We previously documented geographic distributions of the optically brightest lightning on Earth – known as “superbolts” – using two space-based instruments: the photodiode detector (PDD) on the Fast On-orbit Recording of Transient Events (FORTE) satellite and the Geostationary Lightning Mapper (GLM) on NOAA’s newest Geostationary Operational Environmental Satellites (GOES). In this study, we further examine the superbolts identified by the PDD and GLM to reconcile the differences between their geographic distributions. We find that both the physical extent of the parent flash and the development speed of its leaders are important for making a superbolt. The oceanic PDD superbolts tend to occur early in flashes that rapidly expand laterally into long-horizontal “megaflashes.” The top GLM superbolts occur over land at later times in particularly large megaflashes that grow more slowly until they extend over multiple hundreds of kilometers. The FORTE PDD missed these delayed superbolts due to limitations in its triggering. Coincident TRMM measurements show that the warm season megaflash superbolts detected by LIS/GLM and Turman’s (1977) wintertime oceanic superbolts also observed by the PDD occur in otherwise similar thunderstorm environments. Both are marked by: low storm heights (< 10 km), widespread rainfall near the surface, small infrared brightness temperature gradients, and low flash rates. We suggest that the vertically-compact, stratiform nature of these clouds allows them to store more charge between flashes, providing favorable conditions for superbolt production.

We previously observed that long-horizontal lightning flashes exceeding 100 km in length, known as “megaflashes,” occur preferentially in certain thunderstorms. In this study, we develop a cluster feature approach for automatically documenting the evolutions of thunderstorm systems from continuous lightning observations provided by the Geostationary Lightning Mapper (GLM) on NOAA’s Geostationary Operational Environmental Satellites (GOES). We apply this methodology to GOES-16 GLM observations from 2018 to mid-2022 to improve our understanding of megaflash-producing storms. We find that megaflashes occur in long-lived (median: 14 hours) storms that grow to exceptional sizes (median: 11,984 km2) while they propagate across long distances (622 km) compared to ordinary storms. The first megaflashes are typically produced within 15 minutes of the storm reaching its peak intensity and extent, describing the transition to mature convection. Most megaflashes occur 13 hours after the initial megaflash activity, and are sufficiently close to convection to suggest initiation in the convective line (where GLM has difficulty detecting faint early light sources from these megaflashes). In-situ generated megaflashes are rare, accounting for 2.7% of the sample using a 50 km convective distance threshold, but also tend to larger than normal megaflashes, possibly due to having direct access to the electrified stratiform cloud through which megaflashes propagate.

It is important to understand connections between society and the natural environment for anticipating environmental hazards and anthropogenic effects on the broader Earth system. In this study, we conduct a detailed exploration of the interactions between oceanic thunderstorms and maritime traffic. Shipping traffic produces aerosols that perturb the otherwise “clean” ocean environment. Prior work proposed these aerosol effects as the cause of increased lightning activity over certain shipping lanes. However, introducing tall well-grounded objects into a high electric field environment might also facilitate lightning discharges, as we see with upward lightning over land. We consider both possibilities in this work. Our analyses of the thunderstorms responsible for the enhanced lightning activity over the shipping lane with the clearest anthropogenic signal indicate that the anthropogenic signature results from an increased frequency of lightning-producing storms. We did not find evidence of variations in the microphysical parameters describing the storms over shipping lanes and other nearby oceanic regions that might suggest aerosol effects. In contrast, matching lightning stroke data with ship transponder events in oceanic regions where public data are available reveals a strong signal from direct ship interactions with lightning that results in a 1-2 orders of magnitude increase in stroke frequency at current ship locations compared to other nearby regions. These results highlight the central role of direct ship interactions in explaining lightning enhancements over shipping lanes. We also document the frequency of these direct lightning interactions across various categories of vessels and on individual ships present in the public data.

The most powerful optical emissions from lightning have been described as “superbolts” since the 1970s. In 2019, Holzworth et al. (2019) applied the superbolt label to the most energetic Radio Frequency (RF) emissions measured by the World Wide Lightning Location Network (WWLLN). In this study, we compare the WWLLN energies to optical measurements by the photodiode detector (PDD) on the Fast On-orbit Recording of Transient Events (FORTE) satellite and the Geostationary Lightning Mappers (GLMs) on NOAA’s Geostationary Operational Environmental Satellites (GOES) to assess whether WWLLN high energy events coincide with optical superbolts. We find no overlap between traditional superbolts and WWLLN high energy events. Optical superbolts are not energetic to WWLLN, while WWLLN superbolts are not optically bright. Additionally, the top WWLLN sources occur in a different meteorological context than superbolts. Despite some similarities in their overall global patterns of occurrence, WWLLN high energy events correspond to a different phenomenon.

