• Albrecht, R. I., S. J. Goodman, D. E. Buechler, R. J. Blakeslee, and H. J. Christian, 2016: Where are the lightning hotspots on earth? Bull. Amer. Meteor. Soc., 97, 20512068, https://doi.org/10.1175/BAMS-D-14-00193.1.

    • 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, https://doi.org/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, https://doi.org/10.1175/2009MWR2975.1.

    • Search Google Scholar
    • Export Citation
  • Antonescu, B., and A. Bell, 2015: Tornadoes in Romania. Mon. Wea. Rev., 143, 689701, https://doi.org/10.1175/MWR-D-14-00181.1.

  • Antonescu, B., S. Burcea, and A. Tănăse, 2013: Forecasting the onset of cloud-to-ground lightning using radar and upper-air data in Romania. Int. J. Climatol., 33, 15791584, https://doi.org/10.1002/joc.3533.

    • Search Google Scholar
    • Export Citation
  • Ashley, W. S., and C. W. Gilson, 2009: A reassessment of U.S. lightning mortality. Bull. Amer. Meteor. Soc., 90, 15011518, https://doi.org/10.1175/2009BAMS2765.1.

    • Search Google Scholar
    • Export Citation
  • Badoux, A., N. Andres, F. Techel, and C. Hegg, 2016: Natural hazard fatalities in Switzerland from 1946 to 2015. Nat. Hazards Earth Syst. Sci., 16, 27472768, https://doi.org/10.5194/nhess-16-2747-2016.

    • Search Google Scholar
    • Export Citation
  • Burcea, S., R. Cică, and R. Bojariu, 2016: Hail climatology and trends in Romania: 1961–2014. Mon. Wea. Rev., 144, 42894299, https://doi.org/10.1175/MWR-D-16-0126.1.

    • Search Google Scholar
    • Export Citation
  • Dewan, A., M. F. Hossain, M. M. Rahman, Y. Yamane, and R. L. Holle, 2017: Recent lightning-related fatalities and injuries in Bangladesh. Wea. Climate Soc., 9, 575589, https://doi.org/10.1175/WCAS-D-16-0128.1.

    • Search Google Scholar
    • Export Citation
  • Dewan, A., E. T. Ongee, M. Rafiuddin, M. M. Rahman, and R. Mahmood, 2018: Lightning activity associated with precipitation and CAPE over Bangladesh. Int. J. Climatol., https://doi.org/10.1002/joc.5286, in press.

    • Search Google Scholar
    • Export Citation
  • Dowdy, A. J., 2016: Seasonal forecasting of lightning and thunderstorm activity in tropical and temperate regions of the world. Sci. Rep., 6, 20 874, https://doi.org/10.1038/srep20874.

    • Search Google Scholar
    • Export Citation
  • Elsom, D. M., 2001: Deaths and injuries caused by lightning in the United Kingdom: Analyses of two databases. Atmos. Res., 56, 325334, https://doi.org/10.1016/S0169-8095(00)00083-1.

    • Search Google Scholar
    • Export Citation
  • Elsom, D. M., and J. D. C. Webb, 2014: Deaths and injuries from lightning in the UK, 1988–2012. Weather, 69, 221226, https://doi.org/10.1002/wea.2254.

    • Search Google Scholar
    • Export Citation
  • Galvonaite, A., 2004: Thunderstorm and lightning formation and continuance in Lithuania. Preprints, 18th Int. Lightning Detection Conf., Helsinki, Finland, Vaisala, 1–6.

  • Gourbière, E., 1999: Lightning injuries to humans in France. 11th Int. Conf. on Atmospheric Electricity, NASA/CPP-1999-209261, Guntersville, AL, NASA, 214–217.

  • Groenemeijer, P., and Coauthors, 2017: Severe convective storms in Europe: Ten years of research and education at the European Severe Storms Laboratory. Bull. Amer. Meteor. Soc., 98, 26412651, https://doi.org/10.1175/BAMS-D-16-0067.1.

    • Search Google Scholar
    • Export Citation
  • Holle, R. L., 2016: A summary of recent national-scale lightning fatality studies. Wea. Climate Soc., 8, 3542, https://doi.org/10.1175/WCAS-D-15-0032.1.

    • Search Google Scholar
    • Export Citation
  • Ioana, M., V. Ivanovici, E. Cordoneanu, D. Banciu, A. Apostu, and B. Ford, 2004: SIMIN–The integrated system for meteorological surveillance, forecast and alert in Romania. 20th Int. Conf. IIPS for Meteorology, Oceanography, and Hydrology, Seattle, WA, Amer. Meteor. Soc, 11.2, https://ams.confex.com/ams/84Annual/techprogram/paper_70954.htm.

  • Ioniţă, A., and O. Penina, 2016: About Romania data on causes of death. Tech. rep., The Human Cause-of-Death Database, 26 pp., http://www.causesofdeath.org/Data/ROU/20160121/ROU_bd.pdf.

  • Kompacher, M., G. Kindermann, and S. Pack, 2012: Fire losses and human accidents caused by lightning—An Austrian overview. 31st Int. Conf. on Lightning Protection, Vienna, Austria, IEEE, 1–5, https://doi.org/10.1109/ICLP.2012.6344276.

  • Loboda, M., 2008: Lightning deaths and injuries in Poland in period of 2001–2006. 29th Int. Conf. on Lightning Protection, Uppsala, Sweden, ICLP, 1–6.

  • López, R. E., and R. L. Holle, 1998: Changes in the number of lightning deaths in the United States during the twentieth century. J. Climate, 11, 20702077, https://doi.org/10.1175/1520-0442-11.8.2070.

    • Search Google Scholar
    • Export Citation
  • Mills, B., D. Unrau, C. Parkinson, B. Jones, J. Yessis, K. Spring, and L. Pentelow, 2008: Assessment of lightning-related fatality and injury risk in Canada. Nat. Hazards, 47, 157183, https://doi.org/10.1007/s11069-007-9204-4.

