This section is divided into two parts. The first part describes a particular thunderstorm that occurred on 10 June 2019. We selected this thunderstorm because on that date, the observer recorded a thunder less than 1 s after he saw the lightning flash at the Milešovka observatory. According to geographical coordinates in EUCLID data, the lightning discharge occurred at a distance of only 65 m from the observatory. This lightning discharge was the second closest in our dataset; on 1 June 2018, there was a lightning detected directly at the observatory, which hit the observatory according to the book of records written by observers. Nevertheless, we do not detail this thunderstorm (on 1 June 2018) in this study since Sokol et al. [33
] already studied it in detail.
Contrary to the first part of this section, which is dedicated to one particular thunderstorm, the second part of this section presents common characteristics, including LDR, that were typical throughout all thunderstorms that occurred in the 38 days in the dataset (Table 3
). It presents results from the comparison of cloud radar data with lightning data in dependence on the distance of lightning discharge to the Milešovka observatory (i.e., NL vs. FL).
3.1. Thunderstorm on 10 June 2019
shows the temporal evolution of rain rates (as derived from C-band weather radar data, Figure 2
a) and lightning discharges (Figure 2
b) during one hour of the thunderstorm on 10 June 2019, when most lightning activity was observed close to the Milešovka hill. The temporal evolution is displayed with a time step of 10 min. Based on rain rates (Figure 2
a), it is clearly visible that the thunderstorm was severe, at least within the central European context.
also shows that the lightning discharges occurred (horizontally) not only within the precipitation areas but also outside of the precipitation cores; i.e., they may have originated in non-precipitating parts of the thundercloud as well. As the system moved in time towards east-northeast, CC flashes had a tendency to precede CG flashes, while lightning flashes (CC + CG together) tended to occur not only during the period of intense rainfall, but also prior to intense rainfall. This has been mentioned in other works as well, e.g., [40
depicts precipitation totals with a time step of 1 min, as registered by a rain gauge with a resolution of 0.1 mm at the Milešovka observatory between 21 and 22 h UTC on 10 June 2019. Note that rain rates (i.e., precipitation intensities, Figure 2
a) cannot be directly compared with recorded precipitation totals (Figure 3
) as they do not represent the same information. Figure 3
shows that the highest 1-min precipitation total occurred between 21:39 and 21:40, while the closest lightning was recorded at 21:37 and 50.341 s. This confirms that lightning flashes may occur prior to heavy rain. Here, we note that measured precipitation totals by the rain gauge might be underestimated. The reason is that at the top of Milešovka hill, strong winds frequently appear during storms, which may result in an underestimation of rainfall totals due to the blowing away of the precipitation from the surface of the rain gauge.
Concerning data of the cloud radar, Figure 4
displays radar reflectivity together with NL (i.e., lightning discharges that occurred not farther than 1 km from the Milešovka observatory). It is obvious from the figure that NL were related to high reflectivity values, although high reflectivity values were also typical for the melting layer and below (i.e., below 2.5 km approximately).
It is worthy of note that between 21:30 and 21:40 approximately, there was a sudden decrease in the vertical span of the thundercloud, according to the cloud radar data (Figure 4
). This is the time when most of rainfall was registered at the Milešovka observatory (Figure 3
). This is probably caused by the fact that the received signal in the lowest gates was too strong due to heavy rain that the radar was unable to capture signal from higher gates.
Furthermore, it is interesting to check the temporal evolution of LDR values during the thunderstorm (Figure 5
) that were not corrected using the integrated cross-polarization ratios [41
]. In Figure 5
, high values of LDR clearly show that the melting layer was around 2500 m above ground in the thunderstorm. Another zone of high LDR values is visible from 21:30 to 21:50 at higher altitudes, which is the time interval of intense rainfall (Figure 3
) and lightning activity near the radar site (Figure 4
). Contrary to very high LDR values in the melting layer, which are commonly associated with melting snow flakes, very high LDR values at higher altitudes, such as 4-7 km, can correspond to non-spherical shape of graupel and/or hail or to aligned ice crystals due to a strong electric field if the crystals are not aligned along with the co-channel, instead they are oriented at angle close to 45° with both the co- and cross-channels [25
]. It is worthy of note that the elevation around 4-7 km, where we observed increased LDR, is also considered as the elevation where the main negatively charged area appears [40
]. We discuss this finding further below.
It should be noted that the LDR data are not available at all gates where we obtained radar reflectivity factor data (for example, after 21:50). This is the consequence of the attenuation of the signal received in the plane perpendicular to the transmission plane.
shows the evolution of hydrometeor distribution (resulting from Hclass, Section 2.3
) on 10 June 2019 from 21:00 to 21:59 UTC. Clearly, the lack of data in the vertical profiles between approximately 21:30 and 21:40 makes the interpretation of the obtained results difficult, especially because many NL discharges occurred at that time. Nevertheless, the majority of the ten closest discharges (Figure 6
) occurred after 21:40, when we had data again, covering almost all the troposphere.
