098
, 130
E. Williams (1999)
245
, 51
C. A. Doswell (1996)
560
, 11
B. N. Grant (1995)
25
, 19
E. Walawender (2017)
229
, 127
A. Van Delden (2001)
89
, 56
M. L. Weisman (1982)
504
, 110
H. J. Christian (1999)
746
C. A. Doswell (1994)
625
, 9
L. Feudale (2013)
77
, 10
K. L. Cummins (2009)
499
, 51
C. Gatzen (2011)
366
, 100
R. Edwards (2013)
641
, 94
D. R. Macgorman (1989)
221
, 46
R. L. Thompson (2012)
1136
, 27
J. T. Allen (2014)
81
, 34
A. V. Chernokulsky (2019)
012012
, 231
P. N. Gatlin (2010)
3
, 27
P. Markowski (2010)
407
G. Anderson (2014)
815
, 14
E. A. DiGangi (2021)
E548
, 103
R. E. Orville (1991)
573
, 119
M. Taszarek (2020)
10 263
, 33
L. Mathias (2017)
1121
, 32
A. K. Anderson-Frey (2016)
1771
, 31
S. J. Goodman (1988)
1185
, 15
K. Grabowska (2001)
181
, 29
J. M. Apke (2018)
3461
, 146
D. Biron (2009)
141
A. Mäkelä (2011)
1323
, 139
B. E. Coffer (2015)
1945
, 143
C. Gatzen (2020)
1335
, 20
T. G. Farr (2007)
RG2004
, 45
R. Edwards (2018)
1825
, 57
E. R. Williams (1989)
13 213
, 94
W. Schulz (2005)
D09101
, 110
Púčik T. (2015)
4805
, 143
M. Stopa (1962)
109
, 34
R. K. Said (2011)
Methodology and performance estimates of the GLD360 lightning detection network
C. A. Doswell (1985)
The operational meteorology of convective weather: Volume II: Storm scale analysis
O. Rodríguez (2021)
E1959
, 41
S. Poręba (2020)
104681
, 232
E. Fluck (2021)
683
, 21
R. L. Thompson (2007)
102
, 22
R. L. Thompson (2003)
1243
, 18
J. B. Cohuet (2011)
603
, 100
Z. Ustrnul (2009)
182
K. Riemann-Campe (2009)
534
, 93
C. Farnell (2017)
130
, 183
M. Taszarek (2019)
1813
, 32
D. Celiński-Mysław (2014)
263
, 149
G. Anderson, D. Klugmann (2014)
A European lightning density analysis using 5 years of ATDnet data
, 14
C. J. Schultz (2017)
275
, 32
R. L. Holle (2014)
1037
, 142
T. Krennert (2018)
71
, 15
A. K. Anderson-Frey, Y. P. Richardson, A. R. Dean, R. L. Thompson, B. T. Smith (2016)
Investigation of near-storm environments for tornado events and warnings
, 31
S. E. Enno (2020)
104769
, 235
A. Seimon (2016)
2069
, 97
C. Farnell (2020)
397
, 11
Y. Lin (2021)
179
, 79
A. M. Haberlie (2019)
1591
, 32
A. Surowiecki (2020)
3471
, 148
D. M. Holley (2014)
3811
, 34
B. Antonescu, S. Burcea (2010)
A cloud-to-ground lightning climatology for Romania
, 138
J. T. Allen (2011)
143
, 61
N. Dotzek (2009)
575
, 93
R. E. Orville (2001)
1179
, 129
R. S. Wacker (1999)
2159
, 104
G. S. Lehmiller (2001)
1968
, 129
J. T. Allen, D. J. Karoly (2014)
A climatology of Australian severe thunderstorm environments 1979–2011: Inter-annual variability and ENSO influence
, 34
S. M. Steiger (2007)
3281
, 135
M. Taszarek (2015)
4285
, 143
K. M. Calhoun (2013)
2199
, 141
C. J. Schultz (2009)
2543
, 48
S. Sulik (2021)
100386
, 34
S. E. Enno (2011)
310
, 100
L. Mathias (2019)
1023
, 19
J. M. Apke, J. R. Mecikalski, K. Bedka, E. W. McCaul, C. R. Homeyer, C. P. Jewett (2018)
Relationships between deep convection updraft characteristics and satellite-based super rapid scan mesoscale atmospheric motion vector–derived flow
, 146
M. Łoboda (2009)
29
, 3
K. C. Wiens (2005)
4151
, 62
H. Lorenc (2005)
116
G. P. Pacey (2021)
237
, 36
B. Czernecki (2019)
249
, 227
W. Pilorz (2015)
45
, 4
D. M. Kingfield (2017)
5203
, 44
R. H. Johns (1992)
588
, 7
H. E. Brooks (2003)
73
, 67–68
M. J. Bunkers (2000)
61
, 15
K. D. Sherburn (2016)
1899
, 31
R. A. Houze (2018)
100 years of research on mesoscale convective systems
P. Groenemeijer (2017)
2641
, 98
Nautical dawn
P. Novák (2011)
318
, 100
R. L. Holle (2016)
2855
, 144
R. E. Bürgesser (2017)
2809
, 143
N. Pilguj (2019)
145
, 218
S. F. Blair (2017)
1101
, 32
C. Price (2008)
157
, 8
B. Geerts (2017)
767
, 98
H. J. Punge (2017)
34
, 198
A. Widawski (2018)
10
, 6
E. Tudurí (1997)
294
, 12
S. M. Verbout (2006)
86
, 21
A. K. Anderson-Frey, Y. P. Richardson, A. R. Dean, R. L. Thompson, B. T. Smith (2019)
Characteristics of tornado events and warnings in the southeastern United States
, 34
D. Celiński-Mysław (2020)
104944
, 240
A. H. Perez (1997)
428
, 12
D. R. Bright (2004)
A physically based parameter for lightning prediction and its calibration in ensemble forecasts
J. T. Allen, M. K. Tippett (2015)
The characteristics of United States hail reports: 1955–2014
, 10
M. E. Gropp (2018)
1045
, 33
T. N. Carlson (1968)
203
, 20
P. Bodzak (2006)
The ratio of cloud to cloud-to-ground flashes in summer seasons for Poland territory on the basis of data from PERUN system
A. T. Westermayer (2017)
59
, 26
B. Czernecki (2016)
118
, 167
Z. Bielec-Bąkowska (2003)
35
, 67–68
Z. Bielec-Bąkowska (2013)
99
, 132
D. W. Reif (2017)
1615
, 145
C. J. Schultz (2011)
744
, 26
A. K. Anderson-Frey (2019)
1017
, 34
B. T. Smith (2012)
1114
, 27
D. R. Poelman (2016)
607
, 16
J. T. Allen, D. J. Karoly, G. Mills (2011)
A severe thunderstorm climatology for Australia and associated thunderstorm environments
, 61
H. E. Brooks (2011)
538
, 100
V. A. Gensini (2011)
1
, 6
W. Deierling (2008)
D16210
, 113

Publisher: American Meteorological Society
Copyright: Copyright © American Meteorological Society
ISSN: 1520-0434
eISSN: 1520-0434
DOI: 10.1175/waf-d-21-0099.1
Publisher site: See Article on Publisher Site

Abstract

AUGUST 2022 PO R E ˛ BA ET A L . 1447 Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland a,b c,d a,b SZYMON PORE ˛ BA, MATEUSZ TASZAREK, AND ZBIGNIEW USTRNUL Department of Climatology, Jagiellonian University, Krakow, ´ Poland Institute of Meteorology and Water Management, National Research Institute, Warsaw, Poland Department of Meteorology and Climatology, Adam Mickiewicz University, Poznan, Poland National Severe Storms Laboratory, Norman, Oklahoma (Manuscript received 14 June 2021, in ﬁnal form 24 January 2022) ABSTRACT: The relationship between convective parameters derived from ERA5 and cloud-to-ground (CG) lightning ﬂashes from the PERUN network in Poland was evaluated. All ﬂashes detected between 2002 and 2019 were divided into intensity categories based on a peak 1-min CG lightning ﬂash rate and were collocated with proximal proﬁles from ERA5 to assess their climatological variability. Thunderstorms in Poland are the most frequent in July, between 1400 and 1600 UTC and over the southeastern parts of the country. The highest median of most unstable convective available potential energy (MUCAPE) for CG lightning ﬂash events is from June to August, between 1400 and 1600 UTC (around 900 J kg ), whereas patterns in 0–6-km wind shear [deep-layer shear (DLS)] are reversed, with the highest median values during winter and night (around 25 m s ). The best overlap of MUCAPE and DLS (MUWMAXSHEAR parameter) is 2 22 in July–August, typically between 1400 and 2000 UTC with median values of around 850 m s . Thunderstorms in Poland 21 21 are the most frequent in MUCAPE below 1000 J kg , and DLS between 8 and 15 m s . Along with increasing MUCAPE and DLS, peak CG lightning ﬂash rates increase as well. Compared to DLS, MUCAPE is a more important parameter in forecasting any lightning activity, but when these two are combined together (MUWMAXSHEAR) they are more reliable in distinguishing between thunderstorms producing small and high CG lightning ﬂash rates. Our results also indicate that higher CG lightning ﬂash rates result in thunderstorms more frequently associated with severe weather reports (hail, tornado, wind). SIGNIFICANCE STATEMENT: Each year severe thunderstorms produce considerable material losses and lead to deaths across central Europe; thus, a better understanding of local storm climatologies and their accompanying environ- ments is important for operational forecasters, emergency managers, and risk estimation. In this research we address this issue by analyzing 18 years of lightning intensity data and collocated atmospheric environments. Thunderstorms in Poland are the most frequent in July between 1400 and 1600 UTC and form typically in environments with low atmo- spheric instability and moderate vertical shear of the horizontal wind. The probability for storms producing intense lightning increases when both of these environmental parameters reach higher values. KEYWORDS: Deep convection; Lightning; Severe storms; Thunderstorms; Forecasting 1. Introduction events can be observed based on the records from the European Severe Weather Database (ESWD; Dotzek et al. A cloud-to-ground (CG) lightning ﬂash is on average regis- 2009), which also indicate that the most common convective tered on 150–160 days per year in Poland (Taszarek et al. threat across central Europe is severe wind, followed by exces- 2015). Thunderstorms are most frequent during the summer sive rainfall, large hail, damaging lightning, and tornadoes (i.e., June–August) and over the southeastern part of the (Groenemeijer et al. 2017; Taszarek et al. 2019). country (Bielec-Ba ˛kowska 2003). The same period is also Although thunderstorm intensity can be measured in vari- linked to the peak occurrence of severe weather outbreaks ous ways, it can be generally assumed that the presence of such as derechos and tornadic supercells that are responsible severe weather reports, high CG lightning ﬂash rates, and/or for considerable material losses in Poland (Celinski-Mysław high radar reﬂectivity depends largely on the vertical velocity and Matuszko 2014; Pilorz 2015; Widawski and Pilorz 2018; and size of the convective updraft (Apke et al. 2018). Pilguj et al. 2019; Pore Rba and Ustrnul 2020; Surowiecki and Regional research studies that focus on evaluating thunder- Taszarek 2020). A similar distribution of severe weather storm intensity and its relation to atmospheric environments are highly important to operational forecasters, as these stud- ies provide forecasters with guidance on the metrics and cor- Denotes content that is immediately available upon publica- responding values that are useful in predicting severe tion as open access. convective storms over speciﬁc areas, times of year, and times of day. For example, an increase in the availability of low-level moisture, midtropospheric lapse rates, and the Corresponding author: Szymon PorReba, szymon.poreba@doctoral. uj.edu.pl degree of convective organization (e.g., supercells, squall DOI: 10.1175/WAF-D-21-0099.1 Ó 2022 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses). 1448 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 lines) governed by sufﬁcient vertical wind shear leads to stronger updrafts (Doswell 2001; Markowski and Richardson 2010; Coffer and Parker 2015; Dennis and Kumjian 2017; Lin and Kumjian 2021). Despite growing numbers of studies on thunderstorm environments, there are still some limitations associated with challenging identiﬁcation and recognition of thunderstorm intensity (further described in section 2b). Thus, in this study we address this issue by using the peak 1-min CG lightning ﬂash rate as a proxy to categorize thun- derstorms according to their lightning intensity and then intercompare them with collocated atmospheric environ- ments. Although studies on thunderstorm environments in ´ ˇ central Europe were conducted in the past (Pucik et al. 2015; Westermayer et al. 2017; Taszarek et al. 2020), none of them evaluated speciﬁc CG lightning ﬂash rates, which in our approach constitutes a new, not previously applied in Europe, thunderstorm intensity proxy. Similar approaches have been tested for the tropics and subtropics (Liu et al. 2020). A pre- liminary comparison of CG lightning ﬂash rate intensity cate- FIG. 1. Mean number of ESWD severe weather reports per grid gories deﬁned in this study with 14 427 ESWD large hail, with speciﬁc thunderstorm intensity category. PERUN CG light- severe wind, heavy rain, and tornado reports from the terri- ning data and ESWD reports were collocated on 0.258 grids within tory of Poland indicates an increasing relative number of 1-h intervals. For each thunderstorm intensity category grids (deﬁned based on a peak 1-min CG lightning ﬂash rate), proximal reports along with increasing CG lightning ﬂash rates (Fig. 1). ESWD reports were counted and then divided by the total number By evaluating 18 years (2002–19) of CG lightning ﬂash rate of grids representing each category. data from the PERUN lightning detection network, we aim to better understand the climatological aspects of severe thun- derstorms and their accompanying environments across Lightning Detection Network (NLDN; Orville 1991; Orville Poland. An improved resolution of the ﬁfth generation of and Hufﬁnes 2001; Orville et al. 2002; Zajac and Rutledge ECMWF reanalysis (ERA5) allows us to study lightning envi- 2001; Cummins and Murphy 2009; Holle 2014; Holle et al. ronments in the way that was not possible with prior genera- 2016; Kingﬁeld et al. 2017; Koehler 2020). These studies show tions of reanalyses. We use hourly intervals (contrary to 3–6-h that the highest annual mean CG lightning ﬂash density steps in previous studies) to evaluate the diurnal, annual, and occurs over Florida, reaching 16 CG lightning ﬂashes km spatial aspects of convective environments collocated with yr and over the coast of Gulf of Mexico with 8–12 CG light- 22 21 speciﬁc CG lightning ﬂash rates}an aspect rarely explored ning ﬂashes km yr . In the same region the number of before. An auxiliary goal constituting the background of the thunderstorm days is the highest with more than 110 thunder- research is also to present an updated 18-yr climatology of storm days per year (Koehler 