A Cloud-to-Ground Lightning Climatology for Poland

Mateusz Taszarek Department of Climatology, Institute of Physical Geography and Environmental Planning, Adam Mickiewicz University, Poznań, Poland

Search for other papers by Mateusz Taszarek in
Current site
Google Scholar
PubMed
Close
,
Bartosz Czernecki Department of Climatology, Institute of Physical Geography and Environmental Planning, Adam Mickiewicz University, Poznań, Poland

Search for other papers by Bartosz Czernecki in
Current site
Google Scholar
PubMed
Close
, and
Aneta Kozioł Institute of Meteorology and Water Management, National Research Institute, Warsaw, Poland

Search for other papers by Aneta Kozioł in
Current site
Google Scholar
PubMed
Close
Full access

We are aware of a technical issue preventing figures and tables from showing in some newly published articles in the full-text HTML view.
While we are resolving the problem, please use the online PDF version of these articles to view figures and tables.

Abstract

This research focuses on the climatology of cloud-to-ground (CG) lightning flashes based on PERUN lightning detection network data from 2002 to 2013. To present various CG lightning flash characteristics, 10 km × 10 km grid cells are used, while for estimating thunderstorm days, circles with radii of 17.5 km in the 1 km × 1 km grid cells are used. A total of 4 328 892 CG lightning flashes are used to analyze counts, density, polarity, peak current, and thunderstorm days. An average of 151 days with thunderstorm (appearing anywhere in Poland) occurs each year. The annual number of days with thunderstorms increases southeasterly from the coast of the Baltic Sea (15–20 days) to the Carpathian Mountains (30–35 days). The mean CG lightning flash density varies from 0.2 to 3.1 flashes km−2 yr−1 with the highest values in the southwest–northeast belt from Kraków-Częstochowa Upland to the Masurian Lake District. The maximum daily CG lightning flash density in this region amounted to 9.1 km−2 day−1 (3 July 2012). The monthly variation shows a well-defined thunderstorm season extending from May to August with July as the peak month. The vast majority of CG lightning flashes were detected during the daytime (85%) with a peak at 1400 UTC and a minimum at 0700 UTC. Almost 97% of all CG lightning flashes in the present study had a negative current, reaching the highest average monthly values in February (55 kA) and the lowest in July (24 kA). The percentage of positive CG lightning flashes was the lowest during the summer (2%–3%) and the highest during the winter (10%–20%).

Corresponding author address: Mateusz Taszarek, Dept. of Climatology, Institute of Physical Geography and Environmental Planning, Adam Mickiewicz University, St. Dzięgielowa 27, 61-680 Poznań, Poland. E-mail: mateusz.taszarek@amu.edu.pl

Abstract

This research focuses on the climatology of cloud-to-ground (CG) lightning flashes based on PERUN lightning detection network data from 2002 to 2013. To present various CG lightning flash characteristics, 10 km × 10 km grid cells are used, while for estimating thunderstorm days, circles with radii of 17.5 km in the 1 km × 1 km grid cells are used. A total of 4 328 892 CG lightning flashes are used to analyze counts, density, polarity, peak current, and thunderstorm days. An average of 151 days with thunderstorm (appearing anywhere in Poland) occurs each year. The annual number of days with thunderstorms increases southeasterly from the coast of the Baltic Sea (15–20 days) to the Carpathian Mountains (30–35 days). The mean CG lightning flash density varies from 0.2 to 3.1 flashes km−2 yr−1 with the highest values in the southwest–northeast belt from Kraków-Częstochowa Upland to the Masurian Lake District. The maximum daily CG lightning flash density in this region amounted to 9.1 km−2 day−1 (3 July 2012). The monthly variation shows a well-defined thunderstorm season extending from May to August with July as the peak month. The vast majority of CG lightning flashes were detected during the daytime (85%) with a peak at 1400 UTC and a minimum at 0700 UTC. Almost 97% of all CG lightning flashes in the present study had a negative current, reaching the highest average monthly values in February (55 kA) and the lowest in July (24 kA). The percentage of positive CG lightning flashes was the lowest during the summer (2%–3%) and the highest during the winter (10%–20%).

Corresponding author address: Mateusz Taszarek, Dept. of Climatology, Institute of Physical Geography and Environmental Planning, Adam Mickiewicz University, St. Dzięgielowa 27, 61-680 Poznań, Poland. E-mail: mateusz.taszarek@amu.edu.pl

1. Introduction

Thunderstorms pose a direct risk to human lives and property. In the United States, in addition to flash-flood phenomena, cloud-to-ground (CG) lightning flashes are among the leading causes of weather-related fatalities (Holle et al. 1999; Curran et al. 2000). On average 10 people are killed in Poland each year by CG lightning flashes, as shown by the data from the Polish National Institute of Statistics. According to the European Severe Weather Database (ESWD; Groenemeijer et al. 2004; Dotzek et al. 2009), between 2012 and 2014 Poland experienced more than 100 damaging lightning events that killed 17 people.

CG lightning flashes are also associated with economic losses; they affect high-voltage power lines, cause forest and infrastructure fires, and also result in transportation disruptions (Wierzchowski et al. 2002; Sasse and Hauf 2003; Larjavaara et al. 2005; Mäkelä et al. 2013). Knowing the spatial and temporal distribution of thunderstorms can improve weather forecasting, and also can help urban planners, insurance companies, and the public to be better prepared (Brooks et al. 2003a).

For decades the main climatological research into thunderstorm spatial and temporary occurrence was based on observations at meteorological stations. Although human observations allow one to analyze long-term changes in the number of thunderstorm days [100-yr climatologies: Changnon and Changnon (2001); Bielec-Bąkowska (2003)], they cannot estimate the intensity of thunderstorms (Rakov and Uman 2003). Storms with either one lightning strike or thousands of flashes are reported as one thunderstorm. This however can be examined using lightning detection that allows one to count the number of flashes on particular days and in particular locations.

CG lightning flash climatologies based on data from ground-based lightning detection networks have been developed for some European countries.

a. Central Europe

Based on data recorded between 1992 and 2001 from the Austrian Lightning Detection and Information System (ALDIS), Schulz et al. (2005) showed that thunderstorms are most likely to occur from May to September, especially over the southern part of Austria, where the topographical and meteorological conditions are most favorable (flash density up to 4 flashes km−2 yr−1, computed within 1 km × 1 km grid cells). Using the same grid resolution in the climatology of lightning characteristics within central Europe, Wapler (2013) estimated the highest CG lightning flash density to be in southern Germany, with more than 30 flashes km−2 yr−1. In the Czech Republic, the number was estimated across 20 km × 20 km grid cells and found to average 1–3 CG lightning flashes km−2 yr−1 each year (Novák and Kyznarová 2011).

b. Southern Europe

A 10-yr period (1992–2001) of lightning data derived from the Spanish Lightning Detection Network (SLDN) was analyzed over 0.2° × 0.2° grid cells by Soriano et al. (2005). They found that the lightning density is mainly related to the topography and the atmospheric circulation, with the maximum found over the Pyrenees and along the coast of Catalonia (density up to 2 flashes km−2 yr−1). The CG lightning flash climatology for Portugal (Santos et al. 2012) revealed that thunderstorms are most likely in May and September between 1600 and 1800 UTC. The maximum CG lightning flash density estimated on the 0.1° × 0.1° grid cells was up to 0.6 flashes km−2 yr−1. In Italy, Biron (2009), taking into account CG lightning flash data over 10 km × 10 km grid cells from the Italian National Meteorological Service Lightning Network (LAMPINET) between 2005 and 2007, concluded that Lake Como, Sardinia, the Gulf of Trieste, and Naples, Liguria, and the central Apennine have the highest average annual CG lightning flash densities. The highest CG flash density, up to 9 flashes km−2 yr−1, computed over 0.02° × 0.03° grid cells, was also found in northeastern Italy by Feudale et al. (2013).

c. Northern Europe

In Scandinavia, the Nordic Lightning Information System (NORDLIS) was used by Mäkelä et al. (2014) to construct a CG lightning flash climatology spanning 2002–11. The average daily number of ground flashes peaked in mid-July and early August while cold season (October–April) thunderstorms were most frequent over the sea. At 0.2° × 0.2° grid resolution, Sonnadara et al. (2006) estimated for Sweden the maximum CG lightning density up to 0.4 km2 yr−1 and highlighted that the main thunderstorm season extends from June to August. Daily CG lightning flash densities in the contiguous United States and Finland across 20 km × 20 km grid cells, as well as the relationship between thunderstorm days and CG lightning flash density, were analyzed by Mäkelä et al. (2011). Enno (2011) investigated the lightning climatology of Estonia for the period 2005–09 and estimated that for 10 km × 10 km grid cells the maximum lightning density was 1 flash km−2 yr−1.

d. Eastern Europe

The lightning climatology of Romania, as derived from the Romanian National Lightning Detection Network (RNLDN), was studied by Antonescu and Burcea (2010). The analyses of the years 2003–05 and 2007 revealed that most of the CG lightning flashes occur over the southern slopes of the central meridional Carpathians (density of up to 3 flashes km−2 yr−1 computed within 20 km × 20 km grid cells) from May to September.

e. Poland

For Poland such a study has been not performed yet. However, integrated European lightning climatologies have been developed with the use of the Arrival Time Differencing Network (ATDnet) for the years 2008–13 (Anderson and Klugmann 2014), and the Vaisala Global Lightning Dataset (GLD360) and the European Cooperation for Lightning Detection (EUCLID) dataset for 2011 (Pohjola and Mäkelä 2013). The results show that the lowest lightning flash density was located along the coast of the Baltic Sea while the highest was placed over the southeastern and south-central parts of the country (density of over 3 flashes km−2 yr−1 in the GLD360 and EUCLID networks, and 4 flashes km−2 yr−1 in ATDnet).

