Cloud Flashes Research Articles

Imaging of 3 bright terrestrial gamma-ray flashes by the atmosphere-space interactions monitor and their parent thunderstorms

The Atmosphere-Space Interactions Monitor (ASIM) on the International Space Station (ISS) includes an instrument designed to geolocate Terrestrial Gamma-ray Flashes (TGF) produced by thunderstorms. It does so with a coded aperture system shadowing the pixelated Low Energy Detector of the Modular X- and Gamma-ray Sensor (MXGS). Additionally, it locates associated lightning flashes with the Modular Multispectral Imaging Array (MMIA). Here we present 3 bright TGFs with very similar duration, fluency and observation distance. The innovative imaging capabilities allow us to determine the TGF position and correlate the TGF-lightning parent event in time and position simultaneously. The accurate position determination and distance to the observer allow us to perform a comparative study of TGF characteristics. These TGFs were produced in association with lightning flashes below the highest cloud tops of developing to mature convective cells. In one event, GLM (Geostationary Lightning Mapper) cloud flash rates were slowing down after the TGF while negative cloud-to-ground flashes suddenly ceased from 10 to 5 min before the TGF. An 8-stroke (strongest: -93 kA) cloud-to-ground flash occurred 31 s before the TGF. Vertical profiles from the ERA5 reanalysis data suggest TGFs may be produced in variety of tropical environments.

Analysis of Inverted Charge Structure and Lightning Activity during the 8.14 Local Hailstorm on the Qinghai–Tibet Plateau

In this paper, the charge structure and lightning activity characteristics of the thunderstorm that occurred on the Qinghai–Tibet Plateau on 14 August 2014 were analyzed using data collected from a three-dimensional (3D) lightning very-high-frequency (VHF) radiation source location system and Doppler weather radar. The analysis results showed that the charge structure of the hailstorm was maintained as an inverted dipole throughout the thunderstorm’s development process. The negatively charged region height was distributed in the 5–7 km range (above ground level (AGL)), and the positively charged region was distributed from 2 to 5 km (AGL). The lightning flash types included only cloud flashes and negative cloud–to–ground (CG) flashes in the hailstorm, and cloud flashes accounted for 93% of the total lightning flashes. Cloud flashes accounted for a high proportion of the total flashes, which may have been related to the deep lower positively charged region observed throughout the thunderstorm process. In the hailstorm development stage, the electric field was dominated by positive polarity. When the hail fell, the electric field changed negatively. When the hail ended, the electric field was dominated by negative polarity. A hail event occurred only once and lasted for a long time in the development stage, but in the mature stage, hail fell many times and every time for only a short time, and in the dissipating stage, hail events also occurred many times and each time for a long time. By comparing the radar echoes of the hailstorm cells and normal thunderstorm cells, we found that the area of the 50 dBZ echo in the hailstorm was small, the occurrence time was late, and the duration was short.

MCS Stratiform and Convective Regions Associated with Sprites Observed from Mt. Fuji

Five sprites produced above the stratiform region and one sprite above the convective region in the mesoscale convective system (MCS) were observed on 22 July 2013 from the summit of Mt. Fuji. Four of the five sprites occurred in the stratiform region after the stratiform region was cut off from the convective region. The five sprite-producing +CGs (SP + CGs) occurred in and close to these cells with low lightning activity, although in-cloud lightning locations extensively spread out in the stratiform region. This suggests that positive charges were sufficient to produce the SP + CGs that accumulated extensively in the stratiform region due to a local (in situ) charge generation mechanism. One sprite occurred above the MCS convective region, and the SP + CG was also located at the MCS convective region. Video images indicated that the optical cloud flash associated with the SP + CG continued at least 150 ms until sprite emission. The optical cloud flash light occurred repeatedly at strong and weak intensities, and this characteristic indicated that the M components in the lightning occurred in the cloud. The SP + CG might have gathered sufficient positive charges to produce the sprite with a continuing current.