    • Search Google Scholar
    • Export Citation
  • Peppas, G. D., K. I. Bekas, I. A. Naxakis, E. C. Prygioti, and V. Charalampakos, 2012: Analysis of lightning impacts in Greece. 31st Int. Conf. on Lightning Protection, Vienna, Austria, IEEE, 1–5, https://doi.org/10.1109/ICLP.2012.6344347.

  • Raga, G. B., M. G. de la Parra, and B. Kucienska, 2014: Deaths by lightning in Mexico (1979–2011): Threat or vulnerability? Wea. Climate Soc., 6, 434444, https://doi.org/10.1175/WCAS-D-13-00049.1.

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

  • Roeder, W. P., B. H. Cummins, K. L. Cummins, R. L. Holle, and W. S. Ashley, 2015: Lightning fatality risk map of the conterminous United States. Nat. Hazards, 79, 16811692, https://doi.org/10.1007/s11069-015-1920-6.

    • Search Google Scholar
    • Export Citation
  • Said, R., 2017: Towards a global lightning locating system. Weather, 72, 3640, https://doi.org/10.1002/wea.2952.

  • Shearman, K. M., and C. F. Ojala, 1999: Some causes for lightning data inaccuracies: The case of Michigan. Bull. Amer. Meteor. Soc., 80, 18831891, https://doi.org/10.1175/1520-0477(1999)080<1883:SCFLDI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Tilev-Tanriover, Ş., A. Kahraman, M. Kadioğlu, and D. M. Schultz, 2015: Lightning fatalities and injuries in Turkey. Nat. Hazards Earth Syst. Sci., 15, 18811888, https://doi.org/10.5194/nhess-15-1881-2015.

    • Search Google Scholar
    • Export Citation
  • Van Den Broeke, M. S., D. M. Schultz, R. H. Johns, J. S. Evans, and J. E. Hales, 2005: Cloud-to-ground lightning production in strongly forced, low-instability convective lines associated with damaging wind. Wea. Forecasting, 20, 517530, https://doi.org/10.1175/WAF876.1.

    • Search Google Scholar
    • Export Citation
  • Wapler, K., 2013: High-resolution climatology of lightning characteristics within central Europe. Meteor. Atmos. Phys., 122, 175184, https://doi.org/10.1007/s00703-013-0285-1.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    (a) Topography of Romania (terrain height shaded according to the scale). The inset map indicates the position of Romania within Europe. (b) Mean annual CG lightning density (flashes per square kilometer per year, shaded according to the scale) for Romania based on data from the RNLDN collected between 2003 and 2007 (except 2006), adapted from Antonescu and Burcea (2010). (c) Mean annual CG lightning density (strokes per square kilometer per year, shaded according to the scale) for Romania based on GLD360 (provided by Vaisala).

  • View in gallery

    (a) The lightning-related fatalities in Romania reported each year between 1999 and 2015. The fatality rate (i.e., fatalities per million inhabitants per year) is shown on top of each bar and is calculated using the population data in (c). (b) Heat map showing the number of fatalities reported each month between 1999 and 2015 (shaded according to the scale). The total number of fatalities for each month is indicated on the right of the heat map. (c) The population of Romania (million inhabitants) for each year between 1999 and 2015 (blue line) together with the urban (dark green) and rural population (light green). The population data were obtained from the Romanian National Institute of Statistics.

  • View in gallery

    The number of lightning-related fatalities (shaded according to the scale) reported in each county between 1999 and 2015.

  • View in gallery

    (a) Population density (based on the average population between 1999 and 2015, shaded according to the scale) and the name of each county in Romania. (b) The number of lightning-related fatalities in each county (shaded according to the scale); the number in each county represents the rank of the county based on the number of lightning fatalities. (c) Fatality rate per million (i.e., fatalities per million inhabitants per year, shaded according to the scale); the number in each county represents the rank of the county based on fatality rate per million. (d) Normalized fatalities by area (i.e., fatalities per square kilometer, shaded according to the scale); the number in each county represents the rank of the county based on normalized fatalities by area.

  • View in gallery

    Percentage of rural population (based the average population between 1999 and 2015, shaded according to the scale) for each county in Romania. The number in each county represents the percentage of fatalities that occurred in rural areas and the total number of fatalities in each county (indicated in parentheses).

  • View in gallery

    The number of male (blue) and female (green) fatalities by age group that occurred in Romania between 1999 and 2015 in (a) rural areas and (b) urban areas. (c) As in (a), but normalized with the average (1999–2015) male and female population by age group living in rural areas. (d) As in (b), but normalized with the average (1999–2915) male and female population by age group living in urban areas.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 1 1 1
PDF Downloads 1 1 1

Lightning-Related Fatalities in Romania from 1999 to 2015

View More View Less
  • 1 School of Earth and Environmental Sciences, University of Manchester, Manchester, United Kingdom, and European Severe Storms Laboratory, Wessling, Germany
  • 2 Romanian National Meteorological Administration, Bucharest, Romania
© Get Permissions
Full access

Abstract

Lightning-related fatalities in Romania are analyzed and presented for the first time using data from the Romanian National Institute of Statistics. The database contains 724 lightning fatalities that occurred between 1999 and 2015 in Romania, corresponding to an average of 42.6 fatalities per year. The annual number of lightning fatalities decreased from 65 fatalities per year between 1999 and 2003 to 23.2 fatalities per year between 2011 and 2015. The majority of fatalities occurred in May–August (42% of all fatalities) with a peak in June (31%) and July (28%). The highest fatality rates (>2.6 fatalities per million inhabitants per year) are observed over southwestern Romania, a region characterized by high values of cloud-to-ground lightning density (>2 flashes per square kilometer per year) and by a relatively high percentage (>40%) of the population living in rural areas. The majority of fatalities (78%) were reported in rural areas. Approximately 78% of the victims were male. The most vulnerable group was males between the ages of 10–39 living in rural areas. To further reduce the lightning fatality rate in Romania, currently one of the highest in Europe, the authors argue that lightning mitigation activities and information campaigns about the risks associated with lightning should be initiated in Romania.