The Polish PERUN lightning detection network (Łoboda et al. 2009), which was introduced in 2002, presented the possibility of performing a national analysis. Therefore, by using the data from this network the main aim of this paper is to present a CG lightning flash climatology for Poland. This is a first of its kind study to be performed in Poland and is a contribution to the European CG lightning climatology. Previously, national thunderstorm characteristics in Poland were studied only with the use of surface synoptic observation (SYNOP) reports. Bielec-Bąkowska (2003), Kolendowicz (2006, 2012), and Czernecki et al. (2015) estimated that within a particular location, from 15 to 33 thunderstorm days occur on average every year in Poland.

The paper is organized as follows. Section 2 describes the data and methods used in this study. The results concerning spatial, annual, monthly, and diurnal distributions of lightning count, density, polarity, peak current and thunderstorm days are presented in section 3. The last section contains a summary and concluding remarks.

2. Data and methods

In this section we describe the PERUN lightning detection network, its structure, detection techniques, detection efficiency, location accuracy, quality of the data, and quality control assumptions. We also discuss computational methods that we use to produce maps with thunderstorm days and characteristics of CG lightning flashes.

a. Lightning data

The lightning detection network in Poland is operated by the Institute of Meteorology and Water Management–National Research Institute (IMGW–PIB), and since 2002 has worked operationally under the name of PERUN (the god of thunder and lightning in Slavic mythology). The system consists of nine Surveillance et Alerte Foudre par Interférométrie Radioélectrique (SAFIR3000) total lightning automatic detection stations located at Białystok, Olsztyn, Toruń, Gorzów Wielkopolski, Kalisz, Częstochowa, Włodawa, and Warszawa (Fig. 1a). The network center’s central processor (CP) unit is situated at IMGW–PIB headquarter in Warszawa.

Fig. 1.
Fig. 1.

(a) Locations of SAFIR3000 lightning sensors in the PERUN network with 100-km buffer zones. (b) Average CG lightning flash location accuracy (km) derived from the PERUN database during 2002–13. Computed in 10 km × 10 km grid cells. Dots denote main meteorological stations (44).

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

With the use of interferometry in the band of very high frequency (VHF), sensors perform angular localization of thunderstorm electric activity for both CG and intracloud (IC) flashes. For measurements of various electrical parameters and discrimination between different types of discharges, detections are also performed at low frequencies (LFs). The system uses a direction-finding (DF) technique (Krider et al. 1980). The system is capable of detecting up to 100 events per second (Betz et al. 2009). Further information on lightning detection techniques and their limitations may be found in MacGorman and Rust (1998) and Rakov and Uman (2003).

To construct a climatology of CG lightning flash, we use PERUN data from 2002 to 2013. The R software package (R Core Team 2014) was used for our computational purposes. The PERUN database contained the information concerning the time of the event (milliseconds), the place [World Geodetic System (WGS-84) map projection], uncertainty related to detection (m), the type of discharge (CG or IC), polarity, peak current estimation (kA), and multiplicity. We reprojected the original WGS-84 projection into a meter-based Polish CS92 (EPSG:2180) coordinate system. The basic unit of detection was strokes; however, in the case of a multistroke flash, we considered only the first located stroke and its current. This meant that in the statistics we included only flashes instead of strokes.

As in previous studies on lightning climatologies (e.g., Antonescu and Burcea 2010; Feudale et al. 2013; Mäkelä et al. 2014), we considered only CG lightning flash data and excluded IC flashes. This can be justified by the relatively low detection efficiency and low quality of IC lightning data, which may yield unreliable climatological results. According to the previous studies of Cummins et al. (1998) and Wacker and Orville (1999a,b), some of the CG positive flashes with the peak current below 10 kA may be considered to be IC flashes; therefore, we also filtered out our data from these flashes and they have been removed. They accounted for around 1.5% of all CG flashes and around 33% of all positive CG flashes in our database.

Since the network is able to detect flashes that may appear far from the Polish border (e.g., numerous cases of lightning detections over Kazakhstan), the data have also been limited to the administrative borders of Poland.

b. Detection efficiency and location accuracy

The spatial distribution of the SAFIR3000 sensors is not homogenous in space; therefore, the detection efficiency and the location accuracy vary across the whole country. Bodzak (2006) estimated that PERUN network has a 95% detection efficiency over the area of Poland and that it is particularly high at distances of up to 100 km from the sensor. Considering the 100-km buffer zones around the SAFIR3000 sensors (Fig. 1a), we can define that the highest detection efficiency is located in the central-eastern part of the country while the lowest falls over the coastal zone and the northwestern and southwestern parts of the country.

Bodzak (2006) stated that the PERUN network reveals the lightning location accuracy in the whole country to be below 1 km. However, by analyzing uncertainty related to lightning location accuracy derived from our database (Fig. 1b), we can estimate that only 38.3% of the country has such a value while the location accuracy of less than 2 km covers almost 76.6%. It can be observed that the spatial distribution of the location accuracy is proportional to the density of the SAFIR3000 sensors (Figs. 1a,b). In the places where the distances between the sensors are smaller than 100 km (70.7% area of the country), the location accuracy is better. The lowest location accuracy (>6 km) falls on the northwestern parts of the country, where the distances to the sensors located in the central-eastern part of the country are the highest (Fig. 1b).

The spatial differences in sensor density suggest that climatological results obtained in the coastal and western regions may be slightly affected by unequal detection efficiency, especially if we take into account low-peak current strokes (Mäkelä et al. 2014). In the climatological examinations the study area should be considered when determining the steady detection efficiency and the location accuracy ratio; however, no correction was made as we analyzed only the measured values.

The performance of the network did not change significantly over the analyzed time frame. In 2009 one sensor (in Toruń) was switched off because of the renovation. There were also some small changes in the configuration of the system (induced after the manufacturer’s recommendations) while in late 2009 the manufacturer’s support of SAFIR3000 sensor types was stopped. In 2012 IMGW–PIB managed to renovate sensors in the network and thus increased the quality of the lightning detections throughout the whole system. Although the detection efficiency and the location accuracy varied somewhat, these changes were small and did not significantly affect the detection efficiency or the location accuracy. The interannual changes in the average location accuracy deviated from the average value up to 17% in 2002 and 2004 while in the remaining years these deviations were lower than 10%.

c. Map computations

Diendorfer (2008) showed that a reliable accuracy for flash density can be achieved when on average of more than 80 flashes occurs in each grid cell. Therefore, most of the modern CG lightning flash climatologies typically use 10 km × 10 km, 20 km × 20 km, or 0.2° × 0.2° (approximately 20 km × 20 km) grid cells (e.g., Soriano et al. 2005; Sonnadara et al. 2006; Tuomi and Mäkelä 2008; Biron 2009; Antonescu and Burcea 2010; Mäkelä et al. 2011; Enno 2011; Santos et al. 2012). In this study we use a resolution of 10 km × 10 km for the grid cells (100 km2 area), and following the study of Diendorfer (2008), we believe that it is the most appropriate for our database (an average of 117 flashes in a grid cell).

To compute thunderstorm days derived from the PERUN database, we considered the flashes in the circle with the radius of 17.5 km from the center of the 1 km × 1 km grid cells. The higher resolution was chosen here to provide more-detailed and smoothed results. To compute such maps, it was necessary to include additional lightning data within a 17.5-km buffer zone away from the Polish borders. The same method used in this study to compute thunderstorm day characteristics has also been used in other studies (Novák and Kyznarová 2011; Wapler 2013; Mäkelä et al. 2014). We used the value of 17.5 km since for Poland it was proven by Czernecki et al. (2015) to provide the best overlap of thunderstorm days derived from the human observations with those estimated within the use of lightning detection data (at least two CG lightning flashes in the circle).

We used SYNOP reports from 44 meteorological stations derived from NOAA/National Climatic Data Center (NCDC) daily summaries for the same period as the data from the PERUN database (2002–13). In total, we distinguished 12 419 daily reports with thunderstorms (1478 unique days with thunderstorms).

3. Results

In this section we first present the statistics for the CG lightning flash data limited to the administrative borders of Poland and we also estimate the intensity of the thunderstorms. In section 3b, we present the spatial distribution of the lightning flash densities and thunderstorm days on different time scales, as these parameters have been frequently used in past CG lightning flash climatologies. Section 3c is devoted to the percentage of nighttime CG lightning flashes. The polarity and peak current characteristics are presented in section 3d.

a. Data statistics

In total, 4 952 203 CG lightning flashes were derived from the PERUN database for the years 2002–13 while 4 328 892 of the flashes were limited to the administrative borders of Poland in order to compute country-scale statistics (Table 1). Among these, almost 97% were negative CG lightning flashes while 3% corresponded to a positive lightning charge. With the use of latitude, longitude, date, and exact time of the detection (counted in seconds), we calculated the angle of the sun for each record and divided the data for detections during the daytime (sun angle ≥ −12°) and nighttime (sun angle < −12°). The value of 12° was used on the basis of NOAA’s astronomical term for nautical dawn [“This is the time at which the sun is 12 degrees below the horizon in the morning. Nautical dawn is defined as that time at which there is just enough sunlight for objects to be distinguishable;” NOAA/NWS (2015)] in order to focus on daytime and nighttime as it is perceived by the human eye. As it turned out, 85% of all flashes in our database were detected during the day while 15% occurred during the nighttime.

Table 1.