Revisiting Lightning Activity and Parameterization Using Geostationary Satellite Observations

The Geostationary Lightning Mapper (GLM) on the Geostationary Operational Environmental Satellite 16 (GOES-16) detects total lightning continuously, with a high spatial resolution and detection efficiency. Coincident data from the GLM and the Advanced Baseline Imager (ABI) are used to explore the correlation between the cloud top properties and flash activity across the continental United States (CONUS) sector from May to September 2020. A large number of collocated infrared (IR) brightness temperature (TBB), cloud top height (CTH) and lightning data provides robust statistics. Overall, the likelihood of lightning occurrence and high flash density is higher if the TBB is colder than 225 K. The higher CTH is observed to be correlated with a larger flash rate, a smaller flash size, stronger updraft, and larger optical energy. Furthermore, the cloud top updraft velocity (w) is estimated based on the decreasing rate of TBB, but it is smaller than the updraft velocity of the convective core. As a result, the relationship between CTH and lightning flash rate is investigated independently of w over the continental, oceanic and coastal regimes in the tropics and mid-latitudes. When the CTH is higher than 12 km, the flash rates of oceanic lightning are 38% smaller than those of both coastal and continental lightning. In addition, it should be noted that more studies are necessary to examine why the oceanic lightning with low clouds (CTH < 8 km) has higher flash rates than lightning over land and coast. Finally, the exponents of derived power relationship between CTH and lightning flash rate are smaller than four, which is underestimated due to the GLM detection efficiency and the difference between IR CTH and 20 dBZ CTH. The results from combining the ABI and GLM products suggest that merging multiple satellite datasets could benefit both lightning activity and parameterization studies, although the parallax corrections should be considered.

Frequency spectra of various events pertinent to lightning cloud flashes obtained from wavelet transform technique and ratified by narrow band measurement technique

In the present study, frequency spectra of the electric field corresponding to the various cloud events, such as, initial breakdown process, regular pulse bursts, chaotic pulse trains and recoil streamers have been analysed. For the purpose, electric field radiated by cloud flashes were obtained simultaneously by a wide bandwidth antenna system and two narrow bandwidth antenna systems tuned at 3 MHz and 30 MHz. The frequency spectra of the broad band electric field signatures were obtained by using the wavelet transform technique and were compared with the magnitudes of the narrow band signals at the given central frequencies. To the best of our knowledge, it is the first study in which frequency spectra is obtained by transforming the time domain signal using wavelet transform technique and ratified by narrow bandwidth system for the cloud flashes. Fifteen cloud flashes pertinent to the Swedish thunderstorms were selected for the purpose. It is found that, the cloud flashes radiate at frequencies as low as 3 kHz to as high as a few tens of Mega Hertz (MHz). Electric field radiation corresponding to the initial breakdown process were found to radiate in the frequency range of 50 kHz to 5 MHz on the average, maximum energy is being radiated in the frequency range of 500 kHz to 5 MHz. Similarly, the final stage corresponding to the regular pulse bursts was found to radiate in the frequency range of 50 kHz to 5 MHz and that corresponding to the chaotic pulse trains was found to be in the range of 100 kHz to 5 MHz. Whereas, the very narrow pulses at the final stage, that can be termed as pulses corresponding recoil streamers (or Q-streamers) were found to radiate in the frequency range of 50 kHz to well above 10 MHz. Q-streamers can be considered as the strongest source of very high frequency and is justified by the simultaneous measurement of the electric fields at very high frequency (30 MHz) narrow bandwidth system. Rectification of the biasness of the conventional wavelet power spectrum has also been performed, however, no significant change in the spectrum was observed. Therefore, this study provides a strong basis for applying the wavelet transform technique without employing large number of narrowband system to acquire frequency domain information of lightning phenomena.