This study uses Fast On-Orbit Detection of Transient Events (FORTE) satellite observations to identify superbolt-class optical lightning events and evaluate their origins. Superbolts have been defined by Turman (1977) as lightning pulses whose peak optical power exceeds 1011 W. However, it has been unclear whether superbolts resulted from particular types of high-energy lightning process or whether they were the result of measurement bias. In the latter case, any decently-bright lightning process could be recorded as a superbolt if the sensor had a particularly clear sight line to the hot channel without thick clouds diluting the optical signals. Our 12-year analysis of FORTE superbolt detections indicates that the lower optical superbolt energy range (~100 GW) is dominated by normal lightning, but brighter cases are predominantly strong +CG strokes that originate from specific types of storms. Oceanic storm systems, particularly during the winter, and especially those located around Japan are shown to produce these intense superbolts. We suggest that some optical superbolts result from favorable viewing conditions and would not be identified as such by another instrument located elsewhere, and that others are associated with a unique set of physics that may merit the “superbolt” distinction.

Optical space-based lightning sensors including NOAA’s Geostationary Lightning Mapper (GLM) detect lightning though its transient illumination of the surrounding clouds. What space-based optical lightning sensors measure is influenced by the physical attributes of the light source, the location of the source within the cloud scene, and the spatial variations in cloud composition. We focus on the lightning channels that serve as optical sources for GLM groups and flashes in this first part of our thundercloud illumination study. We match Lightning Mapping Array (LMA) sources with GLM groups and flashes during two thunderstorms to examine channel segments that are active during optical emission. We find that in each storm, the LMA sources matched with LMA groups are small (median: 2-3 km) compared to GLM pixels (nominal: 8 km), and preferentially come from high altitudes in the cloud (>8-10 km). The detection advantage for high-altitude sources permits GLM to resolve faint optical pulses near the cloud top that might be missed from lower altitudes. However, the most energetic groups can be detected from all altitudes, and the largest groups largely originate at low altitudes. The relationship between group brightness and illuminated area depends on flash development within the cloud medium, and flash development into different cloud regions can be identified by tracking GLM metrics of cloud illumination over time.

Optical instruments such as the Geostationary Lightning Mapper (GLM) detect lightning based on transient changes in cloud illumination. The horizontal location of lightning is determined from the coordinates of the pixels on the imaging array illuminated during the flash. However, the vertical position of the lightning pulses (approximated by GLM “groups”) below the cloud-top cannot be routinely determined from a single space-based instrument. In our prior work, we have developed a machine learning algorithm that can infer optical source altitude for a given pulse based on how the optical energy is distributed across the group footprint. In this fourth part of our thundercloud illumination study, we leverage these source altitudes to generate volumetric GLM imagery of a Colombia thunderstorm. We find that 3D versions of the current GLM meteorological imagery products (that describe thunderstorm kinematics) and thundercloud imagery products (that depict how the flash appears from space) provide additional insights into lightning activity in the thunderstorm are lost in the vertical integration used to generate the current 2D GLM gridded products. This new volumetric imaging capability provides a more comprehensive picture of where lightning occurs in the storm, how its physical characteristics vary across three-dimensional space, and how its optical emissions interact with surrounding the cloud medium.

Lightning is measured from space using optical instruments that detect transient changes in the illumination of the cloud top. How much of the flash (if any) is recorded by the instrument depends on the instrument detection threshold. NOAA’s Geostationary Lightning Mapper (GLM) employs a dynamic threshold that varies across the imaging array and changes over time. This causes flashes in certain regions and at night to be recorded in greater detail than other flashes, and threshold inconsistencies will impose biases on all levels of GLM data products. In this study, we quantify the impact of the varying GLM threshold on event / group detection, flash clustering, and gridded product generation by imposing artificial thresholds on the event data taken from a thunderstorm with a low instrument threshold (~0.7 fJ). We find that even modest increases in threshold severely impact event (60% loss by 2 fJ, 90% loss by 10 fJ) and group (25% loss by 2 fJ, 81% loss by 10 fJ) detection by suppressing faint illumination of the cloud-top from weak sources and scattering. Flash detection is impacted less by threshold increases (4% loss by 2 fJ), but reductions are still significant at higher thresholds (35% loss by 10 fJ, or 44% if single-group flashes are removed). Undetected pulses cause individual flashes to be split and severely impact the construction of gridded products. All these factors complicate the interpretation of GLM data, particularly when trended over time under a changing threshold.