2020). CG lightning ﬂashes for Poland. In Europe, research on lighting climatologies has been con- ducted using the Advanced Technology Demonstration Network (ATDnet), European Cooperation for Lightning 2. Prior research Detection (EUCLID), and ZEUS networks (Anderson and a. Lightning climatologies Klugmann 2014; Poelman et al. 2016; Taszarek et al. 2019, Lightning climatologies are based on data from lightning 2020; Enno et al. 2020). These studies showed that peak light- detectors, which can be ground-based and satellite-based, ning ﬂash rate in Europe is in the afternoon (1400–1600 UTC) global or regional. Measurements of global lightning activity during summer. The highest annual lightning ﬂash rates are in 22 21 are applied with the ground-based lightning sensors systems, the Alps and over northeastern Italy reaching 7–8km yr such as World Wide Lightning Location Network (WWLLN; (Anderson and Klugmann 2014; Enno et al. 2020). The highest annual mean number of thunderstorm days is associated with http://wwlln.net) and Vaisala GLD360 (Said et al. 2011)or satellite-based instruments like Lightning Imaging Sensor mountainous regions and Mediterranean boundary layer mois- (LIS; Christian et al. 1999). Both sources conﬁrm that the ture, especially along the Alps, Dinaric Alps, and Apennines highest frequency of lightning occurs over continents in tropi- with values more than 60 (Taszarek et al. 2019). In regional cal areas (Virts et al. 2013; DiGangi et al. 2021). Global studies for selected countries, the results are consistent with ground-based systems make it possible to monitor lightning the previously mentioned distributions with mean annual peak 22 21 CG lightning ﬂash rates reaching 15 ﬂashes km yr over the over a broader area including higher latitudes, but their detec- border between Italy and Slovenia (Schulz et al. 2005), 9 ﬂashes tion efﬁciency is lower than regional or satellite-based systems 22 21 22 21 (Burgesser 2017; DiGangi et al. 2021). km yr in Italy (Feudale et al. 2013), 7.43 ﬂashes km yr ´ ´ Regionally, a large number of lightning climatologies have in the Czech Republic (Novak and Kyznarova 2011), 22 21 been performed for the United States with the National 3.06 ﬂashes km yr over the Carpathians in Romania AUGUST 2022 PO R E ˛ BA ET A L . 1449 22 21 (Antonescu and Burcea 2010), 1.01 ﬂashes km yr in large hail (Goodman et al. 1988; Macgorman et al. 1989; 22 21 Estonia (Enno 2011), and around 0.9 ﬂashes km yr in Williams et al. 1989, 1999; Wiens et al. 2005; Steiger et al. ¨ ¨ Finland (Makela et al. 2011). However, these numbers should 2007; Deierling and Petersen 2008; Gatlin and Goodman be interpreted with caution when making comparisons between 2010; Schultz et al. 2016). Moreover, severe thunderstorms regions as each lightning detection network features different are often associated with lightning jumps, which are good detection efﬁciency and spatial inhomogeneities (Price 2008). predecessors of severe weather (Perez et al. 1997; Schultz et al. In Poland, research on the temporal and spatial variability 2009, 2011, 2017; Farnell et al. 2017, 2018; Farnell and Rigo of thunderstorms was limited for many years to observations 2020). A prominent example might be strong, long-lived from manned surface synoptic stations (Stopa 1962; Grabowska supercell thunderstorms that feature powerful updrafts speeds 2001; Bielec-Ba˛kowska 2002, 2003, 2013; Kolendowicz 2006; and are well known for producing severe weather (Smith et al. Ustrnul and Czekierda 2009). These studies showed that 2012). These thunderstorms may have lightning ﬂash rates thunderstorms in Poland occur most often over southeastern exceeding 200 per minute (Lang et al. 2004; Markowski and Poland (30–35 thunderstorm days per year), and least fre- Richardson 2010). On the other hand, there are reports of tor- quently along the coast of the Baltic Sea (10–15 days). nadoes and severe convective wind gusts associated with However, such data have limitations (associated with human weaker and shallower convection, which rarely produce any perception) in thunderstorm identiﬁcation (Czernecki et al. lightning (Pacey et al. 2021). In central Europe, such events 2016; Koehler 2020; DiGangi et al. 2021), spatial biases, and typically occur in cold seasons and narrow cold-frontal rain- no method to determine storm intensity. The development of bands (NCFR; Gatzen 2011; Surowiecki and Taszarek 2020). the PERUN lightning detection system in the early twenty- For these events, the estimation of thunderstorm intensity ﬁrst century (Łoboda et al. 2009) has allowed for the moni- based on ﬂash rates may be not possible and represents a limi- toring of thunderstorms with a time lag of less than 1 min and tation of using lightning data for identiﬁcation of severe con- enhanced understanding of the spatial and temporal distribu- vective storms. tion of convective storms in Poland. An 11-yr period c. Convective parameters (2002–13) of CG lightning ﬂashes data from the PERUN was analyzed by Taszarek et al. (2015), who conﬁrmed that the Severe convective storms can be predicted and studied with largest number of thunderstorm days is over southeastern convective parameters that, when applied in the NWP mod- Poland (in accordance with estimates from manned surface els, may serve as a proxy of expected thunderstorm intensity. observations), but indicated a peak lightning ﬂash rate over Studies conﬁrmed that thunderstorm severity is driven mainly 22 21 eastern part reaching 3.1 ﬂashes km yr . Taszarek et al. by increasing convective instability and vertical wind shear, and (2015) also concluded that the highest fraction of nocturnal that these metrics can be used to discriminate between severe lightning is over western Poland. and nonsevere convection (Brooks et al. 2003; Thompson et al. 2012; Allen et al. 2011; Pu ´ci ˇ k et al. 2015; Taszarek et al. 2020). b. Thunderstorm intensity However, despite the growing usage of convective indices in The determination of thunderstorm intensity, which can be operational forecasting, they feature certain limitations. Some based on the evaluation of lightning frequency, radar reﬂec- indices, especially composite, merge meteorological elements tivity, or the occurrence of severe weather, is a considerable such as convective instability and shear without physical basis methodological challenge. One of the methods uses severe (Doswell and Schultz 2006). Calculations and especially the weather reports collected from weather observers and media scaling of parameters are in many ways arbitrary and vary sources (Dotzek et al. 2009; Edwards et al. 2013; Elmore et al. depending on the geographical location. Additionally, convec- 2014; Seimon et al. 2016; Krennert et al. 2018). However, the tive parameters demonstrate only possible atmospheric envi- determination of thunderstorm intensity based on these ronments, but do not predict if convection will be initiated. reports is quite arbitrary (Doswell 1985), and depends, inter Due to these factors, one should interpret the values of convec- alia, on the population density and local reporting efﬁciency tive indices with caution (Doswell and Schultz 2006). More- inducing spatial and temporal inhomogeneities (Doswell over, most convective indices were developed in the United 1985; Verbout et al. 2006; Allen and Tippett 2015; Blair et al. States, where convective environments feature higher mois- 2017; Groenemeijer et al. 2017; Edwards et al. 2018; Taszarek ture and convective instability, while European environ- et al. 2019). The ESWD database for the area of Poland con- ments are much drier with steeper lapse rates in the lower sists of more than 22 000 large hail, tornado, and severe wind troposphere (Brooks et al. 2007; Riemann-Campe et al. reports for the period 2008–20. 2009; Taszarek et al. 2020). Additionally, the obtained Teledetection data such as radar, lightning detectors, or sat- results will also depend on the choice of the parcel used for ellites provide more homogenous information and offer an CAPE calculations and application of the virtual temperature attractive alternative for the determination of thunderstorm correction (Doswell and Rasmussen 1994). A better agree- intensity (Punge et al. 2017; Enno et al. 2020; Fluck et al. ment between continents is observed for wind shear. Low- 2021). Well-organized thunderstorms with broad and strong level lapse rates, wind shear, and synoptic-scale lift [which updrafts have larger CG lightning ﬂash rates compared to are included in parameters such as SHERB (severe hazards ordinary cells with weaker updrafts, and large ﬂash rates are in environments with reduced buoyancy) or MOSH (modi- typically associated with the occurrence of tornadoes and ﬁed SHERB)] have overall better skill in discriminating 1450 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 between severe and nonsevere convection in low CAPE and high shear environments that are common in Europe (Hanstrum et al. 2002; Sherburn and Parker 2014; Sherburn ´ ´ et al. 2016; Celinski-Mysław et al. 2020; Rodrıguez and Bech 2021; Pacey et al. 2021). Many factors inﬂuencing thunderstorm development such as vertical wind proﬁle, moisture availability, convective mode, and the regional climatological aspects of atmospheric environments require extensive research aimed at assessing speciﬁc values of convective parameters (e.g., Gensini and Ashley 2011; Allen and Karoly 2014; Anderson-Frey et al. 2016; Chernokulsky et al. 2019; Ingrosso et al. 2020). Such studies demonstrate the pronounced utility of convective indi- ces in forecasting different types of thunderstorms. Thunder- storms in Europe occur most often in CAPE below 500 J kg , and higher convective instability is typically associ- ated with severe weather such as large hail and heavy precipi- tation. Given an unstable environment, wind shear is a good discriminator between nonsevere and extremely severe storms (∼20 m s can be used as a proxy for very large hail FIG. 2. Hypsometric map of Poland based on the Digital Terrain ´ ˇ and signiﬁcant tornadoes; Pucik et al. 2015; Taszarek et al. Model (USGS/SRTM; Farr et al. 2007). Points indicate the location 2020). However, one should be aware that it is difﬁcult to of the PERUN lightning detection sensors (red points indicate sen- compare results between studies using different reanalysis sors added to the PERUN network after 2014). datasets, as each of them represents a different underlying cli- matology of convective environments, and the values of cer- tain parameters may change along with changing resolution. coverage, with the lowest detection accuracy over northwest- Although many of the previously mentioned studies ern Poland (Taszarek et al. 2015). As new sensors have been focused on comparing severe wind, large hail, heavy rainfall, added to the network, the detection accuracy has increased and tornadoes with accompanying ambient environments, since 2014, especially in the southwestern part of the country. only few evaluated different classes of CG lightning ﬂash Each CG lightning ﬂash detection in the database included rates, which is the main focus of this work. Although CG the time, location, and a peak current of the discharge. To lightning ﬂash rates are not typically predicted by operational provide better homogeneity of the data we took into account forecasters, a correlation between speciﬁc CG lightning ﬂash only situations when a ﬂash had a peak current of at least rates and the relative number of ESWD reports can be 15 kA, as lower values are typically linked to intracloud detec- observed (Fig. 1). tions (Wacker and Orville 1999; Cummins and Murphy 2009; Koehler 2020). After this ﬁlter, a total of 8 251 273 CG light- 3. Data and methods ning ﬂashes (limited to the area of Poland) were used in this study. a. Lightning data The total number of CG lightning ﬂashes within a speciﬁc In this study we used all the available CG lightning ﬂashes thunderstorm depends not only on the evolution of updraft detected by the PERUN system on the territory of Poland intensity, but also on the duration and spatial coverage of the between 2002 and 2019 (18 years). PERUN is a Polish light- entire convective system. Thus, the total number of CG light- ning detection network working operationally since 2002. Its ning ﬂashes over a broader area (e.g., associated with meso- name corresponds to the Slavic god of lightning and thunder scale convective systems; Houze 2018) may not represent (Gieysztor 2006). At the beginning, the system consisted of 9 thunderstorm intensity well in comparison to local severe con- sensors, which after 2014 were expanded to 12 (Fig. 2). At the vective cells with a strong updraft (e.g., supercells), which can same time, the sensors began upgrading from SAFIR3000 produce a comparable number of CG lightning ﬂashes but (Surveillance et Alerte Foudre par Interferometrie Radio- over a much smaller area (Wiens et al. 2005; Steiger et al. electrique) to TLS200 (Total Lightning Sensor). As of 2021, 2007; Calhoun et al. 2013). Thus, to avoid biases associated the PERUN system consists of 8 TLS200 and 4 SAFIR3000 with increasing the thunderstorm area but not intensity, we sensors and is supported by the data from sensors in the decided to use a peak of 1-min ﬂash rate per 100 km . neighboring countries. CG lightning ﬂash detection efﬁciency To obtain information about the CG lightning ﬂash rates, estimated in 2006 reached 95% within the area of Poland all detections were gridded to 10 km 3 10 km boxes}a metric (Bodzak 2006; Bodzak et al. 2006). Since the sensors are not commonly used in prior work (e.g., Biron 2009; Enno 2011; distributed evenly throughout the country, only 38.3% of the Sulik 2021). Then, within these boxes, we investigated the country’s coverage has lightning location accuracy ﬁner than peak 1-min ﬂash count. In this way, a rate of CG lightning 1 km. Accuracy ﬁner than 2 km is found for 76.6% of the ﬂashes per 100 km per minute was obtained, and then AUGUST 2022 PO R E ˛ BA ET A L . 