Statistics for 2002–13 CG lightning flashes derived from the PERUN lightning detection network. Data have been limited to the administrative borders of Poland.

Table 1.

To estimate the intensity of thunderstorms, we used daily sums of flashes. We did not include in the analysis days with only one detected lightning flash because of the possibility of false detection and thus unreliable climatological results (these kinds of single discharges may originate either from electromagnetic noise of anthropogenic origin or be lightning from large distances reflected by the ionosphere). Depending on the number of diurnal flashes, we have distinguished in our database thunderstorm days (lightning anywhere in Poland) with more than 1 flash (1815 days), 10 flashes (1354 days), 100 flashes (980 days), 1000 flashes (542 days), and 10 000 flashes (123 days), which gave an annual average of 151 days with a thunderstorm occurring anywhere in Poland (Table 2).

Table 2.

Statistics for days during 2002–13 with detected CG lightning flashes derived from the PERUN lightning detection network. Data have been limited to the administrative borders of Poland. Days with one detected lightning flash have been omitted.

Table 2.

There were 438 days during which the number of CG lightning flashes was between 101 and 1000. These days accounted for only 3.8% of all flashes in the database while 123 days with their daily CG lightning flashes counts exceeding 10 000 consisted represented 60% of all of the lightning data (Table 2). Moreover, thunderstorms on only the 10 days with the highest daily number of CG lightning flashes (Table 3) generated in total 545 071 CG lightning flashes, which accounted for 12.6% of the whole analyzed dataset. Six of these days appeared during the years 2011–13. At this point it is also worth mentioning that the majority of thunderstorm outbreaks occurred in July, the peak month (7 cases in the top 10 days and 41 cases in the top 100 days). The second-most intense month was June with 2 cases in the top 10 days and 26 in the top 100 days. August places third, with 1 case in top the 10 days and 22 in the top 100 days.

Table 3.

Top 10 days with the highest daily number of CG lightning flashes detected by the PERUN lightning detection network from 2002 to 2013. Data have been limited to the administrative borders of Poland.

Table 3.

b. Average annual number of thunderstorm days

The average annual number of thunderstorm days (Fig. 2b) with at least two CG lightning flashes increases generally from the northwest to the southeast with the lowest values along the Baltic Sea coast (15–20 days) and the highest in the Carpathian Mountains (30–35 days). The mean annual number of thunderstorm days averaged for the whole country amounted to 24.21 and was reasonably close to the value of 24.3 obtained from SYNOP reports for Poland by Kolendowicz (2012) through long-term observations (1951–2010). Similar results for the spatial distribution of thunderstorm days in the years 1885–2000 were also found by Bielec-Bąkowska (2003).

Fig. 2.
Fig. 2.

(a) Hypsometric map of Poland based on Shuttle Radar Topography Mission Global Coverage (SRTM3) data (Farr et al. 2007). (b) The average annual number of thunderstorm days during 2002–13. Lightning location data are computed within a radius of 17.5 km from the bin center (within a surface area of 962 km2) in 1 km × 1 km grid cells. (c) The average annual number of CG lightning flashes km−2. (d) The maximum daily number of CG lightning flashes km−2. Lightning densities are computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period 2002–13. Dots denote main meteorological stations (44).

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

c. Average annual CG lightning flash density

The spatial distribution of the mean CG lightning flash density over 10 km × 10 km grid cells varied from 0.2 to 3.1 flashes km−2 yr−1 (Fig. 2c). Similar densities were also observed in, for example, Austria, Spain, Romania, and the Czech Republic (Schulz et al. 2005; Soriano et al. 2005; Antonescu and Burcea 2010; Novák and Kyznarová 2011). In Poland, the lowest values were found over the Baltic Sea while the highest occurred in the middle-eastern part of the country [southwest (SW)–northeast (NE) belt from Kraków-Częstochowa Upland to Masurian Lake District; Figs. 2a,c]. Although the previous climatological study of Taszarek and Brooks (2015) pointed to exactly the same area as the most vulnerable for tornado occurrence, this pattern does not overlap with the spatial distribution of thunderstorm days. Such a distribution may be presumably correlated with PERUN’s spatial lightning detection efficiency, which is the highest in the middle-eastern part of the country and thus accounts for more flash detections (Bodzak 2006; Figs. 1a,b). However, the results of the European lightning density analyses obtained by Anderson and Klugmann (2014) through the use of ATDnet for the years 2008–13, by Pohjola and Mäkelä (2013) from GLD360 and EUCLID for the year 2011, and by the Blitzortung network (Wanke 2011) for the year 2011 pointed exactly to the same region in Poland where the highest CG lightning flash density occurred.

d. Maximum daily CG lightning flash density

The analysis of the maximum daily number of CG lightning flashes per kilometer squared revealed the places where the most intense or multiple thunderstorms occurred within one day (Fig. 2d). The maximum daily CG lightning flash density varied from 0.2 to 9.1 km−2 day−1 (3 July 2012; Table 3). This meant that very intense thunderstorms were capable of producing locally in only one day more CG lightning flashes that on average occur during the whole year (Fig. 2c). These kinds of storms are most likely related to mesoscale convective systems (MCSs; Houze 2004)—thunderstorms that are capable of producing large numbers of CG lightning flashes. The maximum daily density of the CG lightning flashes is the highest in the belt from Kraków-Częstochowa Upland to Masurian Lake District (Fig. 2a), similar to the average annual CG lightning flash density (Fig. 2c). Another similar pattern was also found in the large-hail report distribution in the study by Taszarek and Suwała (2015).

The spatial distribution of CG lightning flashes in the 10 days with the highest number of detected CG lightning flashes in our dataset (Table 3, Fig. 3) revealed that the thunderstorm activity was often covering more than half of the country, with 4–8 locations having CG lightning flash density exceeding 3–5 km−2 day−1. The maximum CG lightning flash density during these days varied from 3 to 9 km−2 day−1 (Table 3). These days also indicate that the belt from Kraków-Częstochowa Upland up to the Masurian Lake District is the most vulnerable region in terms of the occurrence of intense thunderstorms in Poland.

Fig. 3.
Fig. 3.

The number of CG lightning flashes km−2 in 10 days with the highest number of detected CG lightning flashes (as in Table 3). Computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period 2002–13.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

e. Annual variations in CG lightning flash density and thunderstorm days

The yearly variations in the total annual numbers of CG lightning flashes show a notable variability, with an average of 360 741 CG lightning flashes per year (Fig. 4) and a spatial average of 1.17 flashes km−2 yr−1 (Fig. 5). The years 2011 and 2012 were characterized by increased thunderstorm activity, with more than 500 000 CG lightning flashes per year (spatially 1.8 flashes km−2 yr−1). For thunderstorms, the years 2004 and 2009 were the least active and produced only 200 000 CG lightning flashes each year (spatially 0.6 flashes km−2 yr−1). Most of the annual peak CG lightning densities (>8 flashes km−2 yr−1) were observed in the middle and the eastern parts of the country. The highest value occurred in 2012 near the Masurian Lake District and exceeded 12 flashes km−2 yr−1.

Fig. 4.
Fig. 4.

Annual CG lightning flash count (bars), percentage of CG lightning flashes occurring during the night (black solid line), and percentage of positive CG lightning flashes (black dashed line). Based on lightning data derived from the PERUN network for the period 2002–13. Data have been limited to the administrative borders of Poland.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

Fig. 5.
Fig. 5.

Annual number of CG lightning flashes km−2 during the years 2002–13. Computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

The interannual number of thunderstorm days (occurring anywhere in Poland), with at least two flashes, varied from 133 (2007) to 171 (2002), with an average of 151 per year (Fig. 6). Human observations performed at meteorological stations indicated a lower frequency (an average of 123 days per year) and were similar to the data sample of thunderstorm days with at least 10 flashes (an average of 113 per year; Fig. 6).

Fig. 6.
Fig. 6.

Annual number of days with detected thunderstorms from SYNOP (44 stations) reports (bars), and annual number of days with the following number of CG lightning flashes detected: >1 (gray dotted line), >10 (gray dashed line), >100 (gray solid line), >1000 (black empty line), and >10 000 (black solid line). Based on lightning data derived from the PERUN network for the period 2002–13. Data have been limited to the administrative borders of Poland.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

The spatial distribution of thunderstorm days in particular years indicated that in almost every year more than 30 thunderstorm days occurred in the southern and southeastern parts of the country (Fig. 7). The highest value (58 days) observed in 2002 also occurred in the same region. The rest of the country was characterized by a greater differentiation, from 5 to even 50 thunderstorm days (Masurian Lake District in 2010). The annual number of thunderstorm days averaged across the whole area of the country amounted to 19.3 in 2008 with up to 27.5 in 2012 (Fig. 7). In the long-term climatology based on the SYNOP reports (Kolendowicz 2012; Czernecki et al. 2015), this value varied between 18 in 1976 and up to 32 in 1963 (based on the average from 44 meteorological stations). Bearing this in mind, the thunderstorm activity analyzed in this paper did not differ from the overall climatological values.

Fig. 7.
Fig. 7.

Annual number of thunderstorm days in 2002–13. Lightning location data are computed within a radius of 17.5 km from the bin center (within a surface area of 962 km2) in 1 km × 1 km grid cells. Based on lightning data derived from the PERUN network.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

f. Percentage of nighttime CG lightning flashes

Interesting results are found when we take into account the percentage of CG lightning flashes occurring during the night (sun angle < −12°; section 2d). We can observe an increase in CG lightning flashes occurring during the night from an average of 12% during the years 2002–07 to as much as 17% during the years 2008–13, with the peak in 2009 (22%; Fig. 4). It is difficult to explain such an increase, but since it is a percentage value, we doubt the fact that changes in network’s performance could affect it. It is possible that in these years conditions were more conducive for providing higher thermodynamic instability during the evening hours. Thunderstorm clouds could then last longer and produce more lightning during the night, especially in the form of MCSs (Nesbitt et al. 2000; Virts et al. 2013). The presence or absence of a few MCSs in a single year due to synoptic-scale factors that naturally vary from year to year could have dominated an average lightning count on many time and spatial scales.

