Lightning-ignited wildﬁres and long-continuing-current lightning in the Mediterranean Basin: Preferential meteorological conditions

. Lightning is the major cause of natural ignition of wildﬁres worldwide and produces the largest wildﬁres in some regions. Lightning strokes produce about 5% of forest ﬁres in the Mediterranean basin and are one of the most important precursors of the largest forest ﬁres during the summer. Lightning-ignited wildﬁres produce signiﬁcant emissions of aerosols, black carbon and trace gases, such as CO, SO 2 , CH 4 and O 3 , affecting air quality. Characterization of the meteorological and cloud conditions of lightning-ignited wildﬁres in the Mediterranean basin can serve to improve ﬁre forecasting models and to 5 upgrade the implementation of ﬁre emissions in atmospheric models. This study investigates the meteorological and cloud conditions of Lightning-Ignited Wildﬁres (LIW) and Long-Continuing-Current (LCC) lightning ﬂashes in the Iberian Peninsula and Greece. LCC lightning and lightning in dry thunderstorms with low precipitation rate have been proposed to be the main precursors of the largest wildﬁres. We use lightning data provided by the World Wide Lightning Location Network (WWLLN), the Earth Network Total Lightning Network (ENTLN) and the Light- ning Imaging Sensor (LIS) onboard the International Space Station (ISS) together with four databases of wildﬁres produced in Spain, Portugal, Southern France and Greece, respectively, in order to produce a climatology of LIW and LCC lightning over the Mediterranean basin. In addition, we use meteorological data provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5-reanalysis data set and by the Spanish State Meteorological Agency (AEMET) together with the Cloud Top Height (CTH) product derived from Meteosat Second Generation (MSG) satellites measurements to investigate the meteorological conditions of LIW and LCC lightning. According to our results, LIW and a signiﬁcant amount of LCC lightning ﬂashes tend to occur in dry thunderstorms with weak updrafts. Our results suggest that lightning-ignited wildﬁres tend to occur in high-based clouds with clouds with high base with (cid:58) a vertical content of moisture lower than the climatological value, as well as with a higher temperature and a lower precipitation rate. Meteorological conditions of LIW from the Iberian Peninsula and Greece are in agreement, although some differences possibly caused by highly variable topography in Greece and a more humid environment are observed. These results show the possibility of using the typical and cloud conditions of lightning ﬂashes as proxy to parameterize the ignition of wildﬁres in atmospheric or forecasting models. the of the clouds clouds with (cid:58)(cid:58)(cid:58)(cid:58) high (cid:58)(cid:58)(cid:58)(cid:58) base (cid:58) favor the inception of LIW. High-based clouds clouds (cid:58)(cid:58)(cid:58)(cid:58) with (cid:58)(cid:58)(cid:58)(cid:58) high (cid:58)(cid:58)(cid:58)(cid:58) base (cid:58) favor the evaporation of precipitation before reaching the surface, increasing the probability of survival and arrival of LIW. High-based clouds clouds (cid:58)(cid:58)(cid:58)(cid:58) with (cid:58)(cid:58)(cid:58)(cid:58) high base (cid:58) and low content of moisture in the low- or mid-level are the typical meteorological conditions of dry thunderstorms in the western United States, as demonstrated by Wallmann (2004) and Nauslar et al. (2013). Our results suggest, for the ﬁrst time, that high-based clouds clouds with high base favor the occurrence and arrival of LIW in the Iberian Peninsula and Greece. The Some previous studies have developed statistical methods, machine learning approaches or parameterizations based on meteorological variables to predict LIW These methods typically use lightning parameterizations to predict the occurrence of lightning and then estimate the risk of LIW according to environmental factors. However, the employed lightning parameterizations do not distin-guish between regular/typical lightning and LCC-lightning ﬂashes. The analysis of meteorological conditions of LIW and LCC(>20 ms)-lightning in the Mediterranean basin suggests that using the updraft between 300 hPa and 450 hPa could serve to improve LIW forecasting methods. The common features of thunderstorms producing LIW and LCC(>20 ms)-lightning 640 ﬂashes conﬁrm that monitoring the occurrence of LCC(>20 ms)-lightning ﬂashes and/or the characteristics of thunderstorms producing them could improve the forecast of ﬁre ignition. In particular, calibrating