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Dr. Bogdan Antonescu, bogdan.antonescu@manchester.ac.uk

Abstract

Lightning-related fatalities in Romania are analyzed and presented for the first time using data from the Romanian National Institute of Statistics. The database contains 724 lightning fatalities that occurred between 1999 and 2015 in Romania, corresponding to an average of 42.6 fatalities per year. The annual number of lightning fatalities decreased from 65 fatalities per year between 1999 and 2003 to 23.2 fatalities per year between 2011 and 2015. The majority of fatalities occurred in May–August (42% of all fatalities) with a peak in June (31%) and July (28%). The highest fatality rates (>2.6 fatalities per million inhabitants per year) are observed over southwestern Romania, a region characterized by high values of cloud-to-ground lightning density (>2 flashes per square kilometer per year) and by a relatively high percentage (>40%) of the population living in rural areas. The majority of fatalities (78%) were reported in rural areas. Approximately 78% of the victims were male. The most vulnerable group was males between the ages of 10–39 living in rural areas. To further reduce the lightning fatality rate in Romania, currently one of the highest in Europe, the authors argue that lightning mitigation activities and information campaigns about the risks associated with lightning should be initiated in Romania.

© 2018 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Dr. Bogdan Antonescu, bogdan.antonescu@manchester.ac.uk

1. Introduction

Our knowledge about the spatial and temporal distribution of cloud-to-ground (CG) lightning over Europe has improved significantly over the recent years through pan-European (e.g., Anderson and Klugmann 2014) and regional studies (e.g., Wapler 2013). In contrast, we know far less about the impact of CG lightning, in particular, about the spatial and temporal distribution of lightning-related fatalities and injuries. Holle (2016), in a review of recent national-scale lightning fatalities studies, indicated that such studies were published for Austria (Kompacher et al. 2012), France (Gourbière 1999), Greece (Peppas et al. 2012), Lithuania (Galvonaite 2004), Poland (Loboda 2008), Turkey (Tilev-Tanriover et al. 2015), and the United Kingdom (Elsom and Webb 2014). More recently, lightning fatality data were also published for Switzerland (Badoux et al. 2016). Thus, lightning fatalities have been documented in only 7 (17%) of the 47 European countries with an area greater than 100 km2. The fatality rates (i.e., number of fatalities per million inhabitants per year) are in general lower in Europe (>0–0.4) compared with North America (0.10–2.7), Asia (>0–1.5), and Africa (0.9–84) (Holle 2016, their Table 1).

For Romania, as far as we are aware, no studies of lightning-related fatalities have been previously published. An estimate of 56 fatalities per year was previously mentioned in an article on the CG lightning climatology of Romania (based on data collected between 2002 and 2007 and provided by the Romanian National Institute of Statistics; Antonescu and Burcea 2010). This estimate of the annual average number of fatalities corresponds to a fatality rate of 2.47 fatalities per million inhabitants per year, which is higher than any other published fatality rates for Asia (except Sri Lanka), Australia, Europe, North and South America (except Mexico) (Holle 2016, their Table 1).

Motivated by this observation on the fatality rate in Romania, in this article we are exploring for the first time a database containing lightning fatalities with the aim of 1) documenting the lightning mortality in Romania using data collected in a consistent way (i.e., death certificates) over a relatively long period of time (i.e., 1999–2015), 2) increasing awareness of the risk posed by lightning to Romania, 3) supporting decision-makers and emergency managers to better understand the occurrence of lightning fatalities, and 4) contributing to a better understanding of lightning fatalities across Europe. This article is structured as follows. Section 2 details the lightning fatalities dataset and discusses its main limitations. Section 3 describes the climatology of CG lightning in Romania. The changes in the annual and monthly number of fatalities during the study period and the causes of these changes are analyzed in sections 4 and 5, respectively. Section 6 documents the spatial distribution of lightning fatalities based on administrative boundaries. Section 7 details how lightning fatalities are distributed by living area, gender, and age groups. Finally, section 8 summarizes the results of this paper.

2. Data and methods

The lightning-related fatality data used in this article were collected from medical death certificates by the Romanian National Institute of Statistics. The lightning fatality data were then verified and added to a database developed and maintained by the National Center for Statistics and Informatics in Public Health (NCSIPH), a division of the Romanian National Institute of Public Health. The NCSIPH database contains information on diseases and health issues grouped based on the International Statistical Classification of Diseases and Related Health Problems (ICD), a medical classification system of the World Health Organization (WHO). Romania had adopted the ICD classification in the 1950s, and in 1994 an abridged classification of the tenth version of the ICD was adopted, followed by the implementation of ICD-10 in 1999 (Ioniţă and Penina 2016). In the ICD-10 classification, lightning fatalities are included in the category “exposure to forces of nature” (X30–X39). From the NCSIPH database, only the entries with the ICD-10 code X33 (victim of lightning) were used in this study. The X33 code does not include victims associated with fire produced by lightning strikes or from fall of a tree or other object caused by lightning. Unlike previous studies that have analyzed lightning fatalities and injuries (e.g., Tilev-Tanriover et al. 2015; Dewan et al. 2017), in this article we focus only on lightning fatalities because this dataset is, in general, more reliable compared with the lightning injuries dataset.

A total of 719 lightning fatalities were included in the NCSIPH database between 1999 and 2015. Each entry on the NCSIPH database contains information about the place and date of death (i.e., county, judeţ in Romanian), living area (i.e., rural or urban), age, gender, and civil status (married, single, divorced, widowed) of the victim. The place of death is also specified in three categories: medical unit, home, and another situation. One advantage of using the NCSIPH database is that the records are standardized (i.e., the same type of information is provided for each record), compared with the records obtained from newspapers that sometimes, for example, do not include exact date and place of some incidents (e.g., Tilev-Tanriover et al. 2015). As shown by Mills et al. (2008), the records obtained from newspapers are limited by 1) the newspaper circulation, 2) the relevance of the lightning incident compared with other news items, 3) reliability of the sources (i.e., public or official accounts), and 4) limitation of electronic searches when extracting the data from newspaper archives. A partial search for the 2015 convective season (May–September) through the archive of one of the leading Romanian national newspapers (i.e., Adevărul) using a keyword search [e.g., trăsnet (cloud-to-ground lightning), trăsnit (struck by lightning), victimă (fatality), mort (dead)] did not result in any additional lightning-fatality cases to the NCSIPH database.