The spatial distribution of the percentage of CG lightning flashes occurring during the nighttime varied in most of the area from 2% to 20% (Fig. 8). The exception was the midwestern and the southwestern parts of the country, where this value exceeded 45%. In the days with the highest number of CG lightning flashes in this region (e.g., 1 July 2012, 2 July 2012, 5 July 2012, 7 July 2012, 22 August 2012, 4 August 2013, and 29 July 2013), we can see that the majority of the flashes were produced by MCSs that were passing through this area during the nighttime hours (not shown). The majority of these cases had a very characteristic pattern. Thunderstorms during the daytime initiated over Germany and/or the Czech Republic and then moved northeasterly/easterly in the form of MCSs that were entering to the west and southwestern parts of Poland in the evening and nighttime hours. We also consider this pattern as one of the reasons why the annual percentage of CG lightning flashes during the nighttime has been higher in recent years (Fig. 4). MCSs that usually last until the late evening hours were more frequent in recent years and, thus, resulted in a higher percentage of nighttime flashes. The same effect was also observed in the studies of Nesbitt et al. (2000) and Virts et al. (2013), where long-lived MCSs were affecting the daily lightning cycle and increasing the percentage of CG lightning flashes during the nighttime.

Fig. 8.
Fig. 8.

The average percentage of CG lightning flashes occurring during the nighttime (sun angle <−12°; section 3a). Results are computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Dots denote the main meteorological stations (44).

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

The percentage of nighttime CG lightning flashes was highest during winter months and peaked in January (90%). The lowest percentage of nighttime flashes was in the early summer (11%), and from that point it was increasing each month through October (46%; Fig. 9). This could be justified by the advection of cold air masses during the late summer that over relatively preheated ground provided higher thermodynamic instability during the nighttime.

Fig. 9.
Fig. 9.

Monthly annual mean number of CG lightning flashes (bars), percentage of CG lightning flashes detected during the nighttime (black solid line), and percentage of positive CG lightning flashes (black dashed line). Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Data have been limited to the administrative borders of Poland.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

g. Monthly variations of CG lightning flash density and thunderstorm days

The monthly variations in CG lightning flash frequency clearly show a well-defined thunderstorm season extending from May to August with July as a peak month (an average of 137 024 CG lightning flashes per year with a maximum flash density in central Poland of up to 20 km−2 month−1; Figs. 9 and 10). Similar thunderstorm seasons have also been found in recent European CG lightning flash climatological studies (e.g., Antonescu and Burcea 2010; Novák and Kyznarová 2011; Feudale et al. 2013; Wapler 2013; Mäkelä et al. 2014). The spatial distribution of the CG lightning flash density in June–August was the highest in the belt from Kraków-Częstochowa Upland up to the Masurian Lake District.

Fig. 10.
Fig. 10.

Average monthly number of CG lightning flashes km−2 during the years 2002–13. Computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

The analysis of thunderstorm days reveals that the daily probability for thunderstorm occurrence anywhere in Poland from May to August exceeds 76% (>23 days with thunderstorm in each month; Fig. 11) with the July as a peak month (87%; 27 days). In April and September this probability decreases to 40% (12 days) while during cold months it varies from around 5% (February; 2 days) to 25% (October; 8 days).

Fig. 11.
Fig. 11.

Monthly mean number of days with a detected thunderstorm from SYNOP (44 stations) reports (bars), and the number of days with CG lightning flashes detected: >1 (gray dotted line), >10 (gray dashed line), >100 (gray solid line), >1000 (black empty line), and >10 000 (black solid line). Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Data have been limited to the administrative borders of Poland.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

While days with at least 2 CG lightning flashes occur all year round, the days with at least 10 CG lightning flashes occur mainly from March to October, and those with at least 100 CG lightning flashes occur from April to September. Most of the intense thunderstorm days with at least 10 000 CG lightning flashes appear during May–August and the number peaks in July (an average of 4.2 days yr−1), the most intense month (Fig. 11).

The monthly number of thunderstorm days averaged across the whole country (Fig. 12) varied from 0.03 (December) to 0.32 (October), thus giving daily probabilities for thunderstorm occurrences in a particular location (in a circle within the radius of 17.5 km, ~962 km2) from 0.1% to 1%. Substantially higher probabilities exceeding 14% (4.4 days) extended from May to August with the peak in July (21%, 6.6 days). In transitional months, the probability amounted to 3% in April (0.9 days) and 4% in September (1.2 days).

Fig. 12.
Fig. 12.

Average monthly number of thunderstorm days during the years 2002–13. Lightning location data are computed within a radius of 17.5 km from the bin center (within a surface area of 962 km2) for 1 km × 1 km grid cells. Based on lightning data derived from the PERUN network.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

From April to August thunderstorms were most frequent in the continental southeastern part of the country (Fig. 12). Studies by Riemann-Campe et al. (2009) and Brooks et al. (2003b) revealed that strong diurnal heating during these months and overlap with the boundary layer’s high moisture content often results in high convective available potential energy (CAPE) environments in this part of the country and, thus, provide good conditions for thunderstorms.

h. Hourly variations of CG lightning flashes

The hourly distribution of CG lightning flashes overlaps well with the diurnal cycle of convective activity, which generally depends on the boundary layer’s temperature and moisture content. The highest CG lightning flash activity peaks at 1400 and 1500 UTC (1600 and 1700 LT during summer months), whereas a flat minimum lies between 2300 and 0900 UTC (0100 and 1100 LT during summer months; Fig. 13). Exactly the same distribution was also found in other lightning climatological studies (e.g., Antonescu and Burcea 2010; Wapler 2013; Virts et al. 2013; Mäkelä et al. 2014).

Fig. 13.
Fig. 13.

Mean diurnal distribution of CG lightning flashes (percentage bars) with a time resolution of 1 h (UTC). Linear plots denote the diurnal distribution of CG lightning flashes on days when 2–10 (gray dotted line), 11–100 (gray dashed line), 101–1000 (gray solid line), 1001–10 000 (black empty line), and >10 000 (black solid line) flashes were detected. Based on lightning data derived from the PERUN network for the period 2002–13. Data have been limited to the administrative borders of Poland.

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

Regardless of the number of CG lightning flashes during the day, thunderstorm activity always starts to increase around 1000 UTC (1200 LT during summer months). We also note that the intense thunderstorms with more than 10 000 CG lightning flashes per day are still very active in the late evening hours [1700–2100 UTC (~1900–2300 LT) during summer months] while at the same time the activity of thunderstorms in days with fewer than 1000 CG lightning flashes sharply decreases (Fig. 13).

i. Polarity and peak current of CG lightning flashes

The percentage of positive CG lightning flashes was the lowest from May to October (around 2%–3%), while from November to April it ranged from 10% to 20% (Fig. 9). Positive flashes occur when the positive charge region in the thunderstorm cloud is closer to the ground compared to the typical scenario (Mäkelä et al. 2014). Therefore, shallow convection that is more likely during cold seasons may provide a higher percentage of positive CG lightning flashes (Williams 2001). A higher percentage of positive CG lightning flashes during the wintertime was also reported by Clodman and Chisholm (1996), Orville and Huffines (2001), Soriano et al. (2005), and Antonescu and Burcea (2010).

In a spatial sense, the percentage of positive CG lightning flashes varied in most of the area from 1% to 4% while in the northwestern, midwestern, and northeastern parts of the country it locally exceeded 6% (Fig. 14a). It is difficult to explain such a distribution; however, it is possible that during the analyzed period these areas experienced a few intense thunderstorms with inverted polarity (Williams 2001) that increased the percentage in a climatological sense. Orville and Silver (1997) pointed out that as the distance from a sensor of a CG lightning flash increases, only CG lightning flashes with higher peak current are detected. Therefore, positive CG lightning flashes that usually have a greater peak current than negative ones produce a higher percentage of CG-positive lightning flashes at long distances from a sensor. Such a pattern was found in Antonescu and Burcea (2010) but in our database this dependency only partially explained the spatial distribution of the positive CG lightning flash percentage.

Fig. 14.
Fig. 14.

(a) The average percentage of positive CG lightning flashes, and (b) the average peak current (kA) of negative CG lightning flashes. Results are computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Dots denote the main meteorological stations (44).

Citation: Monthly Weather Review 143, 11; 10.1175/MWR-D-15-0206.1

A high correlation with network’s lightning detection efficiency was found in a spatial pattern of an average negative peak current (Fig. 14b). Lowest values (15–25 kA) overlapped with the area of the high PERUN’s lightning detection efficiency (Fig. 1a). This can be explained by the fact that the closer the lightning is to the sensor, the lower the current of the lightning that can be detected will be (Orville and Silver 1997). The highest average peak current values (40–60 kA) that were observed along the coast of the Baltic Sea were presumably due to larger peak electric fields over the highly conductive sea surface (Orville et al. 2011).

In the monthly distribution of the average peak current of negative flashes (not shown) the highest values peaked in February (55 kA) with the lowest attributable to July (24 kA). The electrical field that initiates the lightning is presumably higher in lower temperatures and therefore produces on average flashes with higher peak currents (Brook 1992). The similar patterns in the monthly distribution of average peak current of negative flashes were also observed in the studies of Brook (1992), Orville and Huffines (2001), Soriano et al. (2005), and Antonescu and Burcea (2010).