However, European studies are rare. Lightning is the major cause of ignition in the European boreal forests (Granström, 2001;40 Larjavaara et al., 2005b, a;Granström and Niklasson, 2008;Rolstad et al., 2017) and one of the main cause of ignition in the Alps (Conedera et al., 2006;Moris et al., 2020). In the Mediterranean basin, lightning causes about 5% of the total number of forest fires (Vázquez and Moreno, 1998;Camia et al., 2010;Koutsias et al., 2013), while the rest of fires are man-caused.
However, LIW are one of the most important precursors of the largest forest fires during the summer (Vázquez and Moreno, 1998;Badia et al., 2002;Amatulli et al., 2007;García-Ortega et al., 2011;Oliveira et al., 2012;San-Miguel-Ayanz et al., 2013; seven cases of fire-igniting lightning strokes with duration between 40 ms and 282 ms. Fuquay et al. (1967) and Adachi et al. (2009) showed that the optical signal emitted by lightning discharges can be related to the duration of the electrical discharge. Bitzer (2017) reported the first tropical and mid-latitude climatology of LCC lightning discharges with a duration larger than 10 ms. Bitzer (2017) provided the total number of LCC lightning with duration up to 40 ms from optical lightning measurements reported by LIS onboard the Tropical Rainfall Measuring Mission (TRMM) satellite following a low-Earth 90 orbit between 1997 and 2015, providing lightning measurements in the range of latitude between 35 • N and 35 • S. According to Bitzer (2017), LCC lightning discharges tend to occur in oceanic and winter thunderstorms, where the updrafts are weaker than in typical summer thunderstorms over land. Bitzer (2017) proposed that thunderstorms with weaker updrafts would produce small charging rates, allowing the charging process to develop larger charge regions before the onset of lightning and providing the discharge with more energy to be transferred. However, the TRMM satellite did not cover the European continent. LIS is 95 now operating onboard the International Space Station (ISS), covering for the first time the Mediterranean basin and providing optical measurements of the duration of lightning pulses.
In this work, we investigate the meteorological characteristics of fire-producing thunderstorms and the electrical characteristics of the lightning-igniting fires in the Mediterranean basin. This approach will serve to improve fire forecasting methods and atmospheric models including fire-emissions. As we cannot directly connect LIW to LCC flashes using LLS, we search for 100 shared meteorological conditions. We develop the first climatology of LCC lightning over Europe from the lightning data provided by LIS onboard the International Space Station (ISS-LIS) with a duration larger than 20 ms and its possible relationship with the occurrence of LIW. This duration is lower than the lowest duration of the continuing phase of a fire-igniting lightning reported by Fuquay et al. (1967) (40 ms), but enough to be considered as a LCC lightning flash. In particular, we focus our analysis on the Iberian Peninsula (Spain, Portugal and the Mediterranean France) and Greece. We do not include data over Italy 105 because they were not accessible. We combine five fire databases from these countries with lightning measurements provided by the World Wide Lightning Location Network (WWLLN) and the Earth Network Total Lightning Network (ENTLN). We use meteorological data provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5-reanalysis data set and by the Spanish State Meteorological Agency (AEMET). For the first time, we combine LCC lightning data provided by LIS onboard the International Space Station (ISS) with meteorological data sets. 110 2 Data and methodology Table 1 shows the main characteristics of the data sets employed in this study. In this section, we describe each of the data set and the employed methodology to investigate the meteorological conditions of LIW and LCC lightning flashes.

Lightning measurements
We use ground-based lightning data provided by the Lightning Locations Systems WWLLN and ENTLN to search the lightning 115 candidates for the forest fires in the periods 2009-2013 and 2014-2019, respectively. In addition, we use lightning optical measurements reported by ISS-LIS between 2017 and 2020 to investigate LCC lightning over Europe. The ground-based WWLLN is composed by a global network of VLF sensors that can provide the position, time of occurrence and energy radiation by lightning discharges (Dowden et al., 2002;Rodger et al., 2005). WWLLN is more sensitive to CG lightning discharges than IC, with a total global stroke Detection Efficiency (DE) between 2009 and 2013 of 8-13% 120 (Hutchins et al., 2012a;Rudlosky and Shea, 2013;Bitzer et al., 2016). The stroke location accuracy of WWLLN is between 5 km and 10 km, while the temporal accuracy is about tens of microseconds (Abreu et al., 2010;Rudlosky and Shea, 2013).
In this work, we use the relationship between the energy of the strokes measured by WWLLN and the peak current of the discharge, as reported by Hutchins et al. (2012b) to estimate the peak current of the fire-igniting lightning.
The ground-based ENTLN is a global network composed by VLF sensors that provide the position, time of occurrence, 125 polarity and peak current of lightning strokes. ENTLN has a DE of about 90% for CG strokes over the US (Zhu et al., 2017;Lapierre et al., 2020) and a total global stroke DE of about 57% (Bitzer et al., 2016). The median stroke location error is 631 m (Mallick et al., 2015). In this work, we use the flash product provided by ENTLN. This product is based on the flash criteria proposed by Liu and Heckman (2011), to cluster these strokes into flashes, in which two strokes are part of the same flash if they occur in a 0.7 s temporal window and in a 10 km spatial window.