The NCSIPH database may also contain errors that result from a misinterpretation of ICD-10 codes (e.g., the code is attributed to fatalities from electrocution, this is further discussed in section 4), from the assignment of the place of death, or from the fact that the information is not received from all hospitals (e.g., Shearman and Ojala 1999). Also, the NCSIPH database does not contain explicit information (i.e., latitude and longitude) of the place of death, but only information about the county in which the death occurred. The NCSIPH database used in this article likely does not include all the cases of lightning fatalities in Romania. A survey of the European Severe Weather Database (ESWD; Groenemeijer et al. 2017) indicated that 28 damaging lightning cases from Romania were included in this database between 1999 and 2015, which resulted in 13 fatalities. In the ESWD, damaging lightning is defined as a lightning strike causing important damage to aircrafts, vehicles, ships, or injuring or killing people or animals. Five of these 13 lightning fatalities (representing 0.7% of all lightning fatalities in the NCSIPH database) were added as additional cases to the NCSIPH database. Thus, the database used in this article, containing 724 lightning fatalities between 1999 and 2015 (42.6 fatalities per year), represents the best data source to understand the vulnerability to lightning in Romania. Together with fatality data, we also used population data (i.e., Population and Demographic Structure, POP107A–Permanent Resident Population) provided by the Romanian National Institute of Statistics (RNIS) and extracted using the TEMPO-online database (http://www.insse.ro/cms/en). The RNIS defines the permanent-resident population as the number of inhabitants with Romanian citizenship and permanent residence (as printed on the identity card and registered by the administrative bodies) in the territory of Romania, delimited by territorial-administrative criteria.

3. Cloud-to-ground lightning climatology

Antonescu and Burcea (2010), using CG lightning data from the Romanian National Lightning Detection Network (RNLDN), showed that between 2003 and 2007 (except 2006) 1.75 million CG lightning flashes were recorded in Romania. They also showed that the CG lightning density in Romania varies from less than 0.34 flashes per square kilometer per year over eastern and northwestern Romania to values greater than 2.38 flashes per square kilometer per year over southwestern and central Romania (Figs. 1a,b). Data between 2012 and 2016 from the Global Lightning Dataset 360 (GLD360; Said 2017) provided by Vaisala, showed a similar pattern compared with the density obtained from RNLDN (Fig. 1c). The two main differences between the GLD360 and RNLDN datasets are 1) the CG lightning density maximum is over the southwestern part of the Carpathian Mountains in the GLD360 dataset and mainly over the southern Carpathians in the RNLDN dataset, and 2) the region of high CG lightning density stretching south from the southern Carpathians is closer to Bucharest (the capital city of Romania) in the GLD360 dataset compared with the RNLD dataset (cf. Figs. 1b and 1c). This is emphasizing the role of local topography in the development of thunderstorms. For example, Albrecht et al. (2016), using space-based Lightning Imaging Sensor observations, showed that the continental maxima in CG lightning density were in the proximity of major mountain ranges.

Fig. 1.
Fig. 1.

(a) Topography of Romania (terrain height shaded according to the scale). The inset map indicates the position of Romania within Europe. (b) Mean annual CG lightning density (flashes per square kilometer per year, shaded according to the scale) for Romania based on data from the RNLDN collected between 2003 and 2007 (except 2006), adapted from Antonescu and Burcea (2010). (c) Mean annual CG lightning density (strokes per square kilometer per year, shaded according to the scale) for Romania based on GLD360 (provided by Vaisala).

Citation: Weather, Climate, and Society 10, 2; 10.1175/WCAS-D-17-0091.1

Note that in Fig. 1b the CG lightning density is based on CG lightning flashes, whereas Fig. 1c is based on CG lightning strokes (i.e., components of the CG lightning discharge). As indicated by Rakov and Uman (2003, their Table 1.1), the typical value for the number of strokes per flash is between 3 and 5. Thus, the maximum CG lightning density based on the GLD360 dataset is greater than 2.8 flashes per square kilometer per year , comparable with the maximum values from Antonescu and Burcea (2010).

4. Interannual distribution of lightning fatalities

The annual distribution of the lightning fatalities shows a decrease in lightning fatality counts since 1999 (Fig. 2a). The number of lightning fatalities decreased from an average of 65 fatalities per year between 1999 and 2003 to an average of 23.2 fatalities per year between 2011 and 2015. This corresponds with a decrease in the fatality rate (i.e., fatalities per million inhabitants) from an average of 2.86 fatalities per million inhabitants between 1999 and 2003 to an average of 1.04 fatalities per million between 2011 and 2015 (Fig. 2a). Similar changes in the lightning fatality rates have been observed previously, but these changes occurred over longer periods of time. For example, for the United States, the average fatality rate changed during the twentieth century from an average of 4.92 fatalities per million inhabitants between 1900 and 1904 to an average 0.34 fatalities per million inhabitants between 1987 and 1991 (López and Holle 1998, their Table 1). Similar changes were also observed in Europe. For example, in Switzerland, the fatality rate decreased from 1.07 fatalities per million inhabitants between 1946 and 1950 to no fatalities between 2011 and 2015 (Badoux et al. 2016, Fig. 3).

Fig. 2.
Fig. 2.

(a) The lightning-related fatalities in Romania reported each year between 1999 and 2015. The fatality rate (i.e., fatalities per million inhabitants per year) is shown on top of each bar and is calculated using the population data in (c). (b) Heat map showing the number of fatalities reported each month between 1999 and 2015 (shaded according to the scale). The total number of fatalities for each month is indicated on the right of the heat map. (c) The population of Romania (million inhabitants) for each year between 1999 and 2015 (blue line) together with the urban (dark green) and rural population (light green). The population data were obtained from the Romanian National Institute of Statistics.

Citation: Weather, Climate, and Society 10, 2; 10.1175/WCAS-D-17-0091.1

Fig. 3.
Fig. 3.

The number of lightning-related fatalities (shaded according to the scale) reported in each county between 1999 and 2015.