4. Summary and final remarks

In contrast to SYNOP reports, CG lightning flash data provide a basis for climatologies that resolve variations in thunderstorm occurrence on different time scales with great accuracy. The main aim of this study was to present the first CG lightning flash climatology of Poland. Although PERUN’s lightning detection efficiency and the location accuracy are not homogenous in a spatial sense, the analysis of 4 952 203 CG lightning flashes derived from the PERUN database for the period 2002–13 yielded numerous conclusions. The most important are listed below.

  1. The average annual number of days with a thunderstorm at a particular location generally increases from the northwest to the southeast, with the lowest values along the coast of the Baltic Sea (15–20 days) and the highest in the Carpathian Mountains (30–35 days). This is consistent with studies forming the thunderstorm climatology of Poland that are based on SYNOP reports and long-term time frames (Bielec-Bąkowska 2003; Kolendowicz 2006, 2012).

  2. The annual average of 360 741 CG lightning flashes occurs each year over Polish territory. This results in 151 days with a thunderstorm appearing anywhere in Poland. Approximately 15% of all CG lightning flashes occur during nighttime hours while around 3% are CG lightning flashes with a positive current.

  3. An increase in CG lightning flashes occurring during the nighttime from an average of 12% in the years 2002–07 to as much as 17% in the years 2008–13 can be observed. This was presumably due to a more frequent occurrence of MCSs that in recent years (e.g., 1 July 2012, 2 July 2012, 5 July 2012, 7 July 2012, 22 August 2012, 4 August 2013, and 29 July 2013) produced large numbers of CG lightning flashes during the nighttime hours.

  4. The spatial distribution of the mean annual CG lightning flash density in 10 km × 10 km grid cells varied from 0.2 to 3.1 flashes km−2 yr−1, reaching its lowest values along the coast of the Baltic Sea and its highest in the SW–NE belt from Kraków-Częstochowa Upland to the Masurian Lake District. Although this region partially coincides with the PERUN network’s lightning detection efficiency, the same area with peak lightning density was obtained in the studies of Pohjola and Mäkelä (2013) and Anderson and Klugmann (2014).

  5. The maximum daily CG lightning flash density varied from 0.2 to 9.1 km−2 day−1 and meant that very intense thunderstorms were capable of producing locally in only one day more CG lightning flashes that on average occur during the whole year. The highest values of maximum daily CG lightning flash density were observed in the central and eastern parts of the country. The day with the highest number of CG lightning flashes during the whole analyzed period was 26 June 2006 (73 549 flashes).

  6. The monthly variation in CG lightning flash frequency clearly showed a well-defined thunderstorm season extending from May to August with July as a peak month (an average of 137 024 CG lightning flashes per year with the maximum flash density in central Poland of up to 20 flashes per km−2 month−1). The days with the most intense thunderstorms (days with over 10 000 CG lightning flashes) occur from May to August and peak in July (4.2 days month−1) as the most intense month.

  7. The vast majority of CG lightning flashes were detected during the daytime with the peak at 1400 UTC and the minimum at 0700 UTC. It was also noticed that the intense thunderstorms in days with more than 10 000 CG lightning flashes were still very active in the late evening hours (1700–2100 UTC) while at the same time the activity of thunderstorms in days with less than 1000 CG lightning flashes was sharply decreasing.

  8. Almost 97% of all CG lightning flashes in our study had a negative current, reaching the highest average monthly values in February (55 kA) and the lowest in July (24 kA). The percentage of positive CG lightning flashes was the lowest from May to October (2%–3%), while from November to April the percentage ranged from 10% to 20%.

  9. Compared to thunderstorm statistics in other parts of Europe, Polish thunderstorms have many similar features. The diurnal CG lightning flash peak around 1400 UTC with the thunderstorm high season extending from May to August was also found in CG lightning flash climatologies of Austria (Schulz et al. 2005), Spain (Soriano et al. 2005), Romania (Antonescu and Burcea 2010), Estonia (Enno 2011), the Czech Republic (Novák and Kyznarová, 2011), Germany (Wapler 2013), and Scandinavia (Mäkelä et al. 2014). Conversely, over the Mediterranean (especially the eastern part) the thunderstorm season shifts toward cold season months (October–March) with increased CG lightning flash activity during the nighttime hours (Altaratz et al. 2003; Virts et al. 2013).

Further research into this topic is necessary, especially concerning the atmospheric conditions during the days with the most intense thunderstorms.

Acknowledgments

We thank the Polish Institute of Meteorology and Water Management–National Research Institute for providing data from the PERUN lightning detection network and allowing us to perform such a study. We also appreciate the comments of anonymous reviewers who helped to improve the study. This research was partly supported by the grant of Polish National Science Centre (UMO-2014/13/N/ST10/01708).

REFERENCES

  • Altaratz, O., Z. Levin, Y. Yair, and B. Ziv, 2003: Lightning activity over land and sea on the eastern coast of the Mediterranean. Mon. Wea. Rev., 131, 20602070, doi:10.1175/1520-0493(2003)131<2060:LAOLAS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Anderson, G., and D. Klugmann, 2014: A European lightning density analysis using 5 years of ATDnet data. Nat. Hazards Earth Syst. Sci., 14, 815829, doi:10.5194/nhess-14-815-2014.

    • Search Google Scholar
    • Export Citation
  • Antonescu, B., and S. Burcea, 2010: A cloud-to-ground lightning climatology for Romania. Mon. Wea. Rev., 138, 579591, doi:10.1175/2009MWR2975.1.

    • Search Google Scholar
    • Export Citation
  • Betz, H. D., K. Schmidt, P. Laroche, P. Blanchet, W. P. Oettinger, E. Defer, Z. Dziewit, and J. Konarski, 2009: LINET—An international lightning detection network in Europe. Atmos. Res., 91, 564573, doi:10.1016/j.atmosres.2008.06.012.

    • Search Google Scholar
    • Export Citation
  • Bielec-Bąkowska, Z., 2003: Long-term variability of thunderstorm occurrence in Poland in the 20th century. Atmos. Res., 67–68, 3552, doi:10.1016/S0169-8095(03)00082-6.

    • Search Google Scholar
    • Export Citation
  • Biron, D., 2009: LAMPINET–Lightning detection in Italy. Lightning: Principles, Instruments and Applications Review of Modern Lightning Research, H. D. Betz, U. Schumann, and P. Laroche, Eds., Springer, 141–159.

  • Bodzak, P., 2006: Detekcja i Lokalizacja Wyładowań Atmosferycznych (Lightning Detection and Localization). Instytut Meteorologii i Gospodarki Wodnej, 135 pp.

  • Brook, M., 1992: Breakdown of electric fields in winter storms. Res. Lett. Atmos. Electr., 12, 4752.

  • Brooks, H. E., C. A. Doswell III, and M. P. Kay, 2003a: Climatological estimates of local daily tornado probability for the United States. Wea. Forecasting, 18, 626640, doi:10.1175/1520-0434(2003)018<0626:CEOLDT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brooks, H. E., J. W. Lee, and J. P. Craven, 2003b: The spatial distribution of severe thunderstorm and tornado environments from global reanalysis data. Atmos. Res., 67–68, 7394, doi:10.1016/S0169-8095(03)00045-0.

    • Search Google Scholar
    • Export Citation
  • Changnon, S. A., and D. Changnon, 2001: Long-term fluctuations in the thunderstorm activity in the United States. Climatic Change, 50, 489503, doi:10.1023/A:1010651512934.

    • Search Google Scholar
    • Export Citation
  • Clodman, S., and W. Chisholm, 1996: Lightning flash climatology in the southern Great Lakes region. Atmos.–Ocean, 34, 345377, doi:10.1080/07055900.1996.9649568.

    • Search Google Scholar
    • Export Citation
  • Cummins, K. L., M. J. Murphy, E. A. Bardo, W. L. Hiscox, R. B. Pyle, and A. E. Pifer, 1998: A combined TOA/MDF technology upgrade of the U.S. National Lightning Detection Network. J. Geophys. Res., 103, 90359044, doi:10.1029/98JD00153.

    • Search Google Scholar
    • Export Citation
  • Curran, E. B., R. L. Holle, and R. E. López, 2000: Lightning casualties and damages in the United States from 1959 to 1994. J. Climate, 13, 34483464, doi:10.1175/1520-0442(2000)013<3448:LCADIT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Czernecki, B., M. Taszarek, L. Kolendowicz, and J. Konarski, 2015: Relationship between human observations of thunderstorms and PERUN lightning detection network in Poland. Atmos. Res., 167, 118–128, doi:10.1016/j.atmosres.2015.08.003.

    • Search Google Scholar
    • Export Citation
  • Diendorfer, G., 2008: Some comments on the achievable accuracy of local ground flash density values. Proc. 29th Int. Conf. on Lightning Protection. Uppsala, Sweden, ICLP Centre, 2-8-1–2-8-6.

  • Dotzek, N., P. H. Groenemeijer, B. Feuerstein, and A. M. Holzer, 2009: Overview of ESSL’s severe convective storms research using the European Severe Weather Database ESWD. Atmos. Res., 93, 575586, doi:10.1016/j.atmosres.2008.10.020.

    • Search Google Scholar
    • Export Citation
  • Enno, S. E., 2011: A climatology of cloud-to-ground lightning over Estonia, 2005–2009. Atmos. Res., 100, 310317, doi:10.1016/j.atmosres.2010.08.024.