An In‐Depth Analysis of Lightning Trends in Hurricane Harvey Using Satellite and Ground‐Based Measurements

AbstractSatellite lightning measurements from the Geostationary Lightning Mapper (GLM) and ground‐based measurements from Earth Networks Total Lightning Network (ENTLN) for Hurricane Harvey (2017) are combined to obtain a better representation of intra‐cloud and cloud‐to‐ground lightning and their relationship to intensification and weakening of the hurricane. Through this combination, detection efficiencies of ENTLN and GLM are calculated for the rainband and eyewall of Harvey. In addition, the combined data set is used to derive the cloud flash fraction (CFF) and cloud pulse fraction (CPF) throughout the study period. The results show that flash and pulse rates increase in both the eyewall and rainband prior to rapid intensification in Harvey. The rainband lightning has the highest correlation to Harvey's intensification, where increases in lightning correspond to the largest periods of wind‐speed intensification. The eyewall CFF is also highly correlated to intensification, with minima in eyewall CFF and CPF coinciding with slower intensification periods in Harvey, while increasing CFF and CPF corresponds to more rapidly intensifying periods. Further, GLM flash characteristics including the events per flash, groups per flash, maximum group area, maximum number of events per group, flash footprint, flash energy, and flash duration are tracked and compared to Harvey's wind‐speeds. Nearly all seven flash characteristics show an increase prior to rapid intensification, with the strongest correlation to intensification found with the rainband lightning spatial characteristics. This research represents a step forward in the application of lightning data that can be used in future studies of tropical cyclone intensification and weakening.

Frequency Spectra of Various Events Pertinent to Lightning Cloud Flashes Obtained from Wavelet Transform Technique and Ratified by Narrow Band Measurement Technique

Frequency Spectra of Various Events Pertinent to Lightning Cloud Flashes Obtained from Wavelet Transform Technique and Ratified by Narrow Band Measurement Technique

Investigation on the Lightning Location and Warning System Using Artificial Intelligence

An in‐depth study on a lighting location system is conducted in this paper. Firstly, the history and application of this system are summarized. The overall structure is detailed, including the detection principle of the lightning location, the orientation method, the detection circuit, the method of discriminating cloud flash and ground lightning signal, the error analysis, the guideline for station deployment, the preprocessing of the central station, and the function and structure of data server and user interface. The development of a lightning monitoring system in China is presented, and the construction of a new generation of a lightning location system in the Hubei Province power grid is introduced. Through the collection of measured data, the performance of the lightning location system in the lightning accident inspection rate, lightning location, and lightning situation statistics are analyzed. Artificial intelligence algorithms are applied in the lightning warning system. The new system has a high predicting accuracy.

Orientation of Initial Breakdown Pulses and Leader Discharges by Magnetic Direction Finder

AbstractFor decades, the magnetic direction finder (MDF) technique has been used to give the azimuth of return stroke (RS) in cloud‐to‐ground lightning (CG). Due to the possible polarization error produced by inclined lightning channels, cloud flashes are excluded by most MDF waveform discrimination procedures. In this study, the capability of the orthogonal magnetic coils in determining the azimuth of the initial breakdown pulses (IBPs) and leaders in CG is analyzed and discussed. First, a lightning orthogonal magnetic field sensor with high sensitivity (about 33 nT/V) as well as broad bandwidth (3 kHz to 2 MHz) has been developed. According to the transmission line model, RSs with current 10 kA occurred within 500 km could be detected by this system theoretically. Thereafter, a theoretical analysis was derived in order to evaluate its polarization error with arbitrarily oriented lightning channels. In order to overcome the effect of random nonvertical channels on orientation results, a statistical MDF method is proposed. A field lightning observation was conducted in the year 2017, in which RSs, leaders, and IBPs in CG or cloud flash from two thunderstorms are oriented by the proposed MDF technique. The average azimuth difference between the mathematic expectation of leaders and corresponding RSs is 2.4°. Most of the azimuth in IBPs lies within the region of radar composite reflectivity above 45 dBZ, which indicates that the initiation location of lightning flashes in this study mostly located in these strong echo regions.

Classification of VLF/LF Lightning Signals Using Sensors and Deep Learning Methods.