Citation: Weather, Climate, and Society 10, 2; 10.1175/WCAS-D-17-0091.1

The decrease in the number of lightning-related fatalities and the changes in the fatality rate in Romania can be attributed to ambiguities in the cause of death, which will result in ICD-10 codes for lightning fatalities to be attributed to other types of fatalities, for example, fatalities from electrocution. In this case, lightning fatalities will be randomly distributed over the study period. But, as Fig. 2b shows, the majority of the lightning fatalities were reported between May and August (93% of all fatalities) every year between 1999 and 2015. The decrease in the lightning fatality rate in Romania can also be attributed to a decrease in the population of Romania from 22.85 million inhabitants in 1999 to 22.30 million in 2015 (Fig. 2c). The population of Romania decreased by 2.45% between 1999 and 2015, with a decrease of 5.36% in the rural population and a decrease of 0.08% in the urban population over the same period. The RNIS defines the urban population as the population that has its residence in cities and municipalities and the rural population as the population with residence in villages, with cities, municipalities, and villages being established by law. Other causes that likely contributed to the decrease in the fatality rate include improvements in weather forecasting, in particular in weather warnings for severe convective storms. In November 2000, the Romanian National Meteorological Administration (RNMA) started a process of modernization of the capabilities of detecting, monitoring, and forecasting weather phenomena over Romania (Ioana et al. 2004). As a part of this process, a lightning detection network and a Doppler radar network were installed that contributed to a better monitoring and forecasting of thunderstorms (Antonescu et al. 2013). Currently, the severe weather warnings issued by the RNMA include information about the lightning threat. Also, the development of emergency rescue services (e.g., the Romanian Mobile Emergency Service for Resuscitation and Extrication, fully operational since 1996 as a part of the Military Fireman Corps, and the Romanian General Inspectorate for Emergency Situations created in 2004) potentially contributed to the decrease in the fatality rate by preventing some of the lightning injuries to result in fatalities.

Despite the decrease since 1999, the lightning fatality rate in Romania (i.e., 1.88 fatalities per million inhabitants averaged over the entire study period and 1.04 fatalities per million inhabitants average between 2011 and 2015) is likely one of the highest in Europe. As indicated by Holle (2016), the highest lightning fatality rate (0.4 fatalities per million inhabitants) in Europe was previously reported for Turkey (Tilev-Tanriover et al. 2015). This high fatality rate is not only associated with a particular distribution of the population of Romania—56% living in urban areas in 2010 compared with approximately 76% living in urban areas in Turkey in 2010, for example—but also to the way in which the lightning threat is perceived. As hypothesized by Ashley and Gilson (2009), in general, people do not perceive lightning as life threatening compared with other severe weather phenomena (e.g., floods), even if lightning is observed more frequently than other severe weather phenomena. This perception can be changed by increasing awareness of the risk associated with lightning. Currently, to our knowledge, there are no programs devoted to increase awareness of the risk that lightning poses or to promote lightning safety in Romania.

5. Monthly distribution of lightning fatalities

The peak in the number of lightning fatalities is in May–June (341 fatalities, 47% of all fatalities; Fig. 2b). The monthly distribution of lightning fatalities is consistent with the climatology of severe storms in Romania. For example, Antonescu and Burcea (2010) showed that 98% of CG lightning flashes were detected between May and September, with a broad maximum in June–August (80% of all CG lightning flashes). Also, the monthly distribution of tornado reports has a peak in May (Antonescu and Bell 2015), and the monthly distribution of hail reports has a maximum in May–June (Burcea et al. 2016). The peak during late spring and early summer in the number of lightning fatalities is also due to a larger number of people engaged in outdoor labor activities during this months, compared with the rest of the year. These outdoor activities are mainly agriculture and shepherding (e.g., the livestock is moved during the summer months in the mountains and returned to the lower plains in the winter). The distribution of the number of lightning fatalities by month and by county shows that in May the fatalities are mainly reported over southwestern Romania (Fig. 3). In June and July, the number of fatalities is increased over central and northern Romania, and by August fatalities are mainly reported over northern Romania.

6. Spatial distribution of lightning fatalities

The spatial distribution of lightning-related fatalities was constructed using administrative boundaries (i.e., counties). A county-based analysis is not providing a detailed picture of the variability of lightning fatalities across Romania. This variability will be revealed, for example, by counting the number of fatalities on a grid (e.g., Ashley and Gilson 2009). Because our database does not contain the location of lightning fatalities, our analysis is restricted to a county-based perspective. Nevertheless, this analysis reveals how lightning fatalities are distributed across Romania and is useful for ranking counties in terms of the number of fatalities, fatality rate, and normalized fatalities (fatalities per 1000 km2).

Figure 4a shows the population density using the average population for each county between 1999 and 2015. Romania has 41 counties and the municipality of Bucharest, which has the same administrative level as that of a county. Bucharest has the smallest area (i.e., 238 km2) and the largest number of inhabitants (i.e., 2.15 million inhabitants) compared with any other county (Table 1). The majority of the counties (31 counties) have a population density lower than 100 inhabitants per km2. The highest population density is in Bucharest (9042 inhabitants per km2) and the lowest is in Tulcea (30 inhabitants per km2) (Fig. 4a). Lightning fatalities were more frequently reported over parts of southwestern, central, and northern Romania compared with other regions of the country (Fig. 4b). The counties of Dolj, Gorj, and Mehedinţi have the highest fatality counts, representing 14.5% of all lightning fatalities analyzed in this article (Fig. 4b, Table 1).

Fig. 4.
Fig. 4.

(a) Population density (based on the average population between 1999 and 2015, shaded according to the scale) and the name of each county in Romania. (b) The number of lightning-related fatalities in each county (shaded according to the scale); the number in each county represents the rank of the county based on the number of lightning fatalities. (c) Fatality rate per million (i.e., fatalities per million inhabitants per year, shaded according to the scale); the number in each county represents the rank of the county based on fatality rate per million. (d) Normalized fatalities by area (i.e., fatalities per square kilometer, shaded according to the scale); the number in each county represents the rank of the county based on normalized fatalities by area.

Citation: Weather, Climate, and Society 10, 2; 10.1175/WCAS-D-17-0091.1

Table 1.