    • Search Google Scholar
    • Export Citation
  • Farr, T. G., and Coauthors, 2007: The Shuttle Radar Topography Mission. Rev. Geophys., 45, RG2004, doi:10.1029/2005RG000183.

  • Feudale, L., A. Manzato, and S. Micheletti, 2013: A cloud-to-ground lightning climatology for north-eastern Italy. Adv. Sci. Res., 10, 7784, doi:10.5194/asr-10-77-2013.

    • Search Google Scholar
    • Export Citation
  • Groenemeijer, P. H., N. Dotzek, F. Stel, H. E. Brooks, C. A. Doswell III, and D. M. Elsom, 2004: ESWD—A standardized, flexible data format for severe weather reports. Preprints, Third European Conf. on Severe Storms, León, Spain, European Severe Storms Laboratory. [Available online at https://www.researchgate.net/publication/224780785_ESWD_-_A_Standardized_Flexible_Data_Format_for_Severe_Weather_Reports.]

  • Holle, R. L., R. E. López, and C. Zimmermann, 1999: Updated recommendations for lightning safety—1998. Bull. Amer. Meteor. Soc., 80, 20352041, doi:10.1175/1520-0477(1999)080<2035:URFLS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Houze, R. A., Jr., 2004: Mesoscale convective systems. Rev. Geophys., 42, RG4003, doi:10.1029/2004RG000150.

  • Kolendowicz, L., 2006: The influence of synoptic situations on the occurrence of days with thunderstorms during a year in the territory of Poland. Int. J. Climatol., 26, 18031820, doi:10.1002/joc.1348.

    • Search Google Scholar
    • Export Citation
  • Kolendowicz, L., 2012: Synoptic patterns associated with thunderstorms in Poland. Meteor. Z., 21, 145156, doi:10.1127/0941-2948/2012/0272.

    • Search Google Scholar
    • Export Citation
  • Krider, E. P., R. C. Noggle, A. E. Pifer, and D. L. Vance, 1980: Lightning direction-finding systems for forest fire detection. Bull. Amer. Meteor. Soc., 61, 980986, doi:10.1175/1520-0477(1980)061<0980:LDFSFF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Larjavaara, M., T. Kuuluvainena, and H. Rita, 2005: Spatial distribution of lightning-ignited forest fires in Finland. For. Ecol. Manage., 208, 177188, doi:10.1016/j.foreco.2004.12.005.

    • Search Google Scholar
    • Export Citation
  • Łoboda, M., H. D. Betz, P. Baranski, J. Wiszniowski, and Z. Dziewit, 2009: New lightning detection networks in Poland—LINET and LLDN. Open Atmos. Sci. J., 3, 29–38, doi:10.2174/1874282300903010029.

    • Search Google Scholar
    • Export Citation
  • MacGorman, D. R., and W. D. Rust, 1998: The Electrical Nature of Storms. Oxford University Press, 432 pp.

  • Mäkelä, A., P. Rossi, and D. M. Schultz, 2011: The daily cloud-to-ground lightning flash density in the contiguous United States and Finland. Mon. Wea. Rev., 139, 13231337, doi:10.1175/2010MWR3517.1.

    • Search Google Scholar
    • Export Citation
  • Mäkelä, A., E. Saltikoff, J. Julkunen, I. Juga, E. Gregow, and S. Niemelä, 2013: Cold-season thunderstorms in Finland and their effect on aviation safety. Bull. Amer. Meteor. Soc., 94, 847858, doi:10.1175/BAMS-D-12-00039.1.

    • Search Google Scholar
    • Export Citation
  • Mäkelä, A., S. E. Enno, and J. Haapalainen, 2014: Nordic Lightning Information System: Thunderstorm climate of northern Europe for the period 2002–2011. Atmos. Res., 139, 4661, doi:10.1016/j.atmosres.2014.01.008.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., E. J. Zipser, and D. J. Cecil, 2000: A census of precipitation features in the tropics using TRMM: Radar, ice scattering, and lightning observations. J. Climate, 13, 40874106, doi:10.1175/1520-0442(2000)013<4087:ACOPFI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • NOAA/NWS, 2015: Nautical dawn. NOAA/NWS Glossary. [Available online at http://w1.weather.gov/glossary/index.php?word=nautical+dawn.]

  • Novák, P., and H. Kyznarová, 2011: Climatology of lightning in the Czech Republic. Atmos. Res., 100, 318333, doi:10.1016/j.atmosres.2010.08.022.

    • Search Google Scholar
    • Export Citation
  • Orville, R. E., and A. C. Silver, 1997: Lightning ground flash density in the contiguous United States: 1992–95. Mon. Wea. Rev., 125, 631638, doi:10.1175/1520-0493(1997)125<0631:LGFDIT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Orville, R. E., and G. R. Huffines, 2001: Cloud-to-ground lightning in the United States: NLDN results in the first decade, 1989–98. Mon. Wea. Rev., 129, 11791193, doi:10.1175/1520-0493(2001)129<1179:CTGLIT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Orville, R. E., G. R. Huffines, W. R. Burrows, and K. L. Cummins, 2011: The North American Lightning Detection Network (NALDN) analysis of flash data: 2001–09. Mon. Wea. Rev., 139, 13051322, doi:10.1175/2010MWR3452.1.

    • Search Google Scholar
    • Export Citation
  • Pohjola, H., and A. Mäkelä, 2013: The comparison of GLD360 and EUCLID lightning location systems in Europe. Atmos. Res., 123, 117128, doi:10.1016/j.atmosres.2012.10.019.

    • Search Google Scholar
    • Export Citation
  • Rakov, V. A., and M. A. Uman, 2003: Lightning: Physics and Effects. Cambridge University Press, 687 pp.

  • R Core Team, 2014: R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. [Available online at http://www.R-project.org/.]

  • Riemann-Campe, K., K. Fraedrich, and F. Lunkeit, 2009: Global climatology of convective available potential energy (CAPE) and convective inhibition (CIN) in ERA-40 re-analysis. Atmos. Res., 93, 534545, doi:10.1016/j.atmosres.2008.09.037.

    • Search Google Scholar
    • Export Citation
  • Santos, J. A., M. A. Reis, J. Sousa, S. M. Leite, S. Correia, M. Janeira, and M. Fragoso, 2012: Cloud-to-ground lightning in Portugal: Patterns and dynamical forcing. Nat. Hazards Earth Syst. Sci., 12, 639649, doi:10.5194/nhess-12-639-2012.

    • Search Google Scholar
    • Export Citation
  • Sasse, M., and T. Hauf, 2003: A study of thunderstorm-induced delays at Frankfurt airport, Germany. Meteor. Appl., 10, 2130, doi:10.1017/S1350482703005036.

    • Search Google Scholar
    • Export Citation
  • Schulz, W., K. Cummins, G. Diendorfer, and M. Dorninger, 2005: Cloud-to-ground lightning in Austria: A 10-year study using data from a lightning location system. J. Geophys. Res., 110, D09101, doi:10.1029/2004JD005332.

    • Search Google Scholar
    • Export Citation
  • Sonnadara, U., V. Cooray, and T. Götschl, 2006: Characteristics of cloud-to-ground lightning flashes over Sweden. Phys. Scr., 74, 541, doi:10.1088/0031-8949/74/5/010.

    • Search Google Scholar
    • Export Citation
  • Soriano, L. R., F. de Pablo, and C. Tomas, 2005: Ten-year study of cloud-to ground lightning activity in the Iberian Peninsula. J. Atmos. Solar-Terr. Phys., 67, 1632–1639, doi:10.1016/j.jastp.2005.08.019.

    • Search Google Scholar
    • Export Citation
  • Taszarek, M., and H. E. Brooks, 2015: Tornado climatology of Poland. Mon. Wea. Rev., 143, 702717, doi:10.1175/MWR-D-14-00185.1.

  • Taszarek, M., and K. Suwała, 2015: Large hail in Poland in 2012. Quaestiones Geogr., 34, 7584, doi:10.1515/quageo-2015-0007.

  • Tuomi, T. J., and A. Mäkelä, 2008: Thunderstorm climate of Finland 1998–2007. Geophysica, 44, 6780.

  • Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H. Holzworth, 2013: Highlights of a new ground-based, hourly global lightning climatology. Bull. Amer. Meteor. Soc., 94, 13811391, doi:10.1175/BAMS-D-12-00082.1.

    • Search Google Scholar
    • Export Citation
  • Wacker, R. S., and R. E. Orville, 1999a: Changes in measured lightning flash count and return stroke peak current after the 1994 U.S. National Lightning Detection Network upgrade. 1. Observations. J. Geophys. Res., 104, 21512157, doi:10.1029/1998JD200060.

    • Search Google Scholar
    • Export Citation
  • Wacker, R. S., and R. E. Orville, 1999b: Changes in measured lightning flash count and return stroke peak current after the 1994 U. S. National Lightning Detection Network upgrade. II. Theory. J. Geophys. Res., 104, 21592162, doi:10.1029/1998JD200059.

    • Search Google Scholar
    • Export Citation
  • Wanke, E., 2011: Blitzortung.org—A low cost time of arrival lightning detection and lightning location network. Universität Düsseldorf, 75 pp. [Available online at http://www.blitzortung.org/Documents/TOA_Blitzortung.pdf.]

  • Wapler, K., 2013: High-resolution climatology of lightning characteristics within Central Europe. Meteor. Atmos. Phys., 122, 175184, doi:10.1007/s00703-013-0285-1.