130
After the end of operation of LIS on the TRMM satellite in 2015, a similar instrument was placed on the ISS for a 2 to 4 year mission starting in March 2017 covering latitudes between 54.3 • N and 54.3 • S (Blakeslee et al., , 2020. LIS detects optical emissions from lightning with a frame integration time of 1.79 ms (Bitzer and Christian, 2015) with a spatial resolution of 4 km (Blakeslee et al., 2020). LIS assorts contiguous events into groups, and clusters groups into flashes with a temporal criteria of 330 ms and an spatial criteria of 5.5 km (Mach et al., 2007). Bitzer (2017) proposed a method to identify LCC 135 lightning flashes from the groups reported by TRMM-LIS. According to Bitzer (2017), optical emissions detected in five or more consecutive frames (time contiguous groups), that are in the same flash, can be classified as a LCC lightning flash. In this work, we use the method proposed by Bitzer (2017) to produce a climatology of LCC lightning flashes over Europe based on ISS-LIS lightning measurements between March 2017 and September 2020. We use 20 ms as the lower limit to classify a flash as a LCC flash in order to avoid introducing CG flashes without continuing currents (typical CG) into our climatology.

Forest fire databases
The data of fires in Spain is provided by the Spanish Ministerio de Agricultura, Pesca y Alimentación (López-Santalla and Hellenic Fire Brigade, collects and publish fire incidence data from 2000 onwards, after getting (in summer of 1998) the responsibility of fire extinction in Greece from Forest Service, including the number of fires and the corresponding are burned (Koutsias et al., 2013). During the last years, x and y coordinates of the fire ignition points are also recorded, unfortunately 160 without recording the fire cause. For three years during the period 2017-2019, these data were further checked for providing high positional accuracy and made available to be used here. The total number of fires in this database between 2017 and 2019 is 62690. We use the criterion described in Section 2.4 to filter out some LIW from this data set, obtaining 1999 LIW (about 3.2%).

165
We analyze the meteorological conditions of thunderstorms favoring fires or LCC lightning using meteorological data from the European Centre for Medium-Range Weather Forecasts (ECMWF) fifth generation reanalysis ERA5) (Hersbach et al., 2020) and the Cloud Top Height (CTH) product provided by EUMETSAT (Schmetz et al., 2002). In addition, for some selected cases we analyze the meteorological products provided by the Spanish State Meteorological Agency (AEMET) (Gutiérrez Núñez et al., 2018) for some selected cases.

170
ERA5 provides 1-hourly meteorological data using a 4D-var assimilation scheme at 139 pressure levels with an horizontal resolution of 0.25 • . The product ERA5-Land provides meteorological data over land by replaying the land component of the ECMWF ERA5 climate reanalysis with an horizontal resolution of 0.1 • (Poli et al., 2016). In this work, we analyzed the 1- the hourly-accumulated precipitation, :: the ::::::: specific ::::::: humidity :: at :::: 450 ::: hPa : and vertical profiles between ground and 200 hPa pres- The AEMET provides meteorological measurements under request. In this work, we selected some fire-igniting lightning 185 flashes and some LCC lightning over Spain and collected radar echo top height measurements provided by AEMET (maximum height of the 12-dBZ echo). These heights correspond to the maximum heights populated by the precipitable particles.
We use the land cover map from 2015 produced by the European Space Agency (ESA) Climate Change Initiative (CCI) from the PROBA-Vegetation (PROBA-V) and Sentinel-3 OLCI (S3 OLCI) time series to classify lightning flashes over coniferous and mixed forests. This classification of lightning flashes ensures that only lightning flashes occurring over areas with vege-190 tation that can be ignited are included in our analysis, as most of forest fires are due to the ignition of conifers or shrublands during the summer season (Pineda and Rigo, 2017). This map has a temporal resolution of one year and a horizontal resolution of 300 m. We consider that a lightning flash is taking place over coniferous or mixed forests if there is at least one 300 m × 300 m grid cell with conifer or mixed trees 10 km around the flash position. In particular, we consider coniferous and mixed forests that including the label "needleleaved" in the D3.3.12-v1.3_PUGS_ICDR_LC_v2.1.x_PRODUCTS_v1.3 product. We 195 use elevation data from the NASA Shuttle Radar Topographic Mission Farr et al. (2007) to get the elevation of LIW.
The CTH product provided by EUMETSAT is based on measurements of the Meteosat Second Generation (MSG) satellites.
The geostationary orbit of MSG satellites is centered at 0 • E, 0 • N, reporting data at the rate of one Earth full-disk scan every 15 min. We investigate the CTH of fire-igniting thunderstorms because it is closely related with the level of convection in thunderstorms and the occurrence of lightning (Price and Rind, 1992). The CTH product is calculated by EUMETSAT from