Area (km2), population (million, average between 1999 and 2015), lightning fatalities counts, fatality rates (fatalities per million people per year), and normalized fatalities by area (fatalities per square kilometer × 1000) by county.

Table 1.

The lightning fatality rates per million inhabitants per year for each county are shown in Fig. 4c. During the study period, the average fatality rate across the counties was 2.18 fatalities per million inhabitants per year. The highest fatality rates were observed over parts of southern and central Romania, with a maximum in Mehedinţi (5.95), Gorj (5.10), and Alba (4.17) and a minimum in Arad (0.49), Bucureşti (0.38), and Vrancea (0.29). Thus, using rankings based on fatality rate results in a shift of the top counties ranked by the number of fatalities toward counties with a lower number of inhabitants (cf. Fig. 4b and Fig. 4c). For example, Mehedinţi has the highest fatality rate, but also one the lowest number of inhabitants in Romania (ranking 37 out of 42 counties; Table 1). When fatalities are normalized by the area of the county, a slightly different distribution emerges (cf. Fig. 4c and Fig. 4d), but one that still emphasizes that southwestern and central Romania, compared with other regions of Romania, are most vulnerable to CG lightning.

For the United States, Ashley and Gilson (2009) have identified a number of spatial corridors characterized by a relatively large number of lightning fatalities (e.g., central and eastern Florida) and showed that a high number of fatalities appear near large population centers (e.g., Chicago, Dallas–Fort Worth). They indicated that the irregular spatial pattern in the fatality distribution results from a combination of risk and human vulnerability. This observation also applies to the spatial pattern of lightning-related fatalities in Romania. In particular, we speculate that the high number of fatalities over southwestern Romania results from a combination of large CG lightning density values in a region with a large percentage of the population living in rural areas, as discussed in the next section.

7. Demographics of lightning fatalities

The majority of the lightning fatalities in Romania (78.5% of all fatalities) were reported in rural areas. For Turkey, Tilev-Tanriover et al. (2015) showed that 86% of lightning-related incidents between 1930 and 2014 occurred in rural areas. A high percentage of lightning fatalities in rural areas (93% of all fatalities) was also reported in Bangladesh (Dewan et al. 2017). Figure 5 shows that in 38 Romanian counties the majority of fatalities occurred in rural areas. The exceptions are Braşov, with 28.6% of all fatalities reported in rural areas, and Bucharest and Vrancea, where fatalities occurred only in urban areas. Even for counties characterized by a low incidence of CG lightning, for example, northeastern Romania (Fig. 1), the majority of fatalities occurred in rural areas. Thus, in these counties, it is likely that people are even less aware of the threat associated with CG lightning and, instead of taking precautions when thunderstorms are close by, they continue their outdoor activities (e.g., farming, shepherding). This observation also applies to people involved in outdoor recreational activities (e.g., trekking, rock climbing, camping). For example, the Carpathian Mountains and Transylvania in central Romania are popular destinations for recreational outdoor activities (Fig. 5). These results are somewhat different from those obtained by Roeder et al. (2015) for the United States. Roeder et al. (2015) showed that for the regions characterized by high CG lightning flash rate (e.g., Florida), the lightning fatality risk decreased in rural areas (e.g., southeast of Orlando, their Fig. 5) due to a lower population density.

Fig. 5.
Fig. 5.

Percentage of rural population (based the average population between 1999 and 2015, shaded according to the scale) for each county in Romania. The number in each county represents the percentage of fatalities that occurred in rural areas and the total number of fatalities in each county (indicated in parentheses).

Citation: Weather, Climate, and Society 10, 2; 10.1175/WCAS-D-17-0091.1

Of the 724 fatalities analyzed in this article, 548 (75.7%) were male and 176 (24.3%) were female. This result is consistent with previous studies. For example, Elsom (2001) indicated a similar percentage of male fatalities in the United Kingdom (73%), and Raga et al. (2014) showed that 79% of the lightning fatalities in Mexico between 1979 and 2011 were male fatalities. A lower percentage of male fatalities was reported by Tilev-Tanriover et al. (2015) for Turkey (67% of all fatalities). Mills et al. (2008) showed that 84% of lightning fatalities in Canada were male, consistent with the percentage (85% male) found in the United States by Ashley and Gilson (2009).

Figure 6 shows the number of fatalities for both male and female by age group for rural and urban areas. The largest number of fatalities by age group is for males between the ages of 10 and 39 living in rural areas (211 fatalities representing 29.1% of all fatalities; Fig. 6a). When weighted by the population (average 1999–2015 male and female rural population for each age group), a different picture emerges (Fig. 6b). In this case, the most vulnerable are males between the ages of 60 and 69 living in rural areas (7.42 fatalities per million inhabitants per year). This shift in the most vulnerable age group is due to a decrease in the male population living in rural areas with the age group 20–39 years. The high number of fatalities for the age groups 10–19 years (5.89 fatalities per million inhabitants per year) and 20–29 years (5.82 fatalities per million inhabitants per year) can be attributed, as indicated by Badoux et al. (2016), to increased risk-taking by young males compared with other groups. This observation can also explain the high number of lightning fatalities for the 10–19 age group (1.8 fatalities per million inhabitants per year) of males living in urban areas (Figs. 6c,d).

Fig. 6.
Fig. 6.

The number of male (blue) and female (green) fatalities by age group that occurred in Romania between 1999 and 2015 in (a) rural areas and (b) urban areas. (c) As in (a), but normalized with the average (1999–2015) male and female population by age group living in rural areas. (d) As in (b), but normalized with the average (1999–2915) male and female population by age group living in urban areas.

Citation: Weather, Climate, and Society 10, 2; 10.1175/WCAS-D-17-0091.1

8. Conclusions

The database developed and maintained by the NCSIPH was explored for the first time to document the lightning-related fatalities in Romania. Between 1999 and 2015, 719 lightning fatalities were included in the NCSIPH database, with five additional reports obtained from ESWD. During the study period, the lightning fatality rate decreased from a maximum of 2.86 fatalities per million inhabitants per year between 1999 and 2003 to a minimum of 1.04 fatalities per million inhabitants per year between 2011 and 2015. This trend in the fatality rate is consistent with trends reported in similar studies, with the only difference being the short period of time in which the decrease occurred in Romania. The decrease can be attributed to 1) a decrease in rural population (5.28% between 1999 and 2015), 2) improvements in severe weather warnings due to technology improvements (e.g., Doppler radars, lightning detection networks), and 3) improvements in medical care and technology (e.g., the Romanian Mobile Emergency Service for Resuscitation and Extrication).