    • Search Google Scholar
    • Export Citation
  • Wierzchowski, J., M. Heathcott, and M. D. Flannigan, 2002: Lightning and lightning fire, central cordillera, Canada. Int. J. Wildland Fire, 11, 4151, doi:10.1071/WF01048.

    • Search Google Scholar
    • Export Citation
  • Williams, E. R., 2001: The electrification of severe storms. Severe Convective Storms, Meteor. Monogr., No. 50, 527–528, doi:10.1175/0065-9401-28.50.527.

Save
  • Altaratz, O., Z. Levin, Y. Yair, and B. Ziv, 2003: Lightning activity over land and sea on the eastern coast of the Mediterranean. Mon. Wea. Rev., 131, 20602070, doi:10.1175/1520-0493(2003)131<2060:LAOLAS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Anderson, G., and D. Klugmann, 2014: A European lightning density analysis using 5 years of ATDnet data. Nat. Hazards Earth Syst. Sci., 14, 815829, doi:10.5194/nhess-14-815-2014.

    • Search Google Scholar
    • Export Citation
  • Antonescu, B., and S. Burcea, 2010: A cloud-to-ground lightning climatology for Romania. Mon. Wea. Rev., 138, 579591, doi:10.1175/2009MWR2975.1.

    • Search Google Scholar
    • Export Citation
  • Betz, H. D., K. Schmidt, P. Laroche, P. Blanchet, W. P. Oettinger, E. Defer, Z. Dziewit, and J. Konarski, 2009: LINET—An international lightning detection network in Europe. Atmos. Res., 91, 564573, doi:10.1016/j.atmosres.2008.06.012.

    • Search Google Scholar
    • Export Citation
  • Bielec-Bąkowska, Z., 2003: Long-term variability of thunderstorm occurrence in Poland in the 20th century. Atmos. Res., 67–68, 3552, doi:10.1016/S0169-8095(03)00082-6.

    • Search Google Scholar
    • Export Citation
  • Biron, D., 2009: LAMPINET–Lightning detection in Italy. Lightning: Principles, Instruments and Applications Review of Modern Lightning Research, H. D. Betz, U. Schumann, and P. Laroche, Eds., Springer, 141–159.

  • Bodzak, P., 2006: Detekcja i Lokalizacja Wyładowań Atmosferycznych (Lightning Detection and Localization). Instytut Meteorologii i Gospodarki Wodnej, 135 pp.

  • Brook, M., 1992: Breakdown of electric fields in winter storms. Res. Lett. Atmos. Electr., 12, 4752.

  • Brooks, H. E., C. A. Doswell III, and M. P. Kay, 2003a: Climatological estimates of local daily tornado probability for the United States. Wea. Forecasting, 18, 626640, doi:10.1175/1520-0434(2003)018<0626:CEOLDT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brooks, H. E., J. W. Lee, and J. P. Craven, 2003b: The spatial distribution of severe thunderstorm and tornado environments from global reanalysis data. Atmos. Res., 67–68, 7394, doi:10.1016/S0169-8095(03)00045-0.

    • Search Google Scholar
    • Export Citation
  • Changnon, S. A., and D. Changnon, 2001: Long-term fluctuations in the thunderstorm activity in the United States. Climatic Change, 50, 489503, doi:10.1023/A:1010651512934.

    • Search Google Scholar
    • Export Citation
  • Clodman, S., and W. Chisholm, 1996: Lightning flash climatology in the southern Great Lakes region. Atmos.–Ocean, 34, 345377, doi:10.1080/07055900.1996.9649568.

    • Search Google Scholar
    • Export Citation
  • Cummins, K. L., M. J. Murphy, E. A. Bardo, W. L. Hiscox, R. B. Pyle, and A. E. Pifer, 1998: A combined TOA/MDF technology upgrade of the U.S. National Lightning Detection Network. J. Geophys. Res., 103, 90359044, doi:10.1029/98JD00153.

    • Search Google Scholar
    • Export Citation
  • Curran, E. B., R. L. Holle, and R. E. López, 2000: Lightning casualties and damages in the United States from 1959 to 1994. J. Climate, 13, 34483464, doi:10.1175/1520-0442(2000)013<3448:LCADIT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Czernecki, B., M. Taszarek, L. Kolendowicz, and J. Konarski, 2015: Relationship between human observations of thunderstorms and PERUN lightning detection network in Poland. Atmos. Res., 167, 118–128, doi:10.1016/j.atmosres.2015.08.003.

    • Search Google Scholar
    • Export Citation
  • Diendorfer, G., 2008: Some comments on the achievable accuracy of local ground flash density values. Proc. 29th Int. Conf. on Lightning Protection. Uppsala, Sweden, ICLP Centre, 2-8-1–2-8-6.

  • Dotzek, N., P. H. Groenemeijer, B. Feuerstein, and A. M. Holzer, 2009: Overview of ESSL’s severe convective storms research using the European Severe Weather Database ESWD. Atmos. Res., 93, 575586, doi:10.1016/j.atmosres.2008.10.020.

    • Search Google Scholar
    • Export Citation
  • Enno, S. E., 2011: A climatology of cloud-to-ground lightning over Estonia, 2005–2009. Atmos. Res., 100, 310317, doi:10.1016/j.atmosres.2010.08.024.

    • Search Google Scholar
    • Export Citation
  • Farr, T. G., and Coauthors, 2007: The Shuttle Radar Topography Mission. Rev. Geophys., 45, RG2004, doi:10.1029/2005RG000183.

  • Feudale, L., A. Manzato, and S. Micheletti, 2013: A cloud-to-ground lightning climatology for north-eastern Italy. Adv. Sci. Res., 10, 7784, doi:10.5194/asr-10-77-2013.

    • Search Google Scholar
    • Export Citation
  • Groenemeijer, P. H., N. Dotzek, F. Stel, H. E. Brooks, C. A. Doswell III, and D. M. Elsom, 2004: ESWD—A standardized, flexible data format for severe weather reports. Preprints, Third European Conf. on Severe Storms, León, Spain, European Severe Storms Laboratory. [Available online at https://www.researchgate.net/publication/224780785_ESWD_-_A_Standardized_Flexible_Data_Format_for_Severe_Weather_Reports.]

  • Holle, R. L., R. E. López, and C. Zimmermann, 1999: Updated recommendations for lightning safety—1998. Bull. Amer. Meteor. Soc., 80, 20352041, doi:10.1175/1520-0477(1999)080<2035:URFLS>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Houze, R. A., Jr., 2004: Mesoscale convective systems. Rev. Geophys., 42, RG4003, doi:10.1029/2004RG000150.

  • Kolendowicz, L., 2006: The influence of synoptic situations on the occurrence of days with thunderstorms during a year in the territory of Poland. Int. J. Climatol., 26, 18031820, doi:10.1002/joc.1348.

    • Search Google Scholar
    • Export Citation
  • Kolendowicz, L., 2012: Synoptic patterns associated with thunderstorms in Poland. Meteor. Z., 21, 145156, doi:10.1127/0941-2948/2012/0272.

    • Search Google Scholar
    • Export Citation
  • Krider, E. P., R. C. Noggle, A. E. Pifer, and D. L. Vance, 1980: Lightning direction-finding systems for forest fire detection. Bull. Amer. Meteor. Soc., 61, 980986, doi:10.1175/1520-0477(1980)061<0980:LDFSFF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Larjavaara, M., T. Kuuluvainena, and H. Rita, 2005: Spatial distribution of lightning-ignited forest fires in Finland. For. Ecol. Manage., 208, 177188, doi:10.1016/j.foreco.2004.12.005.

    • Search Google Scholar
    • Export Citation
  • Łoboda, M., H. D. Betz, P. Baranski, J. Wiszniowski, and Z. Dziewit, 2009: New lightning detection networks in Poland—LINET and LLDN. Open Atmos. Sci. J., 3, 29–38, doi:10.2174/1874282300903010029.

    • Search Google Scholar
    • Export Citation
  • MacGorman, D. R., and W. D. Rust, 1998: The Electrical Nature of Storms. Oxford University Press, 432 pp.

  • Mäkelä, A., P. Rossi, and D. M. Schultz, 2011: The daily cloud-to-ground lightning flash density in the contiguous United States and Finland. Mon. Wea. Rev., 139, 13231337, doi:10.1175/2010MWR3517.1.

    • Search Google Scholar
    • Export Citation
  • Mäkelä, A., E. Saltikoff, J. Julkunen, I. Juga, E. Gregow, and S. Niemelä, 2013: Cold-season thunderstorms in Finland and their effect on aviation safety. Bull. Amer. Meteor. Soc., 94, 847858, doi:10.1175/BAMS-D-12-00039.1.

    • Search Google Scholar
    • Export Citation
  • Mäkelä, A., S. E. Enno, and J. Haapalainen, 2014: Nordic Lightning Information System: Thunderstorm climate of northern Europe for the period 2002–2011. Atmos. Res., 139, 4661, doi:10.1016/j.atmosres.2014.01.008.

    • Search Google Scholar
    • Export Citation
  • Nesbitt, S. W., E. J. Zipser, and D. J. Cecil, 2000: A census of precipitation features in the tropics using TRMM: Radar, ice scattering, and lightning observations. J. Climate, 13, 40874106, doi:10.1175/1520-0442(2000)013<4087:ACOPFI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • NOAA/NWS, 2015: Nautical dawn. NOAA/NWS Glossary. [Available online at http://w1.weather.gov/glossary/index.php?word=nautical+dawn.]

  • Novák, P., and H. Kyznarová, 2011: Climatology of lightning in the Czech Republic. Atmos. Res., 100, 318333, doi:10.1016/j.atmosres.2010.08.022.