The results of Hclass indicate that the highest LDR values at the elevation from 4 to 7 km (Figure 5
) correspond to a mixture of several hydrometeor species with a predominance of ice and snow particles and graupel. These are the species which play major roles in the process of cloud electrification by collisions of hydrometeors according to currently accepted theories [44
]. The mixture of many hydrometeor species is also evident during very close lightning activity (between 2400 s and almost 3000 s in Figure 6
The electrification process by collisions of hydrometeors at an elevation of 4–7 km is also supported by Figure 7
, which presents values of the Doppler spectrum width (σ). High values of σ, i.e., large variability of vertical velocities, just after 21:40 confirm the coexistence of various hydrometeors and support the existence of collisions of hydrometeors (light species collide with heavier species having larger terminal velocity). The obtained results of high LDR and sigma values together with the presence of diverse hydrometeor species may bring us to the conclusion that around 21:40, collisions of hydrometeors caused a strong electrification of the thundercloud near the radar site.
3.2. Common Characteristics of Analyzed Thunderstorms
This subsection focuses on results related tall thunderstorms in the dataset (Table 3
). It shows their common (different) features and compares them with recent knowledge on lightning processes. Our intention was to compare NL with FL, when clouds were present above the observatory. Therefore, in the statistical evaluation, we used data from only those gates, where Hclass identified at least one hydrometeor species.
summarizes the results throughout the analyzed thunderstorms at the Milešovka observatory. It depicts radar-derived quantities for NL discharges compared to that for FL discharges. It clearly shows that on average, hail, rain and graupel occurred in lower gates more frequently during NL as compared to FL. For FL, rain and graupel were almost not detected at all. For NL, hail concentration was higher at an elevation of 2000 to 2500 m above ground. This is the level which roughly corresponds to the melting layer (Figure 4
). Rain concentration was higher at lower elevations, at 1800 m approximately. Thereby, it can be suggested that the closer the lightning, the higher the concentration of rain and hail. This agrees with our previous results based on 10 thunderstorms [45
In addition, Figure 8
displays the results of AV for NL vs. FL. It shows that in the case of NL, the downward motion of the air substantially prevails at lower altitudes; from the ground to 1000 m and from 2000 to 3000 m. The layer between 1000 and 2000 m above ground is characterized by fluctuations in AV, which can be related to an interchange of up- and downdrafts. Updrafts mostly dominate the elevation from 3000 m upwards. Slow updrafts are typical for very upper vertical levels (above 9500 m).
Overall, AV seems to be quite low, which is caused by averaging. The variability in AV among gates seems high for NL. This is caused by much lower number of NL discharges (990) as compared to FL discharges (171,754), as shown in Figure 9
. Concerning FL, AV does not fluctuate much on average between neighboring gates, which is due to large number of processed data. Figure 8
also shows that for FL, downward motion prevails from the ground up to 3000 m, while upward motion dominates the layers above 3000 m on average.
Taking into account the distance of the lightning from the observatory, we do not know whether the radar measurements took place in the frontal or back side of the thunderstorms or on their lateral sides. The placement within the thunderstorm may lead to diverse directions and values of AV, which can be confirmed by high variability of AV (not depicted). In addition to the uncertainty regarding the localization of measurements with respect to the movement and development of thunderstorms, it should be emphasized that we present results and quantities that are derived indirectly (i.e., not directly measured). Therefore, the results cannot be explicitly verified. However, we can state that the obtained results are in accordance with the general knowledge about thunderstorms. Therefore, we believe that the technique used to calculate the vertical air velocity and to classify hydrometeors give realistic results.
3.3. LDR during Analyzed Thunderstorms
Averages of LDR are depicted in Figure 10
for NL, as compared to FL. For both NL and FL, the melting layer is not pronounced in LDR averages; there is no obvious increase in LDR averages in lowest gates. This is very likely related to the fact that the height of the melting layer depends on current atmospheric conditions, which change from one thunderstorm to another and might also change during one particular thunderstorm. As a consequence, the height of the melting layer becomes smooth in averaging, making it imperceptible in the figure.
The character of the curves in Figure 10
(their oscillation) is influenced by the number of averaged cases (Figure 9
). This is especially true for the red curve representing NL discharges. The isolated maxima of LDR averages are probably random. However, Figure 10
clearly depicts that at an elevation of 4 to 6.5 km approximately, there are large LDR averages, which show little oscillations for NL, thus they do not correspond to random processes. These averages are much larger than the LDR averages for FL. As in Section 3.1
, we attribute it to electrification by collisions and alignment of ice crystals.
To better assign the cause of increased LDR averages in the middle troposphere, Figure 11
shows 1 km layers of frequency of LDR in profiles with similar distribution of graupel and hail (i.e., rounded hydrometeors). Because concentrations of graupel and hail are similar in both NL and FL (ice or snow being present almost everywhere in these 1 km layers, Figure 8
), it is obvious that there had to be another process that made the higher LDR more frequent in the case of NL as compared to FL. We suggest that the additional process could be the alignment of ice crystals observed by other researchers, e.g., by Melnikov et al. [25
]. However, we are aware that this hypothesis cannot be exactly verified.
At an elevation between 8 and 9 km, significantly higher averages of LDR for NL, as compared to FL, could be rather random because of high oscillations of that for NL. On the other hand, the high oscillations of LDR averages for NL can also be related to the orientation of aligned ice crystals in an electrified field. LDR can increase if the particles align at an angle close to 45° from both the co- and cross-channels, while it can decrease if the particles align along with the co-channel (LDR reaches large negative values). Thus, the LDR of non-spherical targets, such as ice crystals, can have strongly different values (large and small) depending on the azimuth direction to the channels. Therefore, the variability of LDR may be increased in the case of NL.
The results also suggest that the clouds producing lightning in the vicinity (NL) are vertically massive and higher than clouds producing FL, at least in our dataset.