The risk of being struck by lightning is highest over southwestern Romania (e.g., the largest number of fatalities by county was recorded in Dolj with 41 fatalities, and the highest fatality rate was observed in Mehedinţi with 5.95 fatalities per million per year). This high risk results from a maximum in the CG lightning activity over southwestern Romania occurring in a region in which a large proportion of the population lives in rural areas. The most lightning fatalities (i.e., 93% of all fatalities) were reported between May and August, which corresponds to a maximum in the monthly distribution of CG lightning between May and September.

The majority of lightning fatalities (i.e., 78% of all fatalities) occurred in rural areas. The number of male fatalities (i.e., 548 fatalities) was 3 times more than the number of female fatalities (i.e., 176 fatalities). The most vulnerable category was males living in rural areas with ages between 10 and 39 years (211 fatalities, 29.1% of all fatalities), while it was males with ages between 60 and 69 years when normalized by population (i.e., 7.42 fatalities per million per year).

This article emphasizes that the risk associated with lightning in Romania should not be ignored and should be included in risk-reduction strategies and programs. Despite the changes in the recent years, the lightning fatality rate in Romania is still one of the highest in Europe (i.e., 6.8 fatalities per year corresponding to a fatality rate of 1.04 fatalities per million per year). As showed in previous studies documenting lightning-related fatalities in the other eight European countries, the highest fatality rate is in Turkey, with 0.4 fatalities per million per year. We believe that the lightning fatality rate for Romania has not reached a minimum and that the rate can be further reduced through mitigation efforts. Currently, the Romanian General Inspectorate for Emergency Situations is conducting campaigns to increase awareness and mitigate the risks associated with severe weather events, and in April 2016 a smartphone application was launched (with 260 000 users as of 1 November 2017) to disseminate the severe weather warnings issued by the RNMA. To our knowledge, lightning mitigation activities have not been previously conducted in Romania. Thus, for example, the Romanian General Inspectorate for Emergency Situations and the RNMA can 1) participate in national information campaigns each spring (e.g., Lightning, or by extension, Severe Weather Awareness week), 2) conduct countywide campaigns to inform the farmers, and 3) develop lightning safety resources (see, e.g., http://www.lightningsafety.noaa.gov/). Furthermore, the lightning fatality rate in Romania can be reduced by analyzing the environmental parameters associated with lightning occurrence in Romania and developing lightning forecasting schemes based on these parameters (e.g., Van Den Broeke et al. 2005; Antonescu et al. 2013; Dewan et al. 2018) and by developing seasonal forecasts for lightning (e.g., Dowdy 2016). These forecasts can be then used by decision-makers, emergency managers, and (re)insurance companies to reduce the impact of thunderstorms.

Acknowledgments

We thank David M. Schultz for his comments on an earlier version of the manuscript. We also thank Claudiu Botezat and Cristian Calomfirescu from the Romanian National Center for Statistics and Informatics in Public Health and Dan Mihai Marius from the Romanian National Institute of Public Health for providing the data on lightning-related fatalities in Romania. We thank Laura Ichim from the Romanian National Institute of Statistics for providing information about the population dataset and Col. Daniel Marian Dragne for providing information about the activities of the Romanian General Inspectorate for Emergency Situations. We thank Ronald L. Holle and Vaisala for the lightning data used in Fig. 2, Alexandre Badoux from the Swiss Federal Research Institute for providing data on lightning fatalities in Switzerland, and Tudor Palela for providing the map showed in Fig. 1a. Partial funding for Antonescu was provided by the Natural Environment Research Council through Grant NE/N003918/1.

REFERENCES

  • Albrecht, R. I., S. J. Goodman, D. E. Buechler, R. J. Blakeslee, and H. J. Christian, 2016: Where are the lightning hotspots on earth? Bull. Amer. Meteor. Soc., 97, 20512068, https://doi.org/10.1175/BAMS-D-14-00193.1.

    • 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, https://doi.org/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, https://doi.org/10.1175/2009MWR2975.1.

    • Search Google Scholar
    • Export Citation
  • Antonescu, B., and A. Bell, 2015: Tornadoes in Romania. Mon. Wea. Rev., 143, 689701, https://doi.org/10.1175/MWR-D-14-00181.1.

  • Antonescu, B., S. Burcea, and A. Tănăse, 2013: Forecasting the onset of cloud-to-ground lightning using radar and upper-air data in Romania. Int. J. Climatol., 33, 15791584, https://doi.org/10.1002/joc.3533.

    • Search Google Scholar
    • Export Citation
  • Ashley, W. S., and C. W. Gilson, 2009: A reassessment of U.S. lightning mortality. Bull. Amer. Meteor. Soc., 90, 15011518, https://doi.org/10.1175/2009BAMS2765.1.

    • Search Google Scholar
    • Export Citation
  • Badoux, A., N. Andres, F. Techel, and C. Hegg, 2016: Natural hazard fatalities in Switzerland from 1946 to 2015. Nat. Hazards Earth Syst. Sci., 16, 27472768, https://doi.org/10.5194/nhess-16-2747-2016.

    • Search Google Scholar
    • Export Citation
  • Burcea, S., R. Cică, and R. Bojariu, 2016: Hail climatology and trends in Romania: 1961–2014. Mon. Wea. Rev., 144, 42894299, https://doi.org/10.1175/MWR-D-16-0126.1.

    • Search Google Scholar
    • Export Citation
  • Dewan, A., M. F. Hossain, M. M. Rahman, Y. Yamane, and R. L. Holle, 2017: Recent lightning-related fatalities and injuries in Bangladesh. Wea. Climate Soc., 9, 575589, https://doi.org/10.1175/WCAS-D-16-0128.1.