1451 TABLE 1. Thunderstorm intensity classes determined on the TABLE 3. Number of unique ERA5 grids in a given basis of 1-min peak CG lightning ﬂashes per 10 km 3 10 km thunderstorm CG lightning ﬂash intensity category over each grid (considered in hour steps) with corresponding percentile month. values, number of collocated ERA5 grids and a mean annual Weak Moderate Intense Extreme number of days given a speciﬁc category. January 493 120 17 3 Weak Moderate Intense Extreme February 351 64 3 4 Peak 1-min CG 12–45–9 .9 March 1669 763 116 25 lightning ﬂash April 11 278 5266 877 266 rate per 100 km May 45 810 32 126 7853 2714 Percentile of entire 49% 50%–84% 85%–94% 95%–100% June 52 946 40 156 12 402 7344 dataset July 73 392 52 917 16 589 8842 Number of ERA5 240 095 173 781 50 646 26 091 August 47 878 33 430 10 672 6066 grids September 12 911 7674 1923 751 Annual mean 161 126 124 55 October 2031 937 162 64 number of days November 635 190 24 7 December 283 138 8 5 subsequently used to determine thunderstorm lightning inten- sity (Table 1). Intensity classes of weak, moderate, intense, For each ERA5 grid, we assigned a thunderstorm category and extreme thunderstorms were deﬁned based on the 50th, based on the highest peak 1-min CG lightning ﬂash rate 75th, 90th, and 95th percentile thresholds of CG lightning within a speciﬁc hour (to match the ERA5 temporal step). ﬂash rate distribution. A limitation of this approach is that That procedure allowed us to merge two datasets with differ- ent resolutions. Weak thunderstorms contributed to the larg- due to the gridding process the thunderstorm intensity data est number of ERA5 grids, while those of extreme intensity refer to the peak ﬂash rates, which is not always associated were the least frequent, representing an annual mean of with one thunderstorm (several thunderstorms could have 161 and 55 days, respectively (Table 1). With the increase in occurred in the temporal and spatial scope of the reanalysis). the thunderstorm intensity category, the number of grids Nocturnal thunderstorms were classiﬁed based on the crite- decreases, especially in winter}in the case of extreme thun- rion of the solar angle below 2128 (computed for the location derstorms in January, only three grids were used (Table 3). and time of CG lightning ﬂash occurrence). This metric is For each reanalysis proﬁle, temperature, humidity, pres- known as the nautical dawn (NOAA/NWS Glossary 2021). sure, geopotential, U and V were interpolated vertically. b. ERA5 reanalysis A mixed-layer (ML) parcel was deﬁned by mixing a layer 0–500 m above ground level (AGL), while a most unstable To investigate thunderstorm environments, we used the ﬁfth (MU) parcel was based on the highest equivalent potential generation of ECMWF atmospheric reanalysis (ERA5; temperature (Q ) in the 0–3 km AGL; both versions used the Hersbach et al. 2020). ERA5 has a horizontal grid spacing of virtual temperature correction (Doswell and Rasmussen 0.258, a 1-h temporal step, and 137 terrain-following hybrid- 1994). For the computations of 0–3 km AGL storm-relative sigma model levels (Table 2), which made it possible to explore helicity (SRH03), the internal dynamics method was applied convective environments and their corresponding climatologies for right-moving supercells (Bunkers et al. 2000). Vertical in a way that was not possible with prior reanalyses, especially wind shear was computed as a magnitude of vector difference considering diurnal cycles. The 1-h time step allowed us to assign between the surface and a speciﬁc height. The indices used in lighting ﬂash events to speciﬁc convective environments with the study are associated with the concept of the ingredient- higher precision compared to previous reanalyses. In this case, based forecasting (Johns and Doswell 1992; Doswell et al. the maximum time difference between the CG lightning ﬂash 1996) focusing on the assessment of humidity characteristics, and the corresponding ERA5 grid was less than 30 min instead convective instability, and a vertical proﬁle of the wind, of 3 h for reanalyses consisting of 6-h steps (e.g., ERA-Interim). which governs the convective mode and storm organization For each grid from the previously described CG lightning (Weisman and Klemp 1982; Thompson et al. 2012). In addi- ﬂash rate database, we assigned a proximal grid from ERA5. tion, we also used composite indices, which turned out to be useful in assessing thunderstorm intensity across central TABLE 2. Characteristics of ERA5 reanalysis domain used in Europe: WMAXSHEAR (a square root of 2 times CAPE this study. multiplied by 0–6-km wind shear; Taszarek et al. 2020)dedi- cated to forecasting severe storms, SCP (supercell composite Horizontal grid spacing 0.258 3 0.258 Vertical levels (sigma) 137 hybrid-sigma levels parameter; Thompson et al. 2003) with an updated formula Temporal resolution Hourly from Gropp and Davenport (2018), and HSI (hail size index; Timeframe 2002–19 Czernecki et al. 2019) dedicated for forecasting large hail. As Total unique grid points 500 195 supercells and large hail are typically associated with strong Latitude extent 48.758–55.258N updrafts that also favor intense lightning, we believe these Longitude extent 13.758–24.508E metrics are worth comparing with other convective parameters 1452 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 and among CG lightning ﬂash ratecategories. A completelist TABLE 4. List of convective parameters used in this study (see the appendix for additional metrics). of variables used in this study is presented in Table 4 with additional metrics provided in the appendix. Parameter Abbreviation Unit c. ESWD reports 21 Most unstable (0–3 km AGL) MUCAPE J kg convective available potential A total of 14 427 ESWD reports from the area of Poland energy were used for comparison with lightning intensity categories Most unstable (0–3 km AGL) MUCIN J kg (Fig. 1). Only ESWD reports with Q1 and Q2 quality level clas- convective inhibition ses (meaning reports from reliable sources or conﬁrmed by sci- Most unstable (0–3 km AGL) MUMIXR g kg entiﬁcstudy; Dotzek et al. 2009), time lag of less than 1 h, and mixing ratio spatial proximity of 0.258 (from the detected speciﬁcCG light- 800–500-hPa temperature lapse LR85 K km rate ning density) were considered. For the higher lightning intensity 0–1 km AGL bulk wind shear LLS m s categories, the number of available cases decreases. We calcu- 0–6 km AGL bulk wind shear DLS m s late a mean number of ESWD reports that are within the afore- 2 22 0–3 km AGL storm-relative SRH03 m s mentioned proximity of the grid with speciﬁc CG lightning ﬂash helicity (right-moving rate. In the end, we divide the number of ESWD reports supercells) assigned to a speciﬁc lightning intensity category by the number 2 22 Product of MUCAPE and DLS MUWMAXSHEAR m s of grids for this speciﬁc category. (Taszarek et al. 2020) Supercell composite parameter SCP } (Thompson et al. 2003; Gropp 4. Results and Davenport 2018) a. Climatological aspects of CG lightning flashes Hail size index (Czernecki et al. HSI } in Poland 2019) Over the period 2002–19, we identiﬁed a total number of 2926 days with at least two CG lightning ﬂashes, which repre- probability for CG lightning ﬂashes around 1400–1500 UTC, and sented a mean of 161 thunderstorm days per year. The highest the lowest around 0700–0800 UTC. This pattern is evidently visi- annual mean number of hours with CG lightning ﬂashes ble during the summer and spring, whereas from October to occurred over southeastern and central Poland, reaching March it is less clear (Figs. 4b and 5a). Thespringfeatures the around 80 h (Fig. 3a), consistent with results from Earth highest fraction (12%) of CG lightning ﬂashes occurring in the Networks Global Lightning Detection Network (ENGLN; afternoon (1200–1500 UTC), while during the summer the peak DiGangi et al. 2021). The regions with the most frequent CG fraction occurs in the same hours, but with lower values of lightning ﬂashes included mainly higher elevation areas such around 8%–9%. At night, CG lightning ﬂashes are considerably as the Carpathian Mountains, the Lublin Upland, and the less frequent, with the lowest fractions in spring. During winter Krakow–Cze Rstochowa Upland, as well as the lowland areas and autumn, the fraction of nocturnal CG lightning ﬂashes is across central Poland (geographical regions of Poland are slightly higher than during summer and spring. provided in Fig. 2). Conversely, the lowest frequency of thun- Considering the interannual variability (Fig. 4c), the highest derstorm hours occurred across northwestern Poland, with an overall number of CG lightning ﬂashes occur every year dur- annual mean of less than 20 h. The highest mean annual CG ing the summer (in 2017 almost 1 000 000) and the lowest in 22 21 lightning ﬂash rate reaching 3.0 CG km yr was recorded the winter (in 2008 only 10). Although the spring typically has over the central part of the country and in the region of the a much higher number of CG lightning ﬂashes compared to Krakow–Cze Rstochowa Upland (Fig. 3b). the autumn, in some years, the number of detections can be The annual cycle of CG lightning ﬂashes in Poland is con- comparable (e.g., in 2006, the autumn had a higher number of sistent with typical climatological distributions of temperature detections compared to spring). Slight increases in the num- and precipitation in this part of Europe (Lorenc 2005). Thun- ber of annual CG lightning ﬂashes are probably due to derstorms are the most frequent in mid-July, but the time- improvements in the PERUN system over time. Strong frame with an enhanced mean number of CG lightning ﬂashes year-to-year variability is also observed for the annual frac- starts in mid-April with more than 7000, and ends in late tion of nocturnal CG lightning ﬂashes with values from 7% in September with 14 700 (Figs. 4a and 5a). During that period, 2006 to as much as 22% in 2009 (Fig. 4d). However, no signiﬁ- the annual mean number of CG lightning ﬂashes is greatest, cant long-term trend can be deﬁned. especially in July, reaching 155 474 (Fig. 4a). June has a The highest daily numberofCG lightning ﬂashes was slightly higher number compared to August with around recorded on 10 August 2017, reaching 141 628 (Table 5), which 15 000 more ﬂashes. Thunderstorms are clearly less frequent is 57 981 more detections than the second ranked 28 June during the autumn, winter, and early spring. From October to 2017. On 10 August 2017, the coverage of thunderstorms was March, the mean number of CG lightning ﬂashes typically also the largest of all analyzed cases and reached 55 791 km does not exceed 1300 (Fig. 4a). Considering the fractional dis- (17.8% of the area of Poland). Considering other days, the tribution of CG lightning ﬂashes in each hour, convection in area covered by grids with CG lightning ﬂashes ranged typi- Poland has a well-deﬁned diurnal cycle with the highest cally from 10.7% to 12.3% of the area of Poland (Table 5). AUGUST 2022 PO R E ˛ BA ET A L . 1453 FIG.3. (a) Mean annual number of hours with at least one CG lightning ﬂash, and (b) mean annual number of CG lightning ﬂashes per 1 km . Calculations are based on data from the PERUN lightning detection system for 2002–19 (10 km 3 10 km grid). Spatial distribution of median of (c) MUCAPE, (d) MUCIN, (e) DLS, and (f) 90th percentile of MUWMAXSHEAR. Convective parameters are derived from ERA5 reanalysis for a period 2002–19 (0.258 3 0.258 grid). Please note that only situations with detected CG lightning ﬂashes are considered. 