Lightning waveform plays an important role in lightning observation, location, and lightning disaster investigation. Based on a large amount of lightning waveform data provided by existing real-time very low frequency/low frequency (VLF/LF) lightning waveform acquisition equipment, an automatic and accurate lightning waveform classification method becomes extremely important. With the widespread application of deep learning in image and speech recognition, it becomes possible to use deep learning to classify lightning waveforms. In this study, 50,000 lightning waveform samples were collected. The data was divided into the following categories: positive cloud ground flash, negative cloud ground flash, cloud ground flash with ionosphere reflection signal, positive narrow bipolar event, negative narrow bipolar event, positive pre-breakdown process, negative pre-breakdown process, continuous multi-pulse cloud flash, bipolar pulse, skywave. A multi-layer one-dimensional convolutional neural network (1D-CNN) was designed to automatically extract VLF/LF lightning waveform features and distinguish lightning waveforms. The model achieved an overall accuracy of 99.11% in the lightning dataset and overall accuracy of 97.55% in a thunderstorm process. Considering its excellent performance, this model could be used in lightning sensors to assist in lightning monitoring and positioning.

Characteristic Features of Electric Fields Radiated by Cloud Flashes in Himalayan Region

Electric fields radiated by cloud flashes that occurred over the rugged terrain of mountainous country Nepal were recorded, analysed, and compared with those from different geographical regions. The total duration of the flash varies from 80 to 469.5 ms. The majority of the cloud flashes were of two stages: the first stage of the majority of the flashes was found to consist of large microsecond scale bipolar pulses having negative initial polarity and the late stage consists of submicrosecond scale pulses having both positive and negative initial polarities. The average durations of the first and second stages are 11.23 and 66.79 ms, respectively, and the time gap between them is 53.57 ms. The cloud pulses led by the negative initial polarity pulses are more compactly distributed and are higher in mountainous countries as compared to those of flashes lead by positive initial polarity pulses, with the average values of interpulse interval being 211.42 and 309.79 μs, respectively.

Lightning flash density in Europe based on 10 years of ATDnet data

Abstract The first decadal lightning flash climatology for Europe is presented using ATDnet long-range lightning detection network operated by the Met Office. More than 148 million lightning flashes were detected by ATDnet in the study area between 2008 and 2017. The dataset contains up to 90% of cloud-to-ground flashes and 25% of cloud flashes in Europe during the study period. All maps were computed with a spatial resolution of 0.2°. The annual flash density was between 0.5 and 3.0 flashes per km−2 y−1 in most of Europe with a maximum of 7.8 flashes km−2 y−1 in the northeast of Italy. Most of continental Europe experienced 20–40 thunderstorm days annually but a small area on the border of Armenia and Turkey experienced up to 87.6 thunderstorm days. This number exceeds previous estimates of European maximum thunderstorm frequency. The North Atlantic Ocean had only up to 10 thunderstorm days and 0.1 flash km−2 y−1. Flash density per thunderstorm day varied from 0.1 fl km−2 in the northeast of Italy. Occasional intense storms were found to happen even in areas without regular thunderstorm activity, including the oceans to the north of the Arctic Circle. Most of the study area had a mid-latitude specific summer peak in lightning activity whereas the Mediterranean experienced an autumn maximum. From March to August, most of the lightning in Europe occurred over land whereas from September to February it was concentrated over the Mediterranean. The highest lightning frequency over land was observed during the afternoon hours and the lowest frequency at night. At sea the diurnal distribution of lightning peaked either at night or in the morning hours.

Performance Analysis of Filtered VHF Signals Captured by Lightning Interferometer System

A lightning interferometer is a system that can determine the location of Very High Frequency (VHF) radiation of lightning. Such system has been studied and used for several decades ago. In this project, VHF Interferometer is used to determine the location of VHF radiations emitted by Narrow Bipolar Event (NBE) lightning flash by correlating the signals that received at three antennas. In order to get the better result, cross-correlation algorithm technique is employed to process the signals. This algorithm can determine the time delay between the antennas baselines and then determine the azimuth and elevation angles which are used to locate the location of the NBE flash. This interferometer system also can act as an earlier warning system as it detects the first cloud flash and a forecast can be made before cloud-to-ground flashes occur. Therefore, this interferometer system is very useful for the public citizens in protecting themselves from lightning strikes.