    • Search Google Scholar
    • Export Citation
  • Dewan, A., E. T. Ongee, M. Rafiuddin, M. M. Rahman, and R. Mahmood, 2018: Lightning activity associated with precipitation and CAPE over Bangladesh. Int. J. Climatol., https://doi.org/10.1002/joc.5286, in press.

    • Search Google Scholar
    • Export Citation
  • Dowdy, A. J., 2016: Seasonal forecasting of lightning and thunderstorm activity in tropical and temperate regions of the world. Sci. Rep., 6, 20 874, https://doi.org/10.1038/srep20874.

    • Search Google Scholar
    • Export Citation
  • Elsom, D. M., 2001: Deaths and injuries caused by lightning in the United Kingdom: Analyses of two databases. Atmos. Res., 56, 325334, https://doi.org/10.1016/S0169-8095(00)00083-1.

    • Search Google Scholar
    • Export Citation
  • Elsom, D. M., and J. D. C. Webb, 2014: Deaths and injuries from lightning in the UK, 1988–2012. Weather, 69, 221226, https://doi.org/10.1002/wea.2254.

    • Search Google Scholar
    • Export Citation
  • Galvonaite, A., 2004: Thunderstorm and lightning formation and continuance in Lithuania. Preprints, 18th Int. Lightning Detection Conf., Helsinki, Finland, Vaisala, 1–6.

  • Gourbière, E., 1999: Lightning injuries to humans in France. 11th Int. Conf. on Atmospheric Electricity, NASA/CPP-1999-209261, Guntersville, AL, NASA, 214–217.

  • Groenemeijer, P., and Coauthors, 2017: Severe convective storms in Europe: Ten years of research and education at the European Severe Storms Laboratory. Bull. Amer. Meteor. Soc., 98, 26412651, https://doi.org/10.1175/BAMS-D-16-0067.1.

    • Search Google Scholar
    • Export Citation
  • Holle, R. L., 2016: A summary of recent national-scale lightning fatality studies. Wea. Climate Soc., 8, 3542, https://doi.org/10.1175/WCAS-D-15-0032.1.

    • Search Google Scholar
    • Export Citation
  • Ioana, M., V. Ivanovici, E. Cordoneanu, D. Banciu, A. Apostu, and B. Ford, 2004: SIMIN–The integrated system for meteorological surveillance, forecast and alert in Romania. 20th Int. Conf. IIPS for Meteorology, Oceanography, and Hydrology, Seattle, WA, Amer. Meteor. Soc, 11.2, https://ams.confex.com/ams/84Annual/techprogram/paper_70954.htm.

  • Ioniţă, A., and O. Penina, 2016: About Romania data on causes of death. Tech. rep., The Human Cause-of-Death Database, 26 pp., http://www.causesofdeath.org/Data/ROU/20160121/ROU_bd.pdf.

  • Kompacher, M., G. Kindermann, and S. Pack, 2012: Fire losses and human accidents caused by lightning—An Austrian overview. 31st Int. Conf. on Lightning Protection, Vienna, Austria, IEEE, 1–5, https://doi.org/10.1109/ICLP.2012.6344276.

  • Loboda, M., 2008: Lightning deaths and injuries in Poland in period of 2001–2006. 29th Int. Conf. on Lightning Protection, Uppsala, Sweden, ICLP, 1–6.

  • López, R. E., and R. L. Holle, 1998: Changes in the number of lightning deaths in the United States during the twentieth century. J. Climate, 11, 20702077, https://doi.org/10.1175/1520-0442-11.8.2070.

    • Search Google Scholar
    • Export Citation
  • Mills, B., D. Unrau, C. Parkinson, B. Jones, J. Yessis, K. Spring, and L. Pentelow, 2008: Assessment of lightning-related fatality and injury risk in Canada. Nat. Hazards, 47, 157183, https://doi.org/10.1007/s11069-007-9204-4.

    • Search Google Scholar
    • Export Citation
  • Peppas, G. D., K. I. Bekas, I. A. Naxakis, E. C. Prygioti, and V. Charalampakos, 2012: Analysis of lightning impacts in Greece. 31st Int. Conf. on Lightning Protection, Vienna, Austria, IEEE, 1–5, https://doi.org/10.1109/ICLP.2012.6344347.

  • Raga, G. B., M. G. de la Parra, and B. Kucienska, 2014: Deaths by lightning in Mexico (1979–2011): Threat or vulnerability? Wea. Climate Soc., 6, 434444, https://doi.org/10.1175/WCAS-D-13-00049.1.

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

  • Roeder, W. P., B. H. Cummins, K. L. Cummins, R. L. Holle, and W. S. Ashley, 2015: Lightning fatality risk map of the conterminous United States. Nat. Hazards, 79, 16811692, https://doi.org/10.1007/s11069-015-1920-6.

    • Search Google Scholar
    • Export Citation
  • Said, R., 2017: Towards a global lightning locating system. Weather, 72, 3640, https://doi.org/10.1002/wea.2952.

  • Shearman, K. M., and C. F. Ojala, 1999: Some causes for lightning data inaccuracies: The case of Michigan. Bull. Amer. Meteor. Soc., 80, 18831891, https://doi.org/10.1175/1520-0477(1999)080<1883:SCFLDI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Tilev-Tanriover, Ş., A. Kahraman, M. Kadioğlu, and D. M. Schultz, 2015: Lightning fatalities and injuries in Turkey. Nat. Hazards Earth Syst. Sci., 15, 18811888, https://doi.org/10.5194/nhess-15-1881-2015.

    • Search Google Scholar
    • Export Citation
  • Van Den Broeke, M. S., D. M. Schultz, R. H. Johns, J. S. Evans, and J. E. Hales, 2005: Cloud-to-ground lightning production in strongly forced, low-instability convective lines associated with damaging wind. Wea. Forecasting, 20, 517530, https://doi.org/10.1175/WAF876.1.

    • Search Google Scholar
    • Export Citation
  • Wapler, K., 2013: High-resolution climatology of lightning characteristics within central Europe. Meteor. Atmos. Phys., 122, 175184, https://doi.org/10.1007/s00703-013-0285-1.

    • Search Google Scholar
    • Export Citation
Save