Search of lightning-candidates for the fires
We search the most probable CG lightning candidate for each fire using the proximity index A proposed by Larjavaara et al. (2005b), (1)

210
The proximity index A combines the distance between each fire and each lightning discharge (D) with the delay between them (T ), also known as holdover. Parameters (T max ) and (D max ) correspond to the maximum holdover and distance between a fire and a lightning discharge to consider the later as the cause of ignition. We set T max = 14 days and D max = 10 km (Larjavaara et al., 2005b). In the case of no CG lightning discharges taking place within the proposed spatio-temporal window with respect to a fire, we consider that the fire was not ignited by lightning. However, the proximity index can be greater than In the case of the database of fires taking place in Spain and Mediterranean France, the cause of ignition is provided. Therefore, we can easily discard the fires with A > 0 that were not ignited by lightning. However, the database of fires in Portugal and Greece do not provide the cause of ignition, and the proximity index can be greater than zero if a lightning discharge preceded a fire even if the discharge did not cause the ignition. In order to discard the fires of Portugal and Greece 220 that were not ignited by lightning, we estimate the threshold value of the A index that ensures that at least 80% of the selected fires are ignited by lightning. To this end, we use the Spanish and French databases of fires. We plot in Fig. 1 the percentage of fires in Spain and France ignited by lightning using different values of A as threshold. According to this figure, setting the minimum value of A as 0.7 ensures that 80% of the selected fires are produced by lightning. Therefore, in our analysis of LIW over Portugal and Greece we only consider fires with A≥0.7. This approach can exclude a significant number of LIW from 225 the analysis. However, it ensures that the sample is not influenced by fires that were not ignited by lightning. Following this approach, we find 359 (0.22% of the total) and 1999 (3.2% of the total) LIW with A≥0.7 in Portugal and Greece, respectively.
We have obtained a lower percentage of LIW in Portugal with respect to all fires than in Greece, Spain and Mediterranean due to the low DE of WWLLN. Another interesting feature of Fig. 1 is that the percentage of LIW in the sample with A values lower than 0.3 is higher in the period 2014-2015 than in 2009-2013. This difference can be explained by the higher DE of 230 ENTLN than WWLLN in the Iberian Peninsula.
As explained in Section 2.3, we will limit our analysis to LIW occurring between May and September and in coniferous and mixed forest.

Classification of LCC lightning flashes
The standard ISS LIS product includes lightning data at event, group and flash level (Mach et al., 2007). Groups are clustered 235 into flashes if they occur within 330 ms and 5.5 km (Mach et al., 2007). In this work, we use group and flash data. We apply the method to classify lightning flashes reported by ISS-LIS as LCC lightning flashes proposed by Bitzer (2017) for TRMM LIS. We consider a flash as a LCC lightning flash if its optical emissions are detected in ten or more consecutive frames, which entails that the detected optical signal is continuous during about 20 ms. In addition, we have classified flashes with optical emissions detected in five or more consecutive frames (optical signal is continuous during about 10 ms (Bitzer, 2017)) 240 to investigate the possible relationships between the duration of the current and the meteorological conditions.