Optical space-based lightning sensors such as the Geostationary Lightning Mapper (GLM) detect and geolocate lightning by recording rapid changes in cloud-top illumination. While lightning locations can be determined to within a pixel on the GLM imaging array, these instruments are not individually able to natively report lightning altitude. It has previously been shown that thunderclouds are illuminated differently based on the altitude of the optical source. In this study, we examine how altitude information can be extracted from the spatial distributions of GLM energy recorded from each optical pulse. We match GLM “groups” with LMA source data that accurately report the 3-D positions of coincident Radio-Frequency (RF) emitters. We then use machine learning methods to predict the mean LMA source altitudes matched to GLM groups using metrics from the optical data that describe the amplitude, breadth, and texture of the group spatial energy distribution. The resulting model can predict the LMA mean source altitude from GLM group data with a median absolute error of < 1.5 km, which is sufficient to determine the location of the charge layer where the optical energy originated. This model is able to capture changes to the source altitude distribution following convective invigoration or maturation, and the GLM predictions can reveal the vertical structure of individual flashes - enabling 3-D flash geolocation with GLM for the first time. Additional work is required to account for differences in thunderstorm charge / precipitation structures and viewing angle across the GLM Field of View.

The optical and VHF instrumentation on the FORTE satellite is used to document the combined phenomenology evolution of a lightning “megaflash” - mesoscale lightning that propagates laterally over exceptional distances. We identify a FORTE flash whose maximum extent was 82 km and inferred length over multiple distinct branches exceeded 100 km. This flash lasted 1.2 s and produced 250 optical and 591 RF events. We find that the channel development mapped by FORTE’s pixelated lightning imager (LLS) occurred at a typical speed of 2.6x10^5 m/s and was accompanied by sustained periods VHF emission that could individually exceed 100 ms in duration. The impulsive IC events generated by the flash indicate that this development occurred at altitudes between 3 and 8 km. Four +CG strokes were identified in the VHF waveform data that are responsible for two of the three highly-radiant LLS groups (the third radiant group came from a possible -CG while 2 of the +CGs were not as optically bright as the others). These strokes occurred at different locations throughout the flash footprint with the most distant strokes separated by approximately 50 km. These space-based observations match previous observations of megaflashes from space as well as ground-based measurements of slow negative leader development during “spider” lightning, suggesting that FORTE is sensing the same phenomena.

While multiple lightning detection systems provide geographical locations of lightning events across the globe, robust lightning altitude measurements on a global scale have proven elusive. Space-based platforms have an advantageous viewing geometry over ground-based systems for making these measurements, but prior studies with the Fast On-orbit Recording of Transient Events (FORTE) satellite were limited to a few thousand events. In this study, we apply the same technique for calculating source altitude from the previous efforts to a large catalog of hundreds of thousands of global FORTE in-cloud lightning events that were coincident with flashes geolocated by its lightning imager between 1997 and 2003. We use this data to document global variations in lightning altitude. As in previous studies, we find that FORTE primarily resolves sources from the upper (positive) charge layer at ~11 km altitude in normal thunderstorms. However, sources are also recorded from other charge layers in the storm, and from leaders developing between layers. In particular, we note a pronounced increase in source altitude in the first 20 ms of FORTE flashes from the negative leader developing upward into the upper positive charge layer. Regions known for wintertime and/or stratiform lightning have increased contributions from low-altitude sources, while tropical regions particularly around Panama and the Maritime Continent have the greatest concentrations of high-altitude sources.

Previous lightning climatologies derived from Lightning Imaging Sensor (LIS) and Optical Transient Detector (OTD) total lightning measurements have quantified lightning frequency as a Flash Rate Density (FRD). This approach assumes that lightning flashes can be represented as points, and quantifies the frequency of lightning centered in each grid cell. However, lightning has a finite extent that can reach hundreds of kilometers. A new climatology based on Flash Extent Density (FED) is constructed for LIS (including ISS-LIS) and OTD that accounts for the horizontal dimension of lightning. The FED climatology documents the frequency that an observer can expect lightning to be visible overhead - regardless of where the flash began or ended. This new FED climatology confirms and elaborates on the previous global LIS / OTD FRD and Americas-only Geostationary Lightning Mapper (GLM) findings. Applying GLM reprocessing codes to LIS and OTD data reveals cases of megaflashes measured from Low Earth Orbit that were artificially split by the LIS / OTD clustering algorithms. The FED climatology maintains Lake Maracaibo as the global lightning hotspot with an average of 389 flashes / day, but designates Karabre in the Democratic Republic of the Congo as the global thunderstorm duty (percent of the total viewtime where lightning is observed) hotspot at 7.29%. Meanwhile, Kuala Lumpur is the national capital city with the most lightning, and its airport (KUL) is the top major airport affected by lightning. The FED seasonal cycle and month-to-month changes in the “center of lightning” for the three continental chimney regions are also discussed.