Optical observations of electrical activity in cloud discharges

Temporal variation of the luminosity of seven natural cloud-to-cloud lightning channels were studied, and results were presented. They were recorded by using a high-speed video camera with the speed of 5000 fps (frames per second) and the pixel resolution of 512 × 512 in three locations in Sri Lanka in the tropics. Luminosity variation of the channel with time was obtained by analyzing the image sequences. Recorded video frames together with the luminosity variation were studied to understand the cloud discharge process. Image analysis techniques also used to understand the characteristics of channels. Cloud flashes show more luminosity variability than ground flashes. Most of the time it starts with a leader which do not have stepping process. Channel width and standard deviation of intensity variation across the channel for each cloud flashes was obtained. Brightness variation across the channel shows a Gaussian distribution. The average time duration of the cloud flashes which start with non stepped leader was 180.83 ms. Identified characteristics are matched with the existing models to understand the process of cloud flashes. The fact that cloud discharges are not confined to a single process have been further confirmed from this study. The observations show that cloud flash is a basic lightning discharge which transfers charge between two charge centers without using one specific mechanism.

Initiation and propagation of cloud-to-ground lightning observed with a high-speed video camera.

Complete evolution of a lightning discharge, from its initiation at an altitude of about 4 km to its ground attachment, was optically observed for the first time at the Lightning Observatory in Gainesville, Florida. The discharge developed during the late stage of a cloud flash and was initiated in a decayed branch of the latter. The initial channel section was intermittently illuminated for over 100 ms, until a bidirectionally extending channel (leader) was formed. During the bidirectional leader extension, the negative end exhibited optical and radio-frequency electromagnetic features expected for negative cloud-to-ground strokes developing in virgin air, while the positive end most of the time appeared to be inactive or showed intermittent channel luminosity enhancements. The development of positive end involved an abrupt creation of a 1-km long, relatively straight branch with a streamer corona burst at its far end. This 1-km jump appeared to occur in virgin air at a remarkably high effective speed of the order of 106 m/s. The positive end of the bidirectional leader connected to another bidirectional leader to form a larger bidirectional leader, whose negative end attached to the ground and produced a 36-kA return stroke.

ATDnet Detection Efficiency and Cloud Lightning Detection Characteristics from Comparison with the HyLMA during HyMeX SOP1

AbstractThis study provides the first comprehensive quantitative detection efficiency (DE) estimation of the Met Office long-range very low-frequency (VLF) lightning location system, the Arrival Time Difference Network ATDnet, with a focus on cloud flash detection. ATDnet uses sensors in and around Europe to detect lightning in Europe, northern Africa, and northern parts of the Atlantic.In the present paper, ATDnet is validated against the Hydrological Cycle in the Mediterranean Experiment (HyMeX) Lightning Mapping Array (HyLMA) deployed in the south of France in autumn 2012 as a part of the HyMeX project Special Observation Period 1 (SOP1). Three storms (on 5, 11, and 25 September 2012) were selected for this study, as their activity was sufficient to provide good statistics but also low enough to avoid ambiguous situations due to temporally overlapping flashes.The overall ATDnet flash DE was approximately 89% for the 281 ground flashes [cloud to ground (CG)] and 24% for the 1324 cloud flashes [inter/intracloud (IC)] during the storms. It was also demonstrated that variations in cloud flash characteristics (e.g., flash vertical extent) have a significant impact on DE between and during the studied storms.

Features of positive ground flashes observed in Kathmandu Nepal

Lightning vertical electric fields pertinent to the subtropical thunderstorms occurring over the rugged terrain have been measured and recorded at a hilly station Kathmandu, Nepal. In the present work, waveforms of the positive ground flashes have been selected from all the records and were analyzed. To the best of our knowledge, this is the first time that fine structure of electric field signature pertinent to the positive return stroke; have been analyzed and presented from Nepal. One hundred and thirty three (133) of the total of four hundred twenty-five (425) flashes were selected from seven thunderstorm days and analyzed. Of the data recorded for seven days, 133 flashes (31.3%) were positive flashes and 276 flashes (64.9%) were cloud flashes. Majority of the positive ground flashes were found to be single stroke ones, whereas, the average number of strokes per flash is found to be 1.1 with a maximum value of 4. Majority of the positive ground flashes were found either lacking the initial breakdown process and the leader stage or these processes could not be detected. The return strokes are found to be succeeded by large in cloud activity in the continuing current portion of the flash. The average zero-crossing time of the positive return strokes was found to be 60.45µs with a range of 447.81µs and the average rise time was found to be 9.44µs with a range of 42.56µs.