Analysis of meteorological conditions
We compare the meteorological variables of fire-igniting lightning flashes to typical CG lightning flashes over coniferous and For forecasting purposes, we collect the maximum value 3 hours before the discharge, as the value of CAPE during the 250 passage of a thunderstorm is significantly lower than the value of CAPE during the development of the storm. For every lightning discharge, we collect the CTH value provided by EUMETSAT. is lower than 0.05 (less than 5% probability of equal median) we reject the hypothesis of equal medians and consider that both samples are statistically different (Kruskal and Wallis, 1952). It is important to mention that different medians does not necessary imply that the analyzed variable would be a good predictor of LIW. islands as the Iberian Peninsula. Fig. 3(a) and 4(a) show the monthly total number of LIW between May and September in the Iberian Peninsula and in Greece, respectively. In the Iberian Peninsula, the occurrence of LIW reaches its peak in July, while in Greece the peak occurs in June and a secondary peak in September.   in the Iberian Peninsula are, respectively, 0.48 hours, 8.13 hours and 29.23 hours, while they are 14.79 hours, 29.43 hours and 48.78 hours in Greece. Differences in the holdover between the Iberian Peninsula and Greece can be due to differences in the detection of fires by local authorities. Fig. 3(d) and 4(d) suggest a diurnal cycle in the holdover that can be due to different conditions for arrival during day, similar as obtained by Pineda and Rigo (2017) for LIW ober Catalonia. At noon, meteorological conditions can favor a rapid arrival after ignition (high temperature and low humidity). Ignitions separated from 290 noon occur under meteorological conditions that do not favor rapid arrival.
We have found that the mean elevation of LIW in the Iberian Peninsula is 775 m, while in the case of Greece the mean elevation is 319 m.
These conditions favor the evaporation of precipitable water before reaching the ground (Nauslar et al., 2013)  would not be good predictors for LIW in Greece, as the plotted distributions of fire-igniting lightning and CG lightning are not substantially different. Differences in RH and temperature at 850 hPa between Greece and the Iberian Peninsula could be due to topographic effects. LIW in the Iberian Peninsula tend to occur at higher elevation than in Greece (775 m and 319 m, respectively). Therefore, meteorological conditions at 850 hPa would have a greater influence on ignition, survival and arrival of fires in the Iberian Peninsula than in Greece. 335 We have also calculated the median value of the variables plotted in 6 using A≥0.8 and A≥0.85. We have obtained the same results as using A≥0.7 in terms of difference between meteorological variables of fire-igniting lightning and typical lightning.
However, the p-values are significantly higher, as the total number of fire-igniting lightning included in the sample decreases significantly when increasing the minimum value of A.
The temperature difference between vertical levels can be used to investigate the instability of the atmosphere. Let us now 340 analyze the temperature differences between 2 m altitude and 850 hPa pressure level and between 850 hPa and 450 hPa pressure levels. We have found that the median temperature differences between 2 m altitude and 850 ::: 700 hPa pressure level for CG lightning in the Iberian Peninsula and Greece are, respectively, 5.6 :::: 16.8 K and 7.5 :::: 19.0 K. In the case of fire-igniting lightning, the median temperature differences between 2 m altitude and 850 ::: 700 hPa pressure level are 6.3 K and 8.5 K, respectively.
The median temperature differences between 850 hPa and 450 hPa pressure levels for CG lightning in the Iberian Peninsula 345 and Greece are 33.5 ::: 18.8 K and 33.1 :::: 19.9 K, respectively, while in the case of fire-igniting flashes the median temperature differences between 850 hPa and 450 hPa pressure level for the Iberian Peninsula and Greece are, respectively, 35.4 K and 33.2 K. These results indicate that the instability of the atmosphere ::::: lower :::::::::: troposphere is higher in fire-igniting lightning flashes than in typical CG lightning flashes. However, we have found some differences between the Iberian Peninsula and Greece.

Meteorological conditions in the upper troposphere
Let us now investigate the meteorological conditions of thunderstorms at altitudes above the 850 hPa pressure level. Hydrometeors (in the shape of water or ice) can play an important role for the lightning activity (Deierling et al., 2005;Finney et al., 380 2014) and for the precipitation rate (Tao et al., 2012). We plot in Fig. 8 and Fig. 9 the vertical profiles of the specific cloud ice water content and the specific rain water content for the CG lightning and the fire-igniting lightning climatologies during May and September in the Iberian Peninsula and Greece, respectively. We plot in the second column of Fig. 8 and 9 the vertical profile of the p-values to indicate the pressure levels where differences are statistically significant.
The total cloud ice is lower for the case of fire-igniting lightning than for the CG lightning climatology (Fig. 8(a) and 385 Fig. 9(a), except at altitudes above 450 hPa pressure levels in the case of Greece. The lower content of ice particles in fireigniting lightning suggests that fire-producing thunderstorms have a lower content of moisture than the climatological median.

