During the early morning of 10 August 2017, a large Mesoscale Convective System (MCS) formed over northwest Mexico. Vaisala’s Global Lightning Dataset GLD360 network detected a total of 290,769 strokes during the period from 0000 to 1200 UTC on 10 August, as shown in Figure 1. The GLD360 data shown are for both cloud-to-ground strokes and in-cloud pulses. Several centers of activity are apparent within this MCS as indicated by lightning.
Figure 1. Lightning events detected by the Global Lightning Dataset GLD360 network from 0000 to 1200 UTC on 10 August 2017.
Satellite imagery in the middle of this period at 0600 UTC are shown in Figure 2, obtained from the Colorado State University RAMSDIS website. It is apparent that the MCS was the largest cloud system in North America at this time. The detail of the spatial arrangement of the convection within the MCS that is provided by the lightning data is not detectable from the satellite image by itself. Note Tropical Storm Franklin in the lower right corner of the image in Figure 2. A loop of this MCS from 0000 to 1200 UTC is also shown here. MCSs also occur on the Great Plains of the United States in the warmer months of the year, and are prolific lightning and rainfall producers. This event over northwest Mexico is similar in intensity.
Figure 2. Color representation of clouds in the southwestern United States and northwest Mexico at 0640 UTC on 10 August 2017.
Since 2007, lightning has killed an average of 30 people per year in the United States, and potentially up to 24,000 people per year around the globe. While the global figures are significant, fatalities have declined in the U.S. in the last 40 years as shown below.
Yearly United States reported lightning fatalities. Data available here>>
As peak lightning season approaches, news stories about lightning will become more common. To improve the understanding of these stories and dispel misconceptions, we will briefly explain the science behind lightning. Lightning science is a complex field, and not all of the answers about lightning, like other fields of meteorology, are known. Here’s what we do know:
Lightning is a very large electric discharge. Categorized into cloud discharges and ground discharges, lightning occurs in part because of a separation of electric charge within a thunderstorm cloud. The electric current measured within lightning can be as high as 700-thousand amps.
Lightning is measured in strokes and flashes, not bolts and strikes. In The Lightning Discharge (Uman, 1987), Martin Uman writes, “The word thunderbolt is commonly used in the nonscientific literature to refer to cloud-to-ground lightning.” Photographic and electromagnetic field evidence has shown that what many people consider to be a “lightning strike” is actually a flash of lightning that is composed of up to 15 individual lightning strokes.
Average flash density per year for the Continental United States for the years 2006-2016.
Average stroke density per year for the Continental United States for the years 2006-2016.
Lightning can be detected, but not predicted. Through magnetic field direction finding, time of arrival of electromagnetic waves, or combinations thereof, lightning can be detected. As mentioned earlier, because lightning occurs in part due to electric charge separation in the thunderstorm cloud, sensors that monitor the atmospheric electric field (electric field mills or electrostatic sensors), can provide advanced alert to the potential of lightning. Murphy, et al. (2008) describes the manner in which electric field mills work, and shows high false alarm rates for lightning warnings based on electric field mills alone. The Occupational Safety and Health Administration warns that “no systems can detect the “first strike”, detect all lightning, or predict lightning strikes.”
Lightning is a beautiful, but dangerous, weather phenomenon. Vaisala continues to research and develop lightning detection technologies and customer services that are effective for protecting people and property. In addition to our research, we provide educational resources that allow you to better understand lightning, and the news stories it generates. We believe in a world where observations improve daily life. As the peak of lightning season approaches, remember to be lightning aware, and “When thunder roars, go indoors!”
Supertyphoon Nepartak traveled for several days in the Northwest Pacific until it struck Taiwan on July 8, 2016. Wind gusts over the open ocean on July 8th were forecast to exceed 200 mph, although no measurements could be made at the surface at that time. Along its track in Figure 1, Vaisala’s Global Lightning Dataset GLD360 network detected a unique signature consisting of lightning in the eyewall that is located directly along and near the track of the center. While this signature has been observed in a few storms for up to 24 hours since GLD360 data became available in 2011, none of the storms had this feature that lasted nearly as long as the 72-hour period in Nepartak (Figure 1). It shows lightning enveloped along the track, a phenomenon which is being observed only to occur in very strong tropical cyclones that lack outer rainbands. Further studies are planned of the few other storms since 2011 that had a least a few hours of this signature. The ability of GLD360 to track lightning continuously far from land makes this type of observation possible. Satellite imagery in Figure 2 shows an intense tropical cyclone with a sharply defined eye within a uniform circular annulus, but no outer rainbands.
FIGURE 1. Map of track and GLD360 strokes for supertyphoon Nepartak in North Pacific Ocean from 03 through 08 July 2016. Strokes are color coded by day, and daily totals are in parentheses. Taiwan is in the upper left portion of the map, and the Philippines are to the southwest.
FIGURE 2. Satellite image of supertyphoon Nepartak at 1200 UTC on 06 July 2016.
Our friends at KGUN9 in Arizona interviewed Vaisala lightning expert, Ron Holle, recently to discuss how to stay safe when lightning is near, and some myths about protecting yourself. Holle reminds us that nowhere is safe outdoors when lightning is near, and that lightning can strike from storms that are then or fifteen miles away. View the entire story at KGUN9's web site:
When you hear thunder, head for your vehicle or the nearest building.
March is the beginning of meteorological spring in the Northern Hemisphere. Frequent spring storms in the United States combined with improvements to Vaisala’s NLDN and GLD360 networks resulted in very large counts of lightning during March 2016. This was true in both the United States and around the globe.
Storms occur in the southeast half of the United States every spring, but this year the March lightning activity was more similar to mid to late April (comparing to past years). The NLDN underwent a substantial upgrade in cloud pulse detection through the summer of 2013, and more than twice as many cloud pulses are now detected since the upgrade than before the summer of 2013.
A particularly intense period of lightning occurrence began on 8