    • Search Google Scholar
    • Export Citation
  • Orville, R. E., and A. C. Silver, 1997: Lightning ground flash density in the contiguous United States: 1992–95. Mon. Wea. Rev., 125, 631638, doi:10.1175/1520-0493(1997)125<0631:LGFDIT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Orville, R. E., and G. R. Huffines, 2001: Cloud-to-ground lightning in the United States: NLDN results in the first decade, 1989–98. Mon. Wea. Rev., 129, 11791193, doi:10.1175/1520-0493(2001)129<1179:CTGLIT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Orville, R. E., G. R. Huffines, W. R. Burrows, and K. L. Cummins, 2011: The North American Lightning Detection Network (NALDN) analysis of flash data: 2001–09. Mon. Wea. Rev., 139, 13051322, doi:10.1175/2010MWR3452.1.

    • Search Google Scholar
    • Export Citation
  • Pohjola, H., and A. Mäkelä, 2013: The comparison of GLD360 and EUCLID lightning location systems in Europe. Atmos. Res., 123, 117128, doi:10.1016/j.atmosres.2012.10.019.

    • Search Google Scholar
    • Export Citation
  • Rakov, V. A., and M. A. Uman, 2003: Lightning: Physics and Effects. Cambridge University Press, 687 pp.

  • R Core Team, 2014: R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. [Available online at http://www.R-project.org/.]

  • Riemann-Campe, K., K. Fraedrich, and F. Lunkeit, 2009: Global climatology of convective available potential energy (CAPE) and convective inhibition (CIN) in ERA-40 re-analysis. Atmos. Res., 93, 534545, doi:10.1016/j.atmosres.2008.09.037.

    • Search Google Scholar
    • Export Citation
  • Santos, J. A., M. A. Reis, J. Sousa, S. M. Leite, S. Correia, M. Janeira, and M. Fragoso, 2012: Cloud-to-ground lightning in Portugal: Patterns and dynamical forcing. Nat. Hazards Earth Syst. Sci., 12, 639649, doi:10.5194/nhess-12-639-2012.

    • Search Google Scholar
    • Export Citation
  • Sasse, M., and T. Hauf, 2003: A study of thunderstorm-induced delays at Frankfurt airport, Germany. Meteor. Appl., 10, 2130, doi:10.1017/S1350482703005036.

    • Search Google Scholar
    • Export Citation
  • Schulz, W., K. Cummins, G. Diendorfer, and M. Dorninger, 2005: Cloud-to-ground lightning in Austria: A 10-year study using data from a lightning location system. J. Geophys. Res., 110, D09101, doi:10.1029/2004JD005332.

    • Search Google Scholar
    • Export Citation
  • Sonnadara, U., V. Cooray, and T. Götschl, 2006: Characteristics of cloud-to-ground lightning flashes over Sweden. Phys. Scr., 74, 541, doi:10.1088/0031-8949/74/5/010.

    • Search Google Scholar
    • Export Citation
  • Soriano, L. R., F. de Pablo, and C. Tomas, 2005: Ten-year study of cloud-to ground lightning activity in the Iberian Peninsula. J. Atmos. Solar-Terr. Phys., 67, 1632–1639, doi:10.1016/j.jastp.2005.08.019.

    • Search Google Scholar
    • Export Citation
  • Taszarek, M., and H. E. Brooks, 2015: Tornado climatology of Poland. Mon. Wea. Rev., 143, 702717, doi:10.1175/MWR-D-14-00185.1.

  • Taszarek, M., and K. Suwała, 2015: Large hail in Poland in 2012. Quaestiones Geogr., 34, 7584, doi:10.1515/quageo-2015-0007.

  • Tuomi, T. J., and A. Mäkelä, 2008: Thunderstorm climate of Finland 1998–2007. Geophysica, 44, 6780.

  • Virts, K. S., J. M. Wallace, M. L. Hutchins, and R. H. Holzworth, 2013: Highlights of a new ground-based, hourly global lightning climatology. Bull. Amer. Meteor. Soc., 94, 13811391, doi:10.1175/BAMS-D-12-00082.1.

    • Search Google Scholar
    • Export Citation
  • Wacker, R. S., and R. E. Orville, 1999a: Changes in measured lightning flash count and return stroke peak current after the 1994 U.S. National Lightning Detection Network upgrade. 1. Observations. J. Geophys. Res., 104, 21512157, doi:10.1029/1998JD200060.

    • Search Google Scholar
    • Export Citation
  • Wacker, R. S., and R. E. Orville, 1999b: Changes in measured lightning flash count and return stroke peak current after the 1994 U. S. National Lightning Detection Network upgrade. II. Theory. J. Geophys. Res., 104, 21592162, doi:10.1029/1998JD200059.

    • Search Google Scholar
    • Export Citation
  • Wanke, E., 2011: Blitzortung.org—A low cost time of arrival lightning detection and lightning location network. Universität Düsseldorf, 75 pp. [Available online at http://www.blitzortung.org/Documents/TOA_Blitzortung.pdf.]

  • Wapler, K., 2013: High-resolution climatology of lightning characteristics within Central Europe. Meteor. Atmos. Phys., 122, 175184, doi:10.1007/s00703-013-0285-1.

    • Search Google Scholar
    • Export Citation
  • Wierzchowski, J., M. Heathcott, and M. D. Flannigan, 2002: Lightning and lightning fire, central cordillera, Canada. Int. J. Wildland Fire, 11, 4151, doi:10.1071/WF01048.

    • Search Google Scholar
    • Export Citation
  • Williams, E. R., 2001: The electrification of severe storms. Severe Convective Storms, Meteor. Monogr., No. 50, 527–528, doi:10.1175/0065-9401-28.50.527.

  • Fig. 1.

    (a) Locations of SAFIR3000 lightning sensors in the PERUN network with 100-km buffer zones. (b) Average CG lightning flash location accuracy (km) derived from the PERUN database during 2002–13. Computed in 10 km × 10 km grid cells. Dots denote main meteorological stations (44).

  • Fig. 2.

    (a) Hypsometric map of Poland based on Shuttle Radar Topography Mission Global Coverage (SRTM3) data (Farr et al. 2007). (b) The average annual number of thunderstorm days during 2002–13. Lightning location data are computed within a radius of 17.5 km from the bin center (within a surface area of 962 km2) in 1 km × 1 km grid cells. (c) The average annual number of CG lightning flashes km−2. (d) The maximum daily number of CG lightning flashes km−2. Lightning densities are computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period 2002–13. Dots denote main meteorological stations (44).

  • Fig. 3.

    The number of CG lightning flashes km−2 in 10 days with the highest number of detected CG lightning flashes (as in Table 3). Computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period 2002–13.

  • Fig. 4.

    Annual CG lightning flash count (bars), percentage of CG lightning flashes occurring during the night (black solid line), and percentage of positive CG lightning flashes (black dashed line). Based on lightning data derived from the PERUN network for the period 2002–13. Data have been limited to the administrative borders of Poland.

  • Fig. 5.

    Annual number of CG lightning flashes km−2 during the years 2002–13. Computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network.

  • Fig. 6.

    Annual number of days with detected thunderstorms from SYNOP (44 stations) reports (bars), and annual number of days with the following number of CG lightning flashes detected: >1 (gray dotted line), >10 (gray dashed line), >100 (gray solid line), >1000 (black empty line), and >10 000 (black solid line). Based on lightning data derived from the PERUN network for the period 2002–13. Data have been limited to the administrative borders of Poland.

  • Fig. 7.

    Annual number of thunderstorm days in 2002–13. Lightning location data are computed within a radius of 17.5 km from the bin center (within a surface area of 962 km2) in 1 km × 1 km grid cells. Based on lightning data derived from the PERUN network.

  • Fig. 8.

    The average percentage of CG lightning flashes occurring during the nighttime (sun angle <−12°; section 3a). Results are computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Dots denote the main meteorological stations (44).

  • Fig. 9.

    Monthly annual mean number of CG lightning flashes (bars), percentage of CG lightning flashes detected during the nighttime (black solid line), and percentage of positive CG lightning flashes (black dashed line). Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Data have been limited to the administrative borders of Poland.

  • Fig. 10.

    Average monthly number of CG lightning flashes km−2 during the years 2002–13. Computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network.

  • Fig. 11.

    Monthly mean number of days with a detected thunderstorm from SYNOP (44 stations) reports (bars), and the number of days with CG lightning flashes detected: >1 (gray dotted line), >10 (gray dashed line), >100 (gray solid line), >1000 (black empty line), and >10 000 (black solid line). Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Data have been limited to the administrative borders of Poland.

  • Fig. 12.

    Average monthly number of thunderstorm days during the years 2002–13. Lightning location data are computed within a radius of 17.5 km from the bin center (within a surface area of 962 km2) for 1 km × 1 km grid cells. Based on lightning data derived from the PERUN network.

  • Fig. 13.

    Mean diurnal distribution of CG lightning flashes (percentage bars) with a time resolution of 1 h (UTC). Linear plots denote the diurnal distribution of CG lightning flashes on days when 2–10 (gray dotted line), 11–100 (gray dashed line), 101–1000 (gray solid line), 1001–10 000 (black empty line), and >10 000 (black solid line) flashes were detected. Based on lightning data derived from the PERUN network for the period 2002–13. Data have been limited to the administrative borders of Poland.

  • Fig. 14.

    (a) The average percentage of positive CG lightning flashes, and (b) the average peak current (kA) of negative CG lightning flashes. Results are computed for 10 km × 10 km grid cells. Based on lightning data derived from the PERUN network for the period from 2002 to 2013. Dots denote the main meteorological stations (44).