1454 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 FIG. 4. (a) Annual mean number of CG lightning ﬂashes per month, (b) hourly fraction of CG lightning ﬂashes per meteorological season, (c) year-to-year variability of CG lightning ﬂashes per meteorological season, and (d) fraction of nocturnal CG lightning ﬂashes over the years. Calculations were based on the data from the PERUN lightning detection system for the years 2002–19. Please note the logarithmic scale in (c). Nocturnal lightning in (d) is deter- mined based on the solar angle (below 2128). Among the 10 days with the highest daily number of CG light- The components of atmospheric convective instability: ning ﬂashes, four cases occurred both in June and July, with MUMIXR (mixing ratio of the most unstable parcel) and two in August. The most intense thunderstorms in terms of LR85 (800–500-hPa temperature lapse rate) are characterized 1-min peak CG lightning ﬂash rate occurred on 21 June 2013, by differing patterns in temporal distributions. Peak values of 18 June 2013, and 10 August 2017, reaching 108, 80, and MUMIXR reach 12 g kg in July and early August, typically 2 21 74 CG lightning ﬂashes per 100 km per minute, respectively around 1800 UTC (Fig. 5b), while LR85 of 7 K km is (Table 6). observed between 1000 and 1600 UTC during spring from March to May (Fig. 5c). MUMIXR has a better relationship b. Annual and diurnal variability of CG lightning flash with the diurnal cycle of CG lightning ﬂashes but is delayed environments by about 3 h. The best overlap of MUMIXR and LR85 is in In this section we present the climatological temporal and late July and early August between 1200 and 1700 UTC, with spatial distribution of ERA5 convective environments for the resulting median most unstable convective available thunderstorms in Poland. The statistics presented in this sec- potential energy (MUCAPE) reaching around 850 J kg tion refer only to situations when a CG lightning ﬂash was (Fig. 5d). Conversely, when no insolation is available during detected, leading to a smaller sample size during winter and the night, MUCAPE has its minimum. Only for the period larger in summer (Table 3). Thus, results should be inter- from early July to late August does nocturnal MUCAPE have preted with caution as nonelectriﬁed convection is not a median of around 400 J kg . During the spring and considered. autumn, peak values of MUCAPE occur around noon, driven AUGUST 2022 PO R E ˛ BA ET A L . 1455 FIG. 5. (a) Annual and daily variability of grids with at least one CG lightning ﬂash, (b) median of MUMIXR, (c) LR85, (d) MUCAPE, (e) DLS, and (f) 90th percentile of MUWMAXSHEAR. Convective parameters are derived from ERA5 reanalysis for a period 2002–19. Vertical axis has hourly resolution while horizontal weekly. Labels indi- cate the ﬁrst week of each month. Please note that only situations with detected CG lightning ﬂashes are considered. by the surface heating and development of steep lapse rates. In comparison to MUCAPE, the distribution of deep-layer Climatological patterns in MUCAPE are well correlated with shear (DLS) features a reversed pattern (Fig. 5e). From April peak CG lightning ﬂash activity and have a well-deﬁned diur- to October, the median of DLS generally does not exceed nal cycle during the summer in opposition to a poor diurnal 10 m s , with the lowest values in the afternoon hours cycle during winter when thunderstorms in Poland are typi- (1200–1600 UTC). However, shear notably increases during cally rare (Figs. 5a,d). nighttime hours by 5–7m s , which is particularly visible in 1456 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 TABLE 5. Top 10 days with the highest number of CG lightning ﬂashes (left), and the highest spatial coverage of 10 km 3 10 km grids with at least one detection (right). Top 10 days with the highest number of Top 10 days with the highest spatial coverage of grids CG lightning flashes with at least one CG lightning flash Area with flashes (km ) Date No. of flashes Date (percent of country coverage) 10 Aug 2017 141 628 10 Aug 2017 55 791 (17.8%) 28 Jun 2017 83 647 25 Jul 2015 38 596 (12.3%) 21 Jun 2013 78 116 03 Jul 2012 38 014 (12.2%) 9 Aug 2013 74 625 15 Aug 2008 36 668 (11.7%) 26 Jun 2006 74 153 26 Jun 2006 36 482 (11.7%) 25 Jul 2015 72 361 19 Jul 2015 35 977 (11.5%) 13 Jun 2019 66 454 13 Jun 2019 34 520 (11.0%) 3 Jul 2012 64 938 28 Jun 2017 34 239 (11.0%) 15 Aug 2008 64 233 31 Jul 2005 33 434 (10.7%) 19 Jul 2015 63 967 29 Jun 2017 33 339 (10.7%) the summer months. In the period with lesser convective Taszarek 2020). Noteworthy also are the increased values of 2 22 occurrence from October to April, DLS has a typically less MUWMAXSHEAR (around 500 m s of 90th percentile) deﬁned diurnal cycle, but much higher overall values during spring and autumn, which are mainly driven by large (a median of 20–30 m s ). DLS and rather marginal MUCAPE. These conditions are so- A combination of thermodynamic convective instability and called high-shear low-CAPE (HSLC) environments, and are vertical wind shear is a useful environmental proxy for assessing often associated with an enhanced potential of squall lines and ´ ˇ thunderstorm severity (Brooks et al. 2003; Pucik et al. 2015; low-topped supercells capable of producing damaging winds Taszarek et al. 2020). MUWMAXSHEAR, which combines- and tornadoes (Sherburn and Parker 2014; Sherburn et al. 2016; both MUCAPE and DLS, indicates peak severe thunderstorm Anderson-Frey et al. 2019; Mathias et al. 2019; Gatzen et al. potential from June to August, with the 90th percentile exceed- 2020). 2 22 ing 800 m s during this period (Fig. 5f). The highest values c. Spatial variability of CG lightning flash environments typically occur between 1600 and 2000 UTC in late July and early August. Although DLS is relatively low in July (median Spatially, median MUCAPE (only for CG lightning ﬂash 10–14 m s ), the highMUCAPE inthat periodisthe main events) increases from the northwest toward the southeast driver of the overall MUWMAXSHEAR value. A similar (Fig. 3c), and thus the lowest values are along the Baltic Sea effect occurs in the diurnal cycle. Although MUCAPE drops coast (450 J kg ) and the highest in the Bieszczady Moun- during nighttime, MUWMAXSHEAR is still high, as it is com- tains over the southeast (700 J kg ). This pattern is consis- pensated by increasing DLS. This reﬂects an enhanced noctur- tent with the frequency of thunderstorm days based on nal potential for severe thunderstorms, that as a result of SYNOP reports (Bielec-Ba ˛kowska 2003; Czernecki et al. increasing shear, often evolve into large and well-organized 2016). mesoscale convective systems (MCS) with the potential to pro- The median of the MUCIN (convective inhibition) has a duce severe wind and heavy rain (Geerts et al. 2017; Reif and spatial pattern reversed to MUCAPE with the strongest Bluestein 2017; Haberlie and Ashley 2019; Surowiecki and inhibition over the western and northwestern parts of the country reaching 230 J kg and weakest over the south (210 J kg ), indicating the area with the lowest resistance to TABLE 6. Time and location of top 10 grid boxes (10 km 3 10 km) convection initiation (Fig. 3d). We hypothesize that the with the highest CG lightning ﬂash rates. enhanced inhibition occurring along the coast of the Baltic Sea is associated with low sea surface temperatures that lead Peak 1-min CG to temperature inversions in typically marginal MUCAPE sit- lightning flash rate 8 8 Date and time Lon ( E) Lat ( N) per 100 km uations. The regions with the strongest inhibition over west- ern Poland experience the lowest average annual rainfall in 1627 UTC 21 Jun 2013 21.000 50.732 108 Poland (Lorenc 2005), which also seems to be consistent with 1312 UTC 21 Jun 2013 20.409 52.619 88 the reduced convective frequency over that area (Fig. 3a). 1545 UTC 18 Jun 2013 21.708 50.730 80 0349 UTC 10 Aug 2017 18.083 52.044 74 Similarly to MUCIN, the spatial patterns in DLS are also 2118 UTC 12 Aug 2019 21.427 50.911 72 reversed with respect to MUCAPE, with the lowest values 2144 UTC 12 Aug 2019 21.855 50.998 70 over southeastern Poland (∼10 m s ) and the highest over 0545 UTC 10 Aug 2017 19.352 53.238 70 the northwest (∼14 m s ; Fig. 3e). This pattern indicates 1628 UTC 21 Jun 2013 21.000 50.732 70 more frequent HSLC environments over northwestern 1601 UTC 18 Jun 2013 21.849 50.639 70 Poland and in turn more common low-shear, high-CAPE 0219 UTC 14 Jul 2016 22.693 50.540 67 (LSHC) situations over the southeast. However, noteworthy AUGUST 2022 PO R E ˛ BA ET A L . 1457 FIG. 6. Number of situations with speciﬁc MUCAPE and DLS values for weak, moderate, intense, and extreme thunderstorms (divided based on a peak 1-min CG lightning ﬂash rate, as in Table 1). Constant values of MUWMAX- SHEAR (a product of MUCAPE and DLS; Taszarek et al. 2020) are indicated by dashed lines. White points indicate the median value for each category. Convective parameters are derived from ERA5 reanalysis for a period 2002–19. Please note that only situations with detected CG lightning ﬂashes are considered. is the northeastern part of the country, where a combination slightly differing thermodynamic convective instability (a dif- 21 21 of climatologically enhanced MUCAPE (∼600 J kg ) and ference of 129 J kg in the medians of MUCAPE). One can DLS (∼13 m s ) indicates increased conditional potential also see a dependence that thunderstorms developing in an for severe thunderstorms in this region, despite their limited environment with increased MUCAPE require lower DLS frequency compared to the southeast (Fig. 3a). This is con- and conversely, a higher DLS is typically accompanied by ﬁrmed by the spatial distribution of the 90th percentile of weaker convective instability. This pattern is well depicted by MUWMAXSHEAR, which in this region reaches even the constant values of MUWMAXSHEAR on the scatter- 2 22 850 m s } the highest values among the entire country (Fig. 3f). plots (dashed lines on Fig. 6). Thunderstorms with both high CAPE and high DLS are very rare in Poland. For intense and d. Variability of convective environments in relation to extreme thunderstorms (Figs. 6c,d), a shift in the number of CG lightning flash rates cases toward higher MUCAPE and a gradual increase in The two-dimensional distributions of MUCAPE and DLS DLS can be recognized. In the extreme thunderstorm for different CG lightning ﬂash rate intensity categories category (Fig. 6d), a median of 834 J kg MUCAPE and (weak, moderate, intense, extreme; Table 1, Fig. 6) indicate 12.7 m s DLS is notably higher compared to other classes. that the conditional probability for larger CG lightning ﬂash However, the largest increase of 153 J kg in the medians of rates increases as convective instability and vertical wind MUCAPE is between the moderate and intense thunder- shear increase. The median for MUWMAXSHEAR increases storm categories. 2 22 for each category with 330, 375, 441, and 519 m s , respec- In summary, although the probability for higher CG light- tively (Fig. 6). This result is broadly consistent with Liu et al. ning ﬂash rates increases with both DLS and MUCAPE, the (2020). increase in the latter is generally more important. A median 2 22 Both weak (Fig. 6a) and moderate (Fig. 6b) thunderstorm MUWMAXSHEAR of 441 m s for the intense thunder- 2 22 categories feature similar kinematic environments but a storm category is similar to the value of 450 m s from 1458 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 FIG. 7. Heidke skill score for MUCAPE, DLS, and MUWMAXSHEAR calculated by comparing the weak thunder- storm intensity category with at least moderate (orange), at least intense (red), and extreme (magenta). Taszarek et al. (2019), which was considered as a discrimina- of HSS indicate that predicting CG lightning ﬂash rate tor between severe and nonsevere thunderstorms across cen- with convective parameters is a very challenging task, and tral Europe (conditional on successful convective initiation). its application in operational forecasting is limited. How- On the other hand, a different pattern of DLS and CAPE ever, forecasting any type of convective hazards across dependency was found in Europe for the relative frequency of Europe is a difﬁcult task (Brooks et al. 2011). lightning (Westermayer et al. 2017). High relative frequency The increase of MUMIXR and MUCAPE corresponds 21 21 was found in low (,10 m s )and high (.20 m s )DLS with increasing conditional probability for a higher CG light- environments, while CAPE values below approximately ning ﬂash rate category. The annual and daily distributions of 200–400 J kg demonstrated low relative frequency. This sug- these indices are similar (Figs. 8a,c), reaching the highest val- 21 21 21 gests that both enhanced CAPE and DLS are necessary for ues (medians from 10 g kg and 500 J kg to 12 g kg and storms producing high CG lightning ﬂash rates, but only a 1000 J kg for weak to extreme thunderstorms) in summer small amount of CAPE is needed for occurrence of any light- (June–August), except for the 90th percentile of MUCAPE ning (given successful convective initiation). However, as for extreme thunderstorm category, which is notably highest thunderstorms with high CG lightning ﬂash rates may not (∼2700 J kg ) in June (compared to July for MUMIXR). always be associated with severe weather and vice versa, cau- Diurnally, for both indices, the highest values are in the tion needs to be taken when interpreting this result. afternoon (Figs. 8b,d). MUMIXR median oscillates around To provide additional statistical tests for MUCAPE, DLS 10–13 g kg during both day and night (Fig. 8b). The highest and MUWMAXSHEAR, we calculated the Heidke skill median in MUCAPE for weak and moderate thunderstorms score (HSS). Peak HSS values were found for MUCAPE of occurs between 1000 and 1400 UTC, for strong between 1200 21 21 1000 J kg (HSS of 0.13), DLS of 12.5 m s (HSS of 0.05), and 1400 UTC, and for extreme between 1400 and 1800 UTC. 2 22 and MUWMAXSHEAR of 600 m s (HSS of 0.16) for This pattern indicates that storms developing in a high extreme versus weak CG lightning ﬂash rate category MUCAPE environment during late afternoon are more likely (Fig. 7). A shift of peak HSS values among speciﬁcCG to become severe, which is possibly linked to increasing verti- lightning ﬂash rate categories was observed with increas- cal wind shear toward late evening and night. ing values of MUCAPE and MUWMAXSHEAR, while Analysis of MUCIN demonstrated that environments for for DLS, differences were marginal. The biggest differ- severe and extreme thunderstorms are associated not only with ences among categories and the highest HSS were higher CAPE and shear but also with stronger inhibition observed for MUWMAXSHEAR, which seems to better (Figs. 8e,f). The strongest inhibition (the most negative values discriminate among CG lightning ﬂash rate categories of MUCIN) occurs in the summer, but in contrast to compared to MUCAPE or DLS alone. Overall low values MUCAPE, value distributions are relatively even from June to AUGUST 2022 PO R E ˛ BA ET A L . 