Cloud Top Height values
Let us now compare the CTH values associated with fire-igniting lightning and with typical lightning using the CTH product provided by EUMETSAT and based on measurements of the MSG satellites. Thunderstorm with high CTH values are related 420 with high level of convection and lightning activity (Price and Rind, 1992). We have collected the value of the CTH for all CG lightning flashes and fire-igniting lightning flashes for the periods between 2012 and 2015 (Iberian Peninsula) and between and when the reported quality of the estimation of CTH is not poor. As an example, we plot in the supplement the CTH map derived from MSG satellite for one fire-igniting thunderstorm taking place in the Mediterranean coast of the Iberian Peninsula 425 in June 15, 2015 at 14:15 UTC.

Lightning flash frequency and density in fire-producing thunderstorms
Lightning flash frequency and density is used as a proxy variable to forecast or parameterize the inception of lightning ignited the lightning activity just starts to increase from the regime with the lowest lightning activity to the regimen with the highest lightning activity. The regime with a higher lightning activity coincides with a higher precipitation rate (Soriano et al., 2001) that contributes to inhibit the possibility of fire ignition. We have found no relationships between the cumulative number of different type and polarity of lightning discharges and the time of ignition.
Several authors have investigated the spatial density of CG flashes 24 h prior to the ignition as a proxy for the occurrence 480 of fires in the US depending on the 100-h fuel moisture information (e.g., Latham, 1989;Hardy et al.;Schultz et al., 2019).
They have found a that the threshold value of the CG spatial flash density to ignite a fire and allow it to reach the arrival phase would strongly depend on the lightning detection system. In this work, we have calculated the CG spatial flash density 5 km Therefore, is the most recent . We obtain a median value of 0.44 flashes per km −2 , a 25th percentile of 0.26 flashes per km −2 , and a 75th percentile of 0.69 flashes per km −2 . In addition to meteorological parameters, these values could be used in regional fire forecasting or atmospheric models to quantify the risk of ignition by thunderstorms, specially if radar measurements of precipitation rate are not available.

Long-Continuing-Current (LCC) lightning
In this section, we investigate the climatology of LCC lightning flashes over Europe derived from ISS-LIS data between March 2017 and September 2020 and the meteorological and cloud conditions under which they take place. This approach will help us to identify possible relationships between LCC lightning and fire-igniting lightning flashes. We plot the obtained geographical distribution of LCC(>20 ms)-lightning flashes in Fig. 15 together with the geographical distribution of lightning flashes reported by ISS-LIS. According to this figure, LCC(>20 ms)-lightning flashes tend to occur 505 over the oceans and over coastal regions, even when the maximum occurrence rate of total lightning flashes is produced over land. Lightning flashes tend to be more energetic over ocean than over land (Said et al., 2013;Holzworth et al., 2019), which is in agreement with the different ratio of LCC(>20 ms)-lightning flashes to total lightning flashes plotted in the top-right panel of Fig. 15.