Investigating the relationship of lightning activity and rainfall: A case study for Crete Island

The relationship of lightning activity and rainfall is investigated for rain events of variable intensity. Rain data from 22 gauging stations over the island of Crete and lightning activity from the Global Lightning Network including both cloud-to-ground and some cloud flashes are analyzed for the period September 2012 to June 2014. Local thunderstorms' characteristics are investigated both individually as well as in groups according to the results of k-means clustering algorithm in 3 dimensions (space (x, y) and time (t)) in which the number of clusters is decided by G-means algorithm. Correlation of non-zero pairs of rain intensity and number of flashes is examined at various time intervals, time lags and effective radii. Also, correlation of flash count within 50km radius around the stations is examined for the rain events of maximum hourly intensity for each gauging station. The highest coincidence of lightning clusters with intense rain events reaches 60% when gauges are 25–30km from the cluster center. Maximum correlation within non-zero pairs of rain intensity and flashes number is obtained for more intense rain (99th percentile) and for increased flash count within the searching area (more than 10 flashes). Also, correlation is stronger for shorter time windows. The findings of this study improve the understanding of thunderstorm events and could provide staple information for the improvement of forecasting extreme events.

Optical and electrical characteristics of in-cloud discharge activity and downward leaders in positive cloud-to-ground lightning flashes

The characteristics of five downward positive cloud-to-ground (CG) flashes were analyzed based on images from a high-speed video camera and electric field (E-field) changes from slow antenna and fast antenna systems. The flashes persisted for 740ms to 1250ms. The luminous durations of the leaders ranged from 4ms to 24ms. The average 2-D development speeds of the positive leaders ranged from 0.3×105m/s to 2.0×105m/s. The propagation speeds of the positive leaders increased as they approached the ground. Approximately 53.9% of the 89 observed positive CG flashes exhibited isolated monopolar pulses during the leader propagation and immediately prior to the return stroke. The average time interval between adjacent leader pulses was 15μs according to the E-field changes. Most leaders of positive CG flashes (approximately 67.4%) began in the presence of prolonged and intense in-cloud discharge (IC) activity that ranged from 100ms to 973ms. A cloud discharge may be conducive to the formation of positive CG flashes. Generally, a higher occurrence of positive CG flashes corresponded to a higher occurrence of cloud flash, which suggests that positive CG flash can be initiated by a cloud discharge. Four positive CG flashes exhibited one return stroke, and only one flash had two strokes.

Variability of CONUS Lightning in 2003–12 and Associated Impacts

AbstractChanges in lightning characteristics over the conterminous United States (CONUS) are examined to support the National Climate Assessment (NCA) program. Details of the variability of cloud-to-ground (CG) lightning characteristics over the decade 2003–12 are provided using data from the National Lightning Detection Network (NLDN). Changes in total (CG + cloud flash) lightning across part of the CONUS during the decade are provided using satellite Lightning Imaging Sensor (LIS) data. The variations in NLDN-derived CG lightning are compared with available statistics on lightning-caused impacts to various U.S. economic sectors. Overall, a downward trend in total CG lightning count is found for the decadal period; the 5-yr mean NLDN CG count decreased by 12.8% from 25 204 345.8 (2003–07) to 21 986 578.8 (2008–12). There is a slow upward trend in the fraction and number of positive-polarity CG lightning, however. Associated lightning-caused fatalities and injuries, and the number of lightning-caused wildland fires and burn acreage also trended downward, but crop and personal-property damage costs increased. The 5-yr mean LIS total lightning changed little over the decadal period. Whereas the CONUS-averaged dry-bulb temperature trended upward during the analysis period, the CONUS-averaged wet-bulb temperature (a variable that is better correlated with lightning activity) trended downward. A simple linear model shows that climate-induced changes in CG lightning frequency would likely have a substantial and direct impact on humankind (e.g., a long-term upward trend of 1°C in wet-bulb temperature corresponds to approximately 14 fatalities and over $367 million in personal-property damage resulting from lightning).