1459 FIG. 8. Box-and-whisker plots presenting the (left) annual and (right) diurnal distributions of thunderstorm intensity classes (weak, moderate, intense, extreme; as in Table 1) for (a),(b) MUMIXR, (c),(d) MUCAPE, and (e),(f) MUCIN. Convective parameters are derived from ERA5 reanalysis for a period 2002–19. Please note that only situa- tions with detected CG lightning ﬂashes are considered. The horizontal line inside the box demonstrates the median, the extent of the box is from the 25th to 75th percentile, and whiskers represent the 10th and 90th percentiles. 21 21 21 September (median for all categories from 25to 215 J kg ). 250 J kg in 5 J kg steps (Table 7). The results indicate that Diurnally, the weakest inhibition (highest values) occurs during for each thunderstorm category, probability for the occurrence strong surface heating between 1000 and 1600 UTC, while is highest for MUCIN values above 210 J kg (weaker inhibi- outside this period a median of MUCIN varies from 25to tion) and lowest below 240 J kg . While in higher MUCIN 220 J kg (Fig. 8f). Convective inhibition is typically stronger (weaker inhibition), conditional probability increases along in environments favorable for higher CG lightning ﬂash rates. with decreasing thunderstorm intensity, an opposite relation- This is conﬁrmed by calculating the conditional probability for ship can be observed for lower MUCIN (stronger inhibition). each thunderstorm category and MUCIN intervals from 0 to Delayed convection initiation during the day driven by 1460 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 TABLE 7. Conditional probability distribution for MUCIN intervals and thunderstorm category. MUCIN intervals (J kg ) [0; 210] (210; 220] (220; 230] (230; 240] (240; 250] Weak 0.6734 0.1637 0.0811 0.0493 0.0325 Moderate 0.6386 0.1760 0.0924 0.0551 0.0378 Intense 0.5887 0.1922 0.1096 0.0652 0.0442 Extreme 0.5441 0.2080 0.1181 0.0773 0.0525 stronger inhibition subsequently promotes stronger moisture lowest frequency of thunderstorms and lowest convective pooling, higher convective instability, better storm organization instability. It is also worth highlighting that, contrary to pre- (shear is higher toward evening hours; Fig. 9b), and a more viously analyzed variables (LR85, MUCAPE, MUCIN, sudden development of convective updrafts promoting higher and LLS), MUWMAXSHEAR has no clearly deﬁned CG lightning ﬂash rates. Moreover, environments favorable for diurnal cycle. This is because of the distribution of its extreme thunderstorms are often associated with advection of components}convective instability, which has a clear diur- very warm air with steep lapse rates and an elevated mixed nal cycle with the maximum in the afternoon, and shear, layer, a situation commonly referred to as a “loaded gun” pro- which has an inverted diurnal cycle with minimum values in ﬁle. These environments in Europe are often associated with the afternoon. The combination of these two results in a “Spanish plume” synoptic setups (Carlson and Ludlam 1968; ﬂattened diurnal distribution. VanDelden2001; Mathias et al. 2017). On the other hand, While we are aware that SCP and HSI are parameters stronger convective inhibition during extreme thunderstorms is designed to forecast supercells and large hail, the strength of also pronounced at night (medians around 220 J kg ), which the updraft in the supercells and conditions favorable for may correspond to well organized, elevated thunderstorms large hail formation are also supportive for lightning genera- (Colman 1990; Grant 1995). These thunderstorms initiate tion. Thus, we evaluate these parameters in our study as well above a stable layer without being inﬂuenced by strong mixed and note that their increasing values are consistent with more layer or surface-based convective inhibition. frequent lightning. The highest values of both SCP and HSI As previously discussed, months with the highest thermody- are observed between May and August. While values of SCP namic convective instability feature the lowest values of kine- during other months rarely exceed 0, the HSI timeframe with matic parameters (Fig. 9). DLS, 0–1-km wind shear [low-level elevated values is from April to October. A rapid increase in wind shear (LLS)], and SRH03 from April to September has a the HSI in April is maintained at a similar level until August 21 21 2 22 similar range of values (∼11 m s , ∼6m s ,and ∼70 m s , with the 90th percentile of every storm category exceeding a respectively) among thunderstorm intensity classes (Figs. 9a,c,e). value of 2 (Fig. 10e), which highlights an elevated risk of large Only for the extreme thunderstorms, this relationship is better hail in this period. For the SCP, 90th percentile for intense pronounced with overall slightly higher values of all three and extreme thunderstorms during the summer exceeds a metrics. The diurnal distributions of DLS, LLS, and SRH03 value of 2 (Fig. 10c), which is a good discriminator between (Figs. 9b,d,f) are characterized by the lowest values in the hours supercell and nonsupercell thunderstorms based on a calibra- with the highest thermodynamic convective instability (from tion derived from environments across the United States 1200 to 1600 UTC), which is especially evident for LLS that has (Thompson et al. 2003, 2007; Gropp and Davenport 2018). a stronger diurnal amplitude compared to DLS and SRH03. At This may lead to the conclusion that conditions supportive for night and in the morning, the extreme thunderstorm category supercell development in Poland are relatively uncommon features higher median values compared to other categories, and occur typically between May and August. Diurnally, SCP 21 21 2 22 reaching 17 m s ,7 m s ,and 140 m s for DLS, LLS, and has no clear peak, but its minima (with the 90th percentile SHR03, respectively. Low differences between thunderstorm below 0.7) can be explicitly depicted between 0000 and 0200 categories, demonstrates limited utility in forecasting speciﬁc UTC (Fig. 10d). In contrast, HSI has a clear diurnal cycle ﬂash rates using only kinematic parameters. (Fig. 10f) with its peak of ∼1.2 (medians) between 1200 and Composite indices combining multiple parameters into one 1400 UTC (hours with the highest lapse rates). For both product aimed at forecasting speciﬁc phenomena offers an parameters, higher values are characteristic in more intense attractive approach for forecasters, and thus we also evalu- thunderstorm categories, but for SCP, the difference between ated them in our study. MUWMAXSHEAR reaches the high- categories is even higher. The distribution of the 90th percen- 2 22 est median during summer months (300–600 m s ) with a 90th tile for HSI demonstrates rather similar values (3.0–3.4 during 2 22 percentile of extreme thunderstorms exceeding 1250 m s summer) for each thunderstorm category, while for SCP, (Fig. 10a). A notion of increasing MUWMAXSHEAR with extreme thunderstorms have clearly higher values compared increasing CG lightning ﬂash rates is valid from April to Octo- to other categories. Based on that, we can conclude that ber, while from November to March, such a dependence is not supercells in Poland can occur at any time of the aforemen- present. Diurnally (Fig. 10b), the biggest difference between the tioned period and that CG lightning ﬂash rates increase with weak and extreme category is between 0400 and 0800 UTC increasing chances for supercells (i.e., values of SCP). This (morning hours of local time in Poland)}the period with the result is unsurprising, as powerful updrafts in supercell AUGUST 2022 PO R E ˛ BA ET A L . 1461 FIG.9.As in Fig. 7, but for (a),(b) DLS, (c),(d) LLS, and (e),(f) SRH03. thunderstorms may reach vertical velocities of even 50 m s of this study. In addition to the parameters evaluated in this (Lehmiller et al. 2001) and drive generation of very large CG study, in the appendix, we provide median, 10th, and 90th per- lightning ﬂash rates. centiles for speciﬁc CG lightning ﬂash rate categories. In the distributions presented, considerable overlap occurred between thunderstorm categories. For convective instability e. Importance of relative humidity on and composite parameters, this overlap was less pronounced, convection initiation suggesting that CG lightning ﬂash rates are more sensitive to The last analyzed meteorological aspect is the relative convective instability parameters than shear. Improving the analysis by the relation of convective parameters with the radar humidity at the isobaric levels of 850, 700, and 500 hPa, and a data would probably reduce this overlap, but is out of the scope mean in the 0–4 km AGL layer (Fig. 11). Relative humidity, 1462 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 FIG.10. As in Fig. 7, but for (a),(b) MUWMAXSHEAR, (c),(d) SCP, and (e),(f) HSI. although it does not inﬂuence thunderstorm intensity, is a Among the previously mentioned pressure levels, the high- crucial factor for the development of any thunderstorms est values of the interquartile range and the median of relative (Westermayer et al. 2017). The lower the relative humidity is, humidity occurred at 850 hPa (approximately 1.5 km MSL; the stronger the dry air entrainment can be, which may Fig. 11b). At this level, 3/4 of all analyzed lightning cases had strongly limit the development of convective updrafts. On the a relative humidity between 70% and 85%. Equally high val- other hand, relatively dry mid tropospheric air but in environ- ues of the interquartile range occurred also in a mean 0–4-km ments with strong synoptic-scale forcing and sufﬁcient humid- relative humidity (Fig. 11a), stretching from 66% to 80%, ity at low levels, can result in releasing potential instability while at the level of 700 hPa (Fig. 11c) this spread varied from and successful convection development. This effect is demon- 65% to 83%. The relative humidity at the level of 500 hPa strated on a rather ﬂat distribution of relative humidity at 500 turned out to be less important as thunderstorms occurred hPa compared to lower levels. in a wide spectrum of values and the median was 56% AUGUST 2022 PO R E ˛ BA ET A L . 1463 FIG. 11. Histograms of relative humidity at 850-, 700-, and 500-hPa isobaric levels and a mean layer of 0–4km AGL for all CG lightning ﬂashes in Poland. Yellow lines indicate the interquartile range, i.e., position of the 25th percentile (Q1) and 75th percentile (Q3), and the red line indicates the median (Q2). Relative humidity values are derived from ERA5 reanalysis for a period 2002–19. (Fig. 11d). A mean 0–4-km relative humidity turned out to be lower resolution of prior reanalyses. We also evaluated if the most consistent in terms of value distribution, and had a convective parameters can serve as predictors of speciﬁc well-deﬁned peak at 75%. These results are consistent with CG lightning ﬂash rate intensity categories (an aspect Westermayer et al. (2017), who also highlighted the impor- rarely studied in prior research) and investigated their spa- tance of relative humidity with respect to convection initiation tiotemporal climatological variability. We proposed a new across Europe. That study concluded that the probability for deﬁnition for thunderstorm intensity categories by using a thunderstorm development drops signiﬁcantly when the rel- 1-min peak CG lightning ﬂash rates. The proposed method- ative humidity is below 50%, which is very similar to the dis- ology features some limitations: severe, shallow convection tributions provided in our work, but was for a longer research may have poor lightning activity, the choice of timeframe period. and gridtodeﬁne ﬂash rates is arbitrary, and intracloud ﬂashes, which are important in lightning jump calculations (Farnell et al. 2018; Farnell and Rigo 2020), were omitted 5. Concluding remarks in our study (due to low quality of this data). Synthesis of In this study we used over 8 million CG lightning ﬂashes PERUN lightning data with ERA5 reanalysis yielded detected across Poland over the course of 18 years (2002–19) numerous ﬁndings, among which the most important are and intercompared speciﬁc CG lightning ﬂash rates with listed below: accompanying convective environments from ERA5. The hourly steps of ERA5 and a large sample size of the PERUN 1) A mean of 161 thunderstorm days occurs each year in lightning database provided the opportunity to explore the Poland, most frequently over the southeast. However, the diurnal cycles of thunderstorm environments and their rela- highest frequency of CG lightning ﬂashes is in eastern 22 21 tion to CG lightning ﬂash rates, which was not possible with a Poland with up to 3 ﬂashes km yr . 