LCC lightning distribution over Europe
Comparison of LIW maps (Fig. 2) and LCC(>20 ms)-lightning maps (Fig. 15) in the Iberian Peninsula and Greece is difficult 510 as a consequence of the low total number of reported LCC(>20 ms)-lightning flashes by ISS-LIS. However, it is interesting to highlight the high occurrence of LCC(>20 ms) flashes and fire-igniting lightning in the Mediterranean coast of the Iberian Peninsula. In addition, we have obtained a high occurrence of LIW and LCC(>20 ms)-lightning flashes over the Greek coastal regions and over some parts of the Pindus Mountains. Fig. 16 shows the monthly occurrence of typical lightning and LCC(>20 ms)-lightning flashes over land in Europe. The 515 peak in the occurrence rate of LCC(>20 ms)-lightning flashes is reached during the summer season, while the peak in the ratio of LCC(>20 ms)-lightning to typical lightning flashes occurs during the winter. The total number of flashes during winter is lower than during summer, causing that the ratio of LCC(>20 ms)-lightning to typical lightning flashes oscillates more during winter than during summer. Winter thunderstorms are characterized by weak updrafts, which suggests that weaker convection in thunderstorms favors the occurrence of LCC(>20 ms)-lightning flashes, as proposed by Bitzer (2017).
We plot as red dots the LCC(>20 ms)-lightning flashes in thunderstorm with CBH>2000 m in the first panel of Fig.15.
The geographical distribution of LCC(>20 ms)-lightning flashes in thunderstorm with CBH>2000 m suggests that they tend to occur over mountains. However, a climatology covering a larger number of years is needed to explain the geographical distribution of LCC(>20 ms)-lightning flashes with CBH>2000 m. 555 Fig. 19 shows the frequency distribution of CTH values of typical and LCC(>20 ms)-lightning flashes reported by ISS-LIS over Europe between 2017 and 2019. The median CTH value of LCC(>20 ms)-lightning flashes is slightly below the median CTH value of typical lightning flashes (11.1 km). These results suggest that LCC(>20 ms)-lightning flashes tend to occur in thunderstorms with weaker convection than the median, as suggested by Bitzer (2017) for LCC(>10 ms)-lightning. The p-value is below 0.05, indicating that differences in medians of the showed distributions are statistically significant.  We plot +CG flashes in the left panel.

Conclusions
The main objective of this work was to shed light into the following questions: What is the relationship between the occurrence of dry thunderstorms and LIW fires in the Mediterranean basin? What is the role of Long-Continuing-Current (LCC) lightning flashes in the ignition? 650 We have investigated the meteorological and cloud conditions of fire-igniting lightning over the Iberian Peninsula and Greece between 2009 and 2017 and between 2017 and 2019, respectively. In addition, we have analyzed for the first time the relation-ship between LCC(>20 ms)-lightning determined from by ISS-LIS data over Europe with a continuous current lasting more than 20 ms and fire-igniting lightning. We have focused our analysis on LCC(>20 ms)-lightning flashes because the continuous phase of the discharge can transfer a significant amount of energy to trees, producing ignition. We have developed the first 655 LCC(>20 ms)-lightning map over the Mediterranean basin based on optical measurements of ISS-LIS between 2017 and 2020.
However, more data are needed to produce a climatology.
We have searched the lightning candidate of each fire using a proximity index that combines the holdover and the distance between flashes and fires. The fire databases used in this study have been provided by national institutions from Spain, Portugal, 2. Both fire-igniting flashes and LCC(>20 ms)-lightning flashes tend to occur during the brief transition phase when the lightning activity increase from the regimen with the lowest lightning activity to the regimen with the highest lightning activity.
4. We have found that the temperature at 2 m altitude and at 850 hPa pressure level is higher than the climatological value in high-based thunderclouds (CBH larger than 2000 m) producing LCC(>20 ms) flashes, coinciding with the meteorological conditions of fire-producing thunderstorms. In total, about a third of all the analyzed LCC(>20 ms)-680 lightning flashes occurs in high-based thunderclouds. This result suggests that parameterizing and/or monitoring the occurrence of LCC(>20 ms)-lightning flashes could serve to improve fire forecasting models and LIW parameterizations.
Our analysis suggests that LCC-lightning occurrence over Europe can be parameterized in atmospheric models using meteorological variables as proxy. Such a parameterization can be used in future studies to improve the modeling of fire occurrence and its atmospheric emissions in such models, where different atmospheric variables are used as proxies for the occurrence 685 of lightning [e.g., (Tost et al., 2007;Murray et al., 2012;Gordillo-Vázquez et al., 2019)]. The launch of the Meteosat Third Generation (MTG) geostationary satellites of the EUropean organization for the exploitation of METeorological SATellites (EUMETSAT) in 2022 will provide for the first time a continuous monitoring of the occurrence of lightning flashes and fires in Europe and Africa through the instruments Lightning Imager (LI) and Flexible Combined Imager (FCI) from 2023, after the commissioning phase (Stuhlmann et al., 2005). MTG-LI will also provide for the first time a climatology of LCC flashes 690 over Europe and Africa, enabling us to investigate the relationships between LIW and LCC flashes in both continents. New flash and fire climatology provided by the MTG-LI and the MTG-FCI together with meteorological measurements will mean a substantial advance in the study of the meteorological conditions of LIW in the Mediterranean basin.
Data availability. All data used in this paper are directly available after a request is made to authors F. J. P. I (FranciscoJavier.          Ratio LCC(>20 ms)-flashes/Normal-flashes