1464 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 2) July is the month with the largest number of CG lightning differences can be found if Poland is compared to regions of the ﬂashes (around 150 000 per year), reaching peak fre- Mediterranean or northwestern Europe that are more heavily quency typically between 1400 and 1600 UTC. Nocturnal inﬂuenced by their proximity to the sea surface and cyclonic CG lightning ﬂashes share an annual fraction of 4%–24% ´ activity (Tudurı and Ramis 1997; Cohuet et al. 2011; Holley for the speciﬁc years. et al. 2014). Mediterranean storms also typically feature much 3) The greatest convective instability occurs from June to higher instability with a peak of the season shifted toward late August between 1400 and 1600 UTC, with the highest low- summer and autumn (Riemann-Campe et al. 2009). Comparing level moisture at 1800 UTC. Thunderstorms occurring from our results to previous research, we noted that while enhanced March to May feature the highest midtropospheric lapse CAPE and DLS are typically necessary for storms producing rates. Patterns in wind shear are reversed to MUCAPE and high CG lightning ﬂash rates, only a small amount of CAPE have the highest values during the winter and at night. The (∼200 J kg ) is needed for the occurrence of any lightning, most conducive conditions for convection initiation in given successful convective initiation (Westermayer et al. 2017). Poland are between 1000 and 1400 UTC as evidenced by We also found that median MUCAPE and MUWMAX- the weakest convective inhibition and steepest lapse rates. SHEAR values for the extreme lightning category during sum- 4) The best overlap of convective instability and wind shear is mer correspond to large hail environments in central Europe in July and August, typically between 1400 and 2000 UTC (Pu ´cik ˇ et al. 2015; Taszarek et al. 2020). On the other hand, a (based on the 90th percentile of MUWMAXSHEAR). median DLS of 12.5 m s for the extreme lightning category 2 22 However, values exceeding 500 m s can occur from was much lower than 20 m s typically required for signiﬁcant March to November and at any time of the day. tornadoes and very large hail. This indicates that while there is 5) Thunderstorms in Poland are the most frequent in MUCAPE some relationship between the relative number of ESWD 21 21 below 1000 J kg and DLS between 8 and 15 m s .When reports and CG lightning ﬂash rates, the use of lightning data greater convective instability is available, DLS is typically as a proxy for convective hazards features certain limitations. lower (warm-season thunderstorms). Conversely, in highly As we showed in this study, predicting CG lightning ﬂash rates sheared environments, convective instability is typically mar- with convective parameters is a very challenging task and its ginal (cold-season thunderstorms). Situations with both high application in operational forecasting is limited. However, it is MUCAPE and DLS are rare in Poland. LSHC environments worth highlighting that forecasting convective hazards such as are the most frequent over southeastern Poland, while HSLC large hail, tornadoes, and severe wind has always been a very are the most common over the northwest. challenging task for operational forecasters across Europe 6) Along with increasing MUCAPE and DLS, peak CG light- (Brooks et al. 2011). The use of CG lightning ﬂash rates for ning ﬂash rates increase as well. However, a combination of studying storm intensity offers a much larger sample size and these two (MUWMAXSHEAR) demonstrated the highest spatial homogeneity compared to severe weather reports that (but still limited) skill compared to MUCAPE or DLS alone. are biased by population density. With improvements in 7) The vast majority of thunderstorms had a low to midlevel ground-based lightning detection networks, we expect similar relative humidity higher than 60%. studies to be developed in future for other regions of the world, 8) Proposed thunderstorm intensity proxy based on CG thus contributing to better understanding of local thunderstorm lightning ﬂash rates demonstrated a correlation with the climatologies and their accompanying environments. number of ESWD severe weather reports. Acknowledgments. This research was funded by the Insti- The results listed above are broadly consistent with prior tute of Meteorology and Water Management - Polish studies concerning the climatological aspects of lightning Research Institute (project S7), the Priority Research Area (Anderson and Klugmann 2014; Poelman et al. 2016; Taszarek Anthropocene under the program “Excellence Initiative} et al. 2019; Enno et al. 2020), convective environments accom- Research University” at the Jagiellonian University in panying central European thunderstorms (Pu ´cik ˇ et al. 2015; Krakow, ´ and supported by a grant from the Polish National Taszarek et al. 2020; Walawender et al. 2017; Westermayer Science Centre (project 2020/39/D/ST10/00768). The reanal- et al. 2017), and the comparison of satellite observations of ysis computations were performed in Poznan Supercomput- thunderstorms with ERA-Interim environments over tropical ing and Networking Center (project 448). and subtropical regions (Liu et al. 2020). Given a complex European orography and coastline that lead to high variability in local thunderstorm climatologies, Data availability statement. ERA5 data (temperature, spe- focusing on smaller areas (e.g., area of Poland) can provide ciﬁc humidity, geopotential, pressure, U, and V) were down- more precise information with respect to what is happening loaded from the European Centre for Medium-Range on the regional scale. This is conﬁrmed by comparison to Weather Forecasts (ECMWF), Copernicus Climate Change Service (C3S) at Climate Data Store (https://cds.climate. prior research that showed maximum thunderstorm activity in copernicus.eu/). PERUN lightning data was provided by the Poland being shifted by more than a month compared to Polish Institute of Meteorology and Water Management - neighboring parts of Europe (Taszarek et al. 2020). Concern- ing spatiotemporal distributions of convective environments, National Research Institute and due to the proprietary nature the area of Poland seems to not differ substantially from the of the data, cannot be made openly available. Contact rafal. area of eastern and central Europe. However, signiﬁcant lewandowski@imgw.pl for usage information. AUGUST 2022 PO R E ˛ BA ET A L . 1465 TABLE A1. The 10th, 50th, and 90th percentiles for selected convective parameters associated with weak (wea), moderate (mod), intense (int), and extreme (ext) CG lightning ﬂash rate categories. EL = equilibrium level; MUEL = MU equilibrium level; WBZ = wet bulb zero; STP = signiﬁcant tornado parameter; CPTP = cloud physics thunder parameter. 10th percentile 50th percentile 90th percentile Wea Mod Int Ext Wea Mod Int Ext Wea Mod Int Ext MUCAPE (J kg ) 62 116 190 248 448 572 724 834 1334 1475 1611 1714 MLCAPE (J kg ) 0 1 5 5 223 298 396 455 925 1048 1180 1265 8 8 MUCAPE from 0 to 220 C(J kg ) 25 54 92 124 205 244 289 320 447 481 515 540 MUCIN (J kg ) 243 248 256 263 26 27 29 211 0000 MLCIN (J kg ) 284 299 2122 2152 212 216 222 229 0000 MULCL (m AGL) 438 430 432 442 975 975 1000 1050 1875 1850 1850 1925 MLLCL (m AGL) 450 450 470 470 875 875 875 875 1575 1550 1525 1500 MULFC (m AGL) 740 850 975 1075 1600 1642 1750 1850 2550 2575 2625 2725 MLLFC (m AGL) 0 270 390 340 1725 1800 1900 2025 2725 2775 2875 3000 MLEL (m AGL) 0 3900 2875 3150 7000 7900 8700 9140 10 700 11 000 11 263 11 400 MUEL (m AGL) 4550 5592 6700 7600 8500 9300 10 100 10 600 11 400 11 564 11 800 11 900 MLEL temperature (8C) 9.5 1.2 21.7 22.0 227.9 233.2 237.7 239.5 2521. 253.6 254.9 255.5 MUEL temperature (8C) 212.7 217.9 224.0 228.4 239.1 243.8 247.8 250.0 256.2 257.3 258.2 258.9 MUMIXR (g kg ) 7.0 7.8 8.7 9.3 10.2 10.8 11.5 12.0 13.3 13.6 14.1 14.4 MLMIXR (g kg ) 6.9 7.6 8.6 9.4 10.1 10.6 11.3 11.8 12.8 13.1 13.4 13.7 0–1-km lapse rate (K km ) 2.7 2.4 2.0 1.4 7.0 6.7 6.3 5.8 9.9 9.8 9.5 9.2 0–3-km lapse rate (K km ) 5.2 5.2 5.1 4.9 6.7 6.6 6.6 6.4 8.1 8.0 7.9 7.8 3–6-km lapse rate (K km ) 5.6 5.7 5.6 5.6 6.2 6.2 6.1 6.1 7.0 6.9 6.8 6.7 2-m dewpoint (8C) 9.0 10.5 12.3 13.6 14.7 15.5 16.4 17.2 18.5 18.9 19.2 19.5 2-m temperature (8C) 14.4 15.6 16.9 17.7 20.0 20.7 21.5 21.9 25.8 26.1 26.5 26.8 850-hPa dewpoint (8C) 1.6 3.2 5.2 6.6 8.1 8.9 10.0 10.8 12.1 12.5 13.0 13.6 850-hPa temperature (8C) 5.7 7.5 9.5 11.0 12.2 13.0 14.1 15.0 16.4 16.8 17.4 17.9 700-hPa dewpoint (8C) 29.9 27.8 25.9 24.8 22.8 22.2 21.3 20.5 1.4 1.7 2.2 2.8 700-hPa temperature (8C) 24.5 22.9 21.1 0.3 1.4 2.1 3.2 4.0 5.2 5.5 5.9 6.3 500-hPa dewpoint (8C) 233.3 231.3 228.8 226.7 222.4 221.4 219.9 218.6 214.9 214.6 214.1 213.5 500-hPa temperature (8C) 221.7 219.9 217.8 216.2 214.2 213.5 212.5 211.8 210.7 210.5 210.2 210.0 Height of WBZ (m AGL) 1850 2100 2375 2575 2800 2900 3100 3250 3400 3450 3550 3600 Effective shear (m s ) 2.41 2.8 3.4 3.9 7.19 7.7 8.5 9.6 14.4 14.9 15.9 17.0 0–1-km wind shear (m s ) 1.7 1.7 1.8 2.1 5.1 5.1 5.4 5.8 10.7 10.5 10.8 11.2 0–3-km wind shear (m s ) 3.1 3.3 3.6 4.2 8.2 8.5 9.1 10.1 11.1 15.8 16.3 16.9 0–6-km wind shear (m s _ 4.1 4.1 4.7 5.4 11.1 11.1 11.6 12.7 21.3 20.5 20.8 21.7 1–3-km mean wind (m s ) 3.3 3.2 3.4 3.8 8.2 8.2 8.4 8.8 14.7 14.3 14.4 14.6 2 22 0–1-km SRH (m s ) 22.5 20.8 0.6 2.6 33.4 35.3 38.5 45.0 117.2 118.9 124.5 139.8 2 22 0–3-km SRH (m s ) 15.7 17.7 20.9 26.9 65.8 69.1 75.7 86.1 167.2 172.5 184.1 204.3 STP (Thompson et al. 2007) 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.1 SCP (Thompson et al. 2007) 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.0 1.0 1.4 2.0 2.8 HSI (Czernecki et al. 2019) 0.0 0.0 0.0 0.1 0.6 0.7 0.8 0.9 1.7 1.8 1.9 2.1 CPTP (Bright et al. 2004) 0.0 0.0 0.0 1.5 18.9 33.9 53.5 68.2 123.7 138.5 153.0 163.2 Precipitable water (mm) 18.7 21.1 24.2 26.9 29.5 31.4 33.9 36.0 39.0 39.9 41.1 42.7 2 22 MUWMAXSHEAR (m s ) 81.2 108.5 142.9 183.2 292.1 331.1 398.4 479.9 676.1 730.3 822.6 915.7 2 22 MLWMAXSHEAR (m s ) 0.2 14.1 29.6 32.9 188.9 216.6 264.1 309.0 518.6 566.0 642.4 712.5 APPENDIX variability and ENSO inﬂuence. Int. J. Climatol., 34,81–97, https://doi.org/10.1002/joc.3667. Statistical Distribution of Convective Indices in }}, and M. K. Tippett, 2015: The characteristics of United States hail reports: 1955–2014. Electron. J. Severe Storms Thunderstorm Intensity Classes Meteor., 10 (3), 1–31. This appendix contains statistical distribution (10th, 50th, }}, D. J. Karoly, and G. Mills, 2011: A severe thunderstorm cli- and 90th percentiles) of all convective indices that were used matology for Australia and associated thunderstorm environ- in this study in function of thunderstorm intensity Table A1. ments. Aust. Meteor. Oceanogr. J., 61, 143–158, https://doi. org/10.22499/2.6103.001. Anderson, G., and D. Klugmann, 2014: A European lightning REFERENCES density analysis using 5 years of ATDnet data. Nat. Hazards Earth Syst. Sci., 14,815–829, https://doi.org/10.5194/nhess-14- Allen, J. T., and D. J. Karoly, 2014: A climatology of Australian 815-2014. severe thunderstorm environments 1979–2011: Inter-annual 1466 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 Anderson-Frey, A. K.,Y.P.Richardson, A. R. Dean,R.L. a new hodograph technique. Wea. Forecasting, 15,61–79, Thompson, and B. T. Smith, 2016: Investigation of near- https://doi.org/10.1175/1520-0434(2000)015,0061:PSMUAN. storm environments for tornado events and warnings. Wea. 2.0.CO;2. Forecasting, 31, 1771–1790, https://doi.org/10.1175/WAF-D- Burgesser, R. E., 2017: Assessment of the World Wide Lightning 16-0046.1. Location Network (WWLLN) detection efﬁciency by com- }}, }}, }}, }},and }}, 2019: Characteristics of tornado parison to the Lightning Imaging Sensor (LIS). Quart. J. events and warnings in the southeastern United States. Wea. Roy. Meteor. Soc., 143, 2809–2817, https://doi.org/10.1002/qj. Forecasting, 34, 1017–1034, https://doi.org/10.1175/WAF-D- 3129. 18-0211.1. Calhoun, K. M., D. R. MacGorman, C. L. Ziegler, and M. I. Big- Antonescu, B., and S. Burcea, 2010: A cloud-to-ground lightning gerstaff, 2013: Evolution of lightning activity and storm climatology for Romania. Mon. Wea. Rev., 138, 579–591, charge relative to dual-doppler analysis of a high-precipita- https://doi.org/10.1175/2009MWR2975.1. tion supercell storm. Mon. Wea. Rev., 141, 2199–2223, https:// Apke,J.M., J. R. Mecikalski, K. Bedka,E. W. McCaulJr.,C.R. doi.org/10.1175/MWR-D-12-00258.1. Homeyer, and C. P. Jewett, 2018: Relationships between Carlson, T. N., and F. H. Ludlam, 1968: Conditions for the occur- deep convection updraft characteristics and satellite-based rence of severe local storms. Tellus, 20, 203–226, https://doi. super rapid scan mesoscale atmospheric motion vector–der- org/10.3402/tellusa.v20i2.10002. ived ﬂow. Mon. Wea. Rev., 146, 3461–3480, https://doi.org/10. Celinski-Mysław, D., and D. Matuszko, 2014: An analysis of 1175/MWR-D-18-0119.1. selected cases of derecho in Poland. Atmos. Res., 149, 263– Bielec-Ba˛kowska, Z., 2002: Zro ´zni ˙ cowanie przestrzenne i 281, https://doi.org/10.1016/j.atmosres.2014.06.016. zmiennos ´c ´ wieloletnia wystRepowania burz w Polsce }}, A. Palarz, and M. Taszarek, 2020: Climatology and atmo- (1949–1998). Wydawnictwo Uniwersytetu Sla˛skiego, 159 pp. spheric conditions associated with cool season bow echo }}, 2003: Long-term variability of thunderstorms occurrence in storms in Poland. Atmos. Res., 240, 104944, https://doi.org/10. Poland in the 20th century. Atmos. Res., 67–68,35–52, https:// 1016/j.atmosres.2020.104944. Chernokulsky, A. V., M. V. Kurgansky, and I. I. Mokhov, 2019: doi.org/10.1016/S0169-8095(03)00082-6. }}, 2013: Burze i grady w Polsce (Thunderstorms and hails in On characteristic reanalysis-based values of convective insta- Poland). Prz. Geogr., 132,99–132, https://doi.org/10.4467/ bility indices for Northern Eurasia tornadoes. IOP Conf. Ser. 20833113PG.13.005.1095. Earth Environ. Sci., 231, 012012, https://doi.org/10.1088/1755- Biron, D., 2009: LAMPINET-Lightning detection in Italy. Light- 1315/231/1/012012. ning: Principles, Instruments and Applications: Review of Christian, H. J., and Coauthors, 1999: The lightning imaging sen- Modern Lightning Research, H. D. Betz, U. Schumann, and sor. 11th Int. Conf. on Atmospheric Electricity, H.J.Christian, E. Laroche, Eds., Springer, 141–159. Ed., NASA Conf. Publ. NASA/CP-1999-209261, 746–749. Blair, S. F., and Coauthors, 2017: High-resolution hail observa- Coffer, B. E., and M. D. Parker, 2015: Impacts of increasing low- tions: Implications for NWS warning operations. Wea. level shear on supercells during the early evening transition. Forecasting, 32, 1101–1119, https://doi.org/10.1175/WAF-D- Mon. Wea. Rev., 143, 1945–1969, https://doi.org/10.1175/ 16-0203.1. MWR-D-14-00328.1. Bodzak, P., 2006: Detekcja i lokalizacja wyładowan ´ atmosferycz- Cohuet, J. B., R. Romero, V. Homar, V. Ducrocq, and C. Ramis, 2011: Initiation of a severe thunderstorm over the Mediterra- nych (Lightning Detection and Localization). Instytut Mete- orologii i Gospodarki Wodnej, 135 pp. nean Sea. Atmos. Res., 100,603–620, https://doi.org/10.1016/j. }}, Z. Dziewit, and G. Wojciech, 2006: The ratio of cloud to atmosres.2010.11.002. cloud-to-ground ﬂashes in summer seasons for Poland terri- Colman, B. R., 1990: Thunderstorms above frontal surfaces in tory on the basis of data from PERUN system. 19th Int. environments without positive CAPE. Part II: Organization Lightning Detection Conf. (ILDC)/1st Int. Lightning Meteo- and instability mechanisms. Mon. Wea. Rev., 118, 1123–1144, rology Conf. (ILMC), Tucson, AR, Vaisala, https://my. https://doi.org/10.1175/1520-0493(1990)118,1123:TAFSIE.2. vaisala.net/en/events/ildcilmc/archive/Pages/ILDC-2006-Archive. 0.CO;2. aspx. Cummins, K. L., and M. J. Murphy, 2009: An overview of light- Bright, D. R., M. S. Wandishin, R. E. Jewell, and S. J. Weiss, ning locating systems: History, techniques, and data uses, with an in-depth look at the U.S. NLDN. IEEE Trans. 2004: A physically based parameter for lightning prediction Electromagn. Compat., 51, 499–518, https://doi.org/10.1109/ and its calibration in ensemble forecasts. 22nd Conf. on Severe Local Storms, Hyannis, MA, Amer. Meteor. Soc., 4.3., TEMC.2009.2023450. http://ams.confex.com/ams/pdfpapers/84173.pdf. Czernecki, B., M. Taszarek, L. Kolendowicz, and J. Konarski, Brooks, H. E., J. W. Lee, and J. P. Craven, 2003: The spatial dis- 2016: Relationship between human observations of thunder- tribution of severe thunderstorm and tornado environments storms and the PERUN lightning detection network in from global reanalysis data. Atmos. Res., 67–68,73–94, Poland. Atmos. Res., 167,118–128, https://doi.org/10.1016/j. https://doi.org/10.1016/S0169-8095(03)00045-0. atmosres.2015.08.003. }}, A. R. Anderson, K. Riemann, I. Ebbers, and H. Flachs, }}, }},M.Marosz, M. Po ´ łrolniczak, L. Kolendowicz, A. Wys- 2007: Climatological aspects of convective parameters from zogrodzki, and J. Szturc, 2019: Application of machine learn- the NCAR/NCEP reanalysis. Atmos. Res., 83, 294–305, ing to large hail prediction - The importance of radar reﬂec- tivity, lightning occurrence and convective parameters https://doi.org/10.1016/j.atmosres.2005.08.005. }}, and Coauthors, 2011: Evaluation of European Storm Fore- derived from ERA5. Atmos. Res., 227, 249–262, https://doi. cast Experiment (ESTOFEX) forecasts. Atmos. Res., 100, org/10.1016/j.atmosres.2019.05.010. 538–546, https://doi.org/10.1016/j.atmosres.2010.09.004. Deierling, W., and W. A. Petersen, 2008: Total lightning activity Bunkers, M. J., B. A. Klimowski, J. W. Zeitler, R. L. Thompson, as an indicator of updraft characteristics. J. Geophys. Res., and M. L. Weisman, 2000: Predicting supercell motion using 113, D16210, https://doi.org/10.1029/2007JD009598. AUGUST 2022 PO R E ˛ BA ET A L . 1467 Dennis, E. J., and M. R. Kumjian, 2017: The impact of vertical Farr, T. G., and Coauthors, 2007: The Shuttle Radar Topography wind shear on hail growth in simulated supercells. J. Atmos. Mission. Rev. Geophys., 45, RG2004, https://doi.org/10.1029/ 2005RG000183. Sci., 74, 641–663, https://doi.org/10.1175/JAS-D-16-0066.1. DiGangi, E. A., M. Stock, and J. Lapierre, 2021: Thunder hours: Feudale, L., A. Manzato, and S. Micheletti, 2013: A cloud-to ground lightning climatology for north-eastern Italy. Adv. Sci. How old methods offer new insights into thunderstorm clima- Res., 10,77–84, https://doi.org/10.5194/asr-10-77-2013. tology. Bull. Amer. Meteor. Soc., 103, E548–E569, https://doi. Fluck, E., M. Kunz, P. Geissbuehler, and S. P. Ritz, 2021: Radar- org/10.1175/BAMS-D-20-0198.1. based assessment of hail frequency in Europe. Nat. Hazards Doswell, C. A., 1985: The operational meteorology of convective Earth Syst. Sci., 21,683–701, https://doi.org/10.5194/nhess-21- weather: Volume II: Storm scale analysis. NOAA Tech. 683-2021. Memo. ERL ESG-15, 240 pp., https://repository.library.noaa. Gatlin, P. N., and S. J. Goodman, 2010: A total lightning trending gov/view/noaa/11215. algorithm to identify severe thunderstorms. J. Atmos. Oceanic }}, 2001: Severe convective storms}An overview. Severe Con- Technol., 27,3–22, https://doi.org/10.1175/2009JTECHA1286. vective Storms, Meteor. Monogr., No. 50, Amer. Meteor. Soc., 1–26, https://doi.org/10.1175/0065-9401-28.50.1. Gatzen, C., 2011: A 10-year climatology of cold-season narrow }}, and E. N. Rasmussen, 1994: The effect of neglecting the vir- cold-frontal rainbands in Germany. Atmos. Res., 100, 366– tual temperature correction on CAPE calculations. Wea. Fore- 370, https://doi.org/10.1016/j.atmosres.2010.09.018. casting, 9, 625–629, https://doi.org/10.1175/1520-0434(1994)009 }}, A. H. Fink, D. M. Schultz, and J. G. Pinto, 2020: An 18- ,0625:TEONTV.2.0.CO;2. year climatology of derechos in Germany. Nat. Hazards }}, and D. M. Schultz, 2006: On the use of indices and parame- Earth Syst. Sci., 20,1335–1351, https://doi.org/10.5194/nhess- ters in forecasting severe storms. Electron. J. Severe Storms 20-1335-2020. Meteor., 1 (3), 1–22. Geerts, B., and Coauthors, 2017: The 2015 Plains Elevated Con- }}, H. E. Brooks, and R. A. Maddox, 1996: Flash ﬂood fore- vection at Night ﬁeld project. Bull. Amer. Meteor. Soc., 98, casting: An ingredients-based methodology. Wea. Forecasting, 767–786, https://doi.org/10.1175/BAMS-D-15-00257.1. 11,560–581, https://doi.org/10.1175/1520-0434(1996)011,0560: Gensini, V. A., and W. S. Ashley, 2011: Climatology of potentially FFFAIB.2.0.CO;2. severe convective environments from North American Dotzek, N., P. Groenemeijer, B. Feuerstein, and A. M. Holzer, regional reanalysis. Electron. J. Severe Storms Meteor., 6 (8), 2009: Overview of ESSL’s severe convective storms research 1–40. using the European Severe Weather Database ESWD. Gieysztor, A., 2006: Mitologia Słowian. Wydawnictwo Uniwersy- Atmos. Res., 93, 575–586, https://doi.org/10.1016/j.atmosres. tetu Warszawskiego, 407 pp. 2008.10.020. Goodman, S. J., D. E. Buechler, P. D. Wright, and W. D. Rust, Edwards, R., J. G. LaDue, J. T. Ferree, K. Scharfenberg, 1988: Lightning and precipitation history of a microburst-pro- C. Maier, and W. L. Coulbourne, 2013: Tornado intensity esti- ducing storm. Geophys. Res. Lett., 15, 1185–1188, https://doi. mation: Past, present, and future. Bull. Amer. Meteor. Soc., org/10.1029/GL015i011p01185. 94,641–653, https://doi.org/10.1175/BAMS-D-11-00006.1. ´ ´ ´ Grabowska, K., 2001: Zmiennoscaktywnosci burzowej w }}, J. T. Allen, and G. W. Carbin, 2018: Reliability and climato- niekto ´rych regionach geograﬁcznych Polski. Prace i Studia logical impacts of convective wind estimations. J. Appl. Geograﬁczne UW, 29, 181–190. Meteor. Climatol., 57, 1825–1845, https://doi.org/10.1175/ Grant, B. N., 1995: Elevated cold-sector severe thunderstorms: A JAMC-D-17-0306.1. preliminary study. Natl. Wea. Dig., 19,25–31. Elmore, K. L., Z. L. Flamig, V. Lakshmanan, B. T. Kaney, Groenemeijer, P., and Coauthors, 2017: Severe convective storms V. Farmer, H. D. Reeves, and L. P. Rothfusz, 2014: MPING: in Europe: Ten years of research at the European Severe Crowd-sourcing weather reports for research. Bull. Amer. Storms Laboratory. Bull. Amer. Meteor. Soc., 98, 2641–2651, Meteor. Soc., 95, 1335–1342, https://doi.org/10.1175/BAMS-D- https://doi.org/10.1175/BAMS-D-16-0067.1. 13-00014.1. Gropp, M. E., and C. E. Davenport, 2018: The impact of the noc- Enno, S. E., 2011: A climatology of cloud-to-ground lightning turnal transition on the lifetime and evolution of supercell over Estonia, 2005–2009. Atmos. Res., 100,310–317, https:// thunderstorms in the Great Plains. Wea. Forecasting, 33, doi.org/10.1016/j.atmosres.2010.08.024. 1045–1061, https://doi.org/10.1175/WAF-D-17-0150.1. }}, J. Sugier, R. Alber, and M. Seltzer, 2020: Lightning ﬂash Haberlie, A. M., and W. S. Ashley, 2019: A radar-based climatol- density in Europe based on 10 years of ATDnet data. Atmos. ogy of mesoscale convective systems in the United States. J. Res., 235, 104769, https://doi.org/10.1016/j.atmosres.2019. Climate, 32,1591–1606, https://doi.org/10.1175/JCLI-D-18- 0559.1. Farnell, C., and T. Rigo, 2020: The lightning jump algorithm for Hanstrum, B. N., G. A. Mills, A. Watson, J. P. Monteverdi, and nowcasting convective rainfall in Catalonia. Atmosphere, 11, C. A. Doswell, 2002: The cool-season tornadoes of California 397, https://doi.org/10.3390/atmos11040397. and Southern Australia. Wea. Forecasting, 17,705–722, }}, }}, and N. Pineda, 2017: Lightning jump as a nowcast pre- https://doi.org/10.1175/1520-0434(2002)017,0705: dictor: Application to severe weather events in Catalonia. TCSTOC.2.0.CO;2. Atmos. Res., 183, 130–141, https://doi.org/10.1016/j.atmosres. Hersbach, H. B., and Coauthors, 2020: The ERA5 global reanaly- 2016.08.021. sis. Quart. J. Roy. Meteor. Soc., 146, 1999–2049, https://doi. }}, }},and }}, 2018: Exploring radar and lightning varia- org/10.1002/qj.3803. bles associated with the Lightning Jump. Can we predict the Holle, R. L., 2014: Diurnal variations of NLDN-reported cloud- size of the hail? Atmos. Res., 202,175–186, https://doi.org/10. to-ground lightning in the United States. Mon. Wea. Rev., 1016/j.atmosres.2017.11.019. 142,1037–1052, https://doi.org/10.1175/MWR-D-13-00121.1. 1468 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 }}, K. L. Cummins, and W. A. Brooks, 2016: Seasonal, 221–251, https://doi.org/10.1175/1520-0469(1989)046,0221:LRRTTS monthly, and weekly distributions of NLDN and GLD360 .2.0.CO;2. ¨ ¨ cloud-to-ground lightning. Mon. Wea. Rev., 144, 2855–2870, Makela, A., P. Rossi, and D. M. Schultz, 2011: The daily cloud-to- https://doi.org/10.1175/MWR-D-16-0051.1. ground lightning ﬂash density in the contiguous United States Holley, D. M., S. R. Dorling, C. J. Steele, and N. Earl, 2014: A cli- and Finland. Mon. Wea. Rev., 139, 1323–1337, https://doi.org/ matology of convective available potential energy in Great 10.1175/2010MWR3517.1. Britain. Int. J. Climatol., 34, 3811–3824, https://doi.org/10. Markowski, P., and Y. Richardson, 2010: Mesoscale Meteorology 1002/joc.3976. in Midlatitudes. John Wiley & Sons, Ltd., 407 pp. Houze, R. A., Jr., 2018: 100 years of research on mesoscale con- Mathias, L., V. Ermert, F. D. Kelemen, P Ludwig, and J. G. Pinto, 2017: Synoptic analysis and hindcast of an intense bow echo in vective systems. A Century of Progress in Atmospheric and Related Sciences: Celebrating the American Meteorological western Europe: The 9 June 2014 storm. Wea. Forecasting, 32, Society Centennial, Meteor. Monogr.,No. 59, Amer.Meteor. 1121–1141, https://doi.org/10.1175/WAF-D-16-0192.1. }}, P. Ludwig, and J. G. Pinto, 2019: Synoptic-scale conditions Soc., https://doi.org/10.1175/AMSMONOGRAPHS-D-18- and convection-resolving hindcast experiments of a cold-sea- 0001.1. Ingrosso, R., P. Lionello, M. M. Miglietta, and G. A. Salvadori, son derecho on 3 January 2014 in western Europe. Nat. Haz- 2020: Statistical investigation of mesoscale precursors of sig- ards Earth Syst. Sci., 19, 1023–1040, https://doi.org/10.5194/ niﬁcant tornadoes: The Italian case study. Atmosphere, 11, nhess-19-1023-2019. NOAA/NWS, 2021: Nautical dawn. NOAA/NWS Glossary, 301, https://doi.org/10.3390/atmos11030301. Johns, R. H., and C. A. Doswell, 1992: Severe local storms fore- https://w1.weather.gov/glossary/index.php?word=nautical+ casting. Wea. Forecasting, 7,588–612, https://doi.org/10.1175/ dawn. Novak ´ , P., and H. Kyznarova ´, 2011: Climatology of lightning in 1520-0434(1992)007,0588:SLSF.2.0.CO;2. Kingﬁeld, D. M., K. M. Calhoun, and K. M. de Beurs, 2017: the Czech Republic. Atmos. Res., 100, 318–333, https://doi. Antenna structures and cloud-to-ground lightning location: org/10.1016/j.atmosres.2010.08.022. 1995–2015. Geophys. Res. Lett., 44, 5203–5212, https://doi.org/ Orville, R. E., 1991: Lightning ground ﬂash density in the contigu- ous United States}1989. Mon. Wea. Rev., 119, 573–577, 10.1002/2017GL073449. Koehler, T. L., 2020: Cloud-to-ground lightning ﬂash density and https://doi.org/10.1175/1520-0493(1991)119,0573:LGFDIT.2.0. thunderstorm day distributions over the contiguous United CO;2. States derived from NLDN measurements: 1993–2018. Mon. }}, and G. R. Hufﬁnes, 2001: Cloud-to-ground lightning in the United States: NLDN results in the ﬁrst decade, 1989–98. Wea. Rev., 148, 313–332, https://doi.org/10.1175/MWR-D-19- 0211.1. Mon. Wea. Rev., 129,1179–1193, https://doi.org/10.1175/1520- Kolendowicz, L., 2006: The inﬂuence of synoptic situations on the 0493(2001)129,1179:CTGLIT.2.0.CO;2. }}, }},W.R.Burrows,R.L.Holle, and K. L. Cummins, occurrence of days with thunderstorms during a year in the 2002: The North American Lightning Detection Network territory of Poland. Int. J. Climatol., 26, 1803–1820, https:// doi.org/10.1002/joc.1348. (NALDN)}First results: 1998–2000. Mon. Wea. Rev., 130, Krennert, T., G. Pistotnik, R. Kaltenberger, and C. Csekits, 2018: 2098–2109, https://doi.org/10.1175/1520-0493(2002)130,2098: TNALDN.2.0.CO;2. Crowdsourcing of weather observations at national meteoro- Pacey, G. P., D. M. Schultz, and L. Garcia-Carreras, 2021: Severe logical and hydrological services in Europe. Adv. Sci. Res., 15,71–76, https://doi.org/10.5194/asr-15-71-2018. convective windstorms in Europe: Climatology, preconvective Lang, T. J., and Coauthors, 2004: The Severe Thunderstorm Elec- environments, and convective mode. Wea. Forecasting, 36, 237–252, https://doi.org/10.1175/WAF-D-20-0075.1. triﬁcation and Precipitation Study. Bull. Amer. Meteor. Soc., Perez, A. H., L. J. Wicker, and R. E. Orville, 1997: Characteristics 85, 1107–1126, https://doi.org/10.1175/BAMS-85-8-1107. Lehmiller, G. S., H. B. Bluestein, P. J. Neiman, F. M. Ralph, and of cloud-to-ground lightning associated with violent torna- W. F. Feltz, 2001: Wind structure in a supercell thunderstorm does. Wea. Forecasting, 12,428–437, https://doi.org/10.1175/ 1520-0434(1997)012,0428:COCTGL.2.0.CO;2. as measured by a UHF wind proﬁler. Mon. Wea. Rev., 129, Pilguj, N., M. Taszarek, Ł. Pajurek, and M. Kryza, 2019: High-res- 1968–1986, https://doi.org/10.1175/1520-0493(2001)129,1968: WSIAST.2.0.CO;2. olution simulation of an isolated tornadic supercell in Poland Lin, Y., and M. R. Kumjian, 2021: Inﬂuences of CAPE on hail on 20 June 2016. Atmos. Res., 218, 145–159, https://doi.org/10. 1016/j.atmosres.2018.11.017. production in simulated supercell storms. J. Atmos. Sci., 79, Pilorz, W., 2015: Very large hail occurrence in Poland from 2007 179–204, https://doi.org/10.1175/JAS-D-21-0054. Liu, N., C. Liu, B. Chen, and E. Zipser, 2020: What are the favor- to 2015. Contemp. Trends. Geosci., 4,45–55, https://doi.org/ able large-scale environments for the highest-ﬂash-rate thun- 10.1515/ctg-2015-0005. Poelman, D. R., W. Schulz, G. Diendorfer, and M. Bernardi, 2016: derstorms on Earth? J. Atmos. Sci., 77, 1583–1612, https://doi. The European lightning location system EUCLID}Part 2: org/10.1175/JAS-D-19-0235.1. Łoboda, M., H. D. Betz, P. Baranski, J. Wiszniowski, and Z. Observations. Nat. Hazards Earth Syst, 16,607–616, https:// Dziewit, 2009: New lightning detection networks in doi.org/10.5194/nhess-16-607-2016. Pore Rba, S., and Z. Ustrnul, 2020: Forecasting experiences associ- Poland}LINET and LLDN. Open Atmos. Sci. J., 3,29–38, ated with supercells over South-Western Poland on July 7, https://doi.org/10.2174/1874282300903010029. Lorenc, H., 2005: Atlas klimatu Polski. Instytut Meteorologii i 2017. Atmos. Res., 232, 104681, https://doi.org/10.1016/j. Gospodarki Wodnej, 116 pp. atmosres.2019.104681. Macgorman, D. R., D.W. Burgess, V.Mazur,W. D.Rust, Price, C., 2008: Lightning sensors for observing, tracking and now- W. L. Taylor, and B. C. Johnson, 1989: Lightning rates relative casting severe weather. Sensors, 8,157–170, https://doi.org/10. to tornadic storm evolution on 22 May 1981. J. Atmos. Sci., 46, 3390/s8010157. AUGUST 2022 PO R E ˛ BA ET A L . 1469 Pu ´cik ˇ T., P. Groenemeijer, D. Ryv ´ a, and M. Kolar, ´ 2015: Proxim- Smith, B. T., R. L. Thompson, J. S. Grams, C. Broyles, and H. E. ity soundings of severe and nonsevere thunderstorms in cen- Brooks, 2012: Convective modes for signiﬁcant severe thun- tral Europe. Mon. Wea. Rev., 143, 4805–4821, https://doi.org/ derstorms in the contiguous United States. Part I: Storm clas- 10.1175/MWR-D-15-0104.1. siﬁcation and climatology. Wea. Forecasting, 27, 1114–1135, Punge, H. J., K. M. Bedka, M. Kunz, and A. Reinbold, 2017: Hail https://doi.org/10.1175/WAF-D-11-00115.1. frequency estimation across Europe based on a combination Steiger, S. M., R. E. Orville, and L. D. Carey, 2007: Total light- of overshooting top detections and the ERA-INTERIM ning signatures of thunderstorm intensity over north Texas. reanalysis. Atmos. Res., 198,34–43, https://doi.org/10.1016/j. Part I: Supercells. Mon. Wea. Rev., 135, 3281–3302, https:// atmosres.2017.07.025. doi.org/10.1175/MWR3472.1. Reif, D. W., and H. B. Bluestein, 2017: A 20-year climatology of Stopa, M., 1962: Burze w Polsce. Prace Instytutu GeograﬁiPAN, nocturnal convection initiation over the central and southern 34,109–179. Great Plains during the warm season. Mon. Wea. Rev., 145, Sulik, S., 2021: Formation factors of the most electrically active thunderstorm days over Poland (2002–2020). Atmos. Res., 34, 1615–1639, https://doi.org/10.1175/MWR-D-16-0340.1. Riemann-Campe, K., K. Fraedrich, and F. Lunkeit, 2009: Global 100386, https://doi.org/10.1016/j.wace.2021.100386. climatology of convective available potential energy (CAPE) Surowiecki, A., and M. Taszarek, 2020: A 10-year radar-based cli- and convective inhibition (CIN) in ERA-40 reanalysis. matology of mesoscale convective system archetypes and der- Atmos. Res., 93, 534–545, https://doi.org/10.1016/j.atmosres. echos in Poland. Mon. Wea. Rev., 148,3471–3488, https://doi. 2008.09.037. org/10.1175/MWR-D-19-0412.1. Rodr´ıguez, O., and J. Bech, 2021: Tornadic environments in the Taszarek, M., B. Czernecki, and A. Kozioł, 2015: A cloud-to- IberianPeninsula andthe Balearic Islands basedonERA5 ground lightning climatology for Poland. Mon. Wea. Rev., reanalysis. Int. J. Climatol., 41,E1959–E1979, https://doi.org/ 143,4285–4304, https://doi.org/10.1175/MWR-D-15-0206.1. 10.1002/joc.6825. }}, and Coauthors, 2019: A climatology of thunderstorms across Said, R.K., M. J. Murphy, N.W.S. Demetriades,K.L. Cummins, Europe from a synthesis of multiple data sources. J. Climate, and U. S. Inan, 2011: Methodology and performance esti- 32, 1813–1837, https://doi.org/10.1175/JCLI-D-18-0372.1. mates of the GLD360 lightning detection network. Proc. 14th }}, J.T.Allen, T. Pu ´c ˇik, K. Hoogewind, and H. E. Brooks, Int. Conf. on Atmospheric Electricity, Rio de Janeiro, Brazil, 2020: Severe convective storms across Europe and the United ICAE, 243 pp., http://www.icae2011.net.br/upload/resumos. States. Part II: ERA5 environments associated with lightning, pdf. large hail, severe wind and tornadoes. J. Climate, 33, 10 263– Schultz, C. J., W. A. Petersen, and L. D. Carey, 2009: Preliminary 10 286, https://doi.org/10.1175/JCLI-D-20-0346.1. development and evaluation of lightning jump algorithms Thompson, R. L., R. Edwards, J. A. Hart, K. L. Elmore, and P. for the real-time detection of severe weather. J. Appl. Markowski, 2003: Close proximity soundings within supercell Meteor. Climatol., 48,2543–2563, https://doi.org/10.1175/ environments obtained from the Rapid Update Cycle. 2009JAMC2237.1. Wea. Forecasting, 18,1243–1261, https://doi.org/10.1175/1520- }}, }},and }}, 2011: Lightning and severe weather: A com- 0434(2003)018,1243:CPSWSE.2.0.CO;2. parison between total and cloud-to-ground lightning trends. }}, C. M. Mead, and R. Edwards, 2007: Effective storm-relative Wea. Forecasting, 26, 744–755, https://doi.org/10.1175/WAF- helicity and bulk shear in supercell thunderstorm environ- D-10-05026.1. ments. Wea. Forecasting, 22, 102–115, https://doi.org/10.1175/ }}, L. D. Carey, E. V. Schultz, and R. J. Blakeslee, 2017: Kine- WAF969.1. matic and microphysical signiﬁcance of lightning jumps versus }}, B. T. Smith, J. S. Grams, A. R. Dean, and C. Broyles, 2012: non-jump increases in total ﬂash rate. Wea. Forecasting, 32, Convective modes for signiﬁcant severe thunderstorms in the 275–288, https://doi.org/10.1175/WAF-D-15-0175.1. contiguous United States. Part II: Supercell and QLCS tor- Schultz, E. V.,C.J. Schultz,L. D.Carey,D.J.Cecil,and M. Bate- nado environments. Wea. Forecasting, 27,1136–1154, https:// man, 2016: Automated Storm tracking and the lightning doi.org/10.1175/WAF-D-11-00116.1. jump algorithm using GOES-R Geostationary Lightning Tudur´ı, E., and C. Ramis, 1997: The environments of signiﬁcant Mapper (GLM) proxy data. J. Operat. Meteor., 4,92–107, convective events in the western Mediterranean. Wea. Forecast- https://doi.org/10.15191/nwajom.2016.0407. ing, 12, 294–306, https://doi.org/10.1175/1520-0434(1997)012,0294: Schulz, W., K. Cummins, G. Diendorfer, and M. Dorninger, 2005: TEOSCE.2.0.CO;2. Cloud-to-ground lightning in Austria: A 10-year study using Ustrnul, Z., and D. Czekierda, 2009: Atlas ekstremalnych zjawisk data from a lightning location system. J. Geophys. Res., 110, meteorologicznych oraz sytuacji synoptycznych w Polsce. D09101, https://doi.org/10.1029/2004JD005332. Instytut Meteorologii i Gospodarki Wodnej, 182 pp. Seimon, A., J. T. Allen, T. A. Seimon, S. J. Talbot, and D. K. Van Delden, A., 2001: The synoptic setting of thunderstorms in Hoadley, 2016: Crowdsourcing the El Reno 2013 tornado: A western Europe. Atmos. Res., 56,89–110, https://doi.org/10. new approach for collation and display of storm chaser imag- 1016/S0169-8095(00)00092-2. ery for scientiﬁc applications. Bull. Amer. Meteor. Soc., 97, Verbout, S. M., H. E. Brooks, L. M. Leslie, and D. M. Schultz, 2069–2084, https://doi.org/10.1175/BAMS-D-15-00174.1. 2006: Evolution of the U.S. tornado database: 1954–2003. Sherburn, K. D., and M. D. Parker, 2014: Climatology and ingre- Wea. Forecasting, 21,86–93, https://doi.org/10.1175/WAF910. dients of signiﬁcant severe convection in high-shear, low- 1. CAPE environments. Wea. Forecasting, 29, 854–877, https:// Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H. Holzworth, doi.org/10.1175/WAF-D-13-00041.1. 2013: Highlights of a new ground-based, hourly global light- }}, }}, J. R. King, and G. M. Lackmann, 2016: Composite ning climatology. Bull. Amer. Meteor. Soc., 94, 1381–1391, environments of severe and nonsevere high-shear, low-CAPE https://doi.org/10.1175/BAMS-D-12-00082.1. convective events. Wea. Forecasting, 31,1899–1927, https:// Wacker, R. S., and R. E. Orville, 1999: Changes in measured doi.org/10.1175/WAF-D-16-0086.1. lightning ﬂash count and return stroke peak current after the 1470 W E A T H E R A N D F O R ECA S T I N G VOLUME 37 1994 U.S. National Lightning Detection Network upgrade: 2. extremely severe and widespread geohazard in Poland. Envi- Theory. J. Geophys. Res., 104, 2159–2162, https://doi.org/10. ron. Socio-Econ. Stud., 6,10–16, https://doi.org/10.2478/ 1029/1998JD200059. environ-2018-0002. Walawender, E., R. Kielar, and Z. Ustrnul, 2017: Use of RegCM Wiens, K. C., S. A. Rutledge, and S. A. Tessendorf, 2005: The 29 gridded dataset for thunderstorm favorable conditions analy- June 2000 supercell observed during STEPS. Part II: Light- sis over Poland}Climatological approach. Theor. Appl. Cli- ning and charge structure. J. Atmos. Sci., 62, 4151–4177, matol., 127,229–240, https://doi.org/10.1007/s00704-015-1620- https://doi.org/10.1175/JAS3615.1. x. Williams, E., and Coauthors, 1999: The behavior of total lightning Weisman, M. L., and J. B. Klemp, 1982: The dependence of activity in severe Florida thunderstorms. Atmos. Res., 51, numerically simulated convective storms on vertical wind 245–265, https://doi.org/10.1016/S0169-8095(99)00011-3. shear and buoyancy. Mon. Wea. Rev., 110, 504–520, https:// Williams, E. R., M. E. Weber, and R. E. Orville, 1989: The rela- doi.org/10.1175/1520-0493(1982)110,0504:TDONSC.2.0. tionship between lightning type and convective state of thun- CO;2. derclouds. J. Geophys. Res., 94, 13 213–13220, https://doi.org/ Westermayer, A. T., P. Groenemeijer, G. Pistotnik, R. Sausen, 10.1029/JD094iD11p13213. and E. Faust, 2017: Identiﬁcation of favorable environments Zajac, B. A., and S. A. Rutledge, 2001: Cloud-to-ground lightning for thunderstorms in reanalysis data. Meteor. Z., 26,59–70, activity in the contiguous United States from 1995 to 1999. https://doi.org/10.1127/metz/2016/0754. Mon. Wea. Rev., 129, 999–1019, https://doi.org/10.1175/1520- Widawski, A., and W. Pilorz, 2018: The mesoscale convective sys- tems with bow echo radar signatures as an example of 0493(2001)129,0999:CTGLAI.2.0.CO;2.

Journal

Weather and Forecasting – American Meteorological Society

Published: Aug 5, 2022

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland

Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland

Diurnal and Seasonal Variability of ERA5 Convective Parameters in Relation to Lightning Flash Rates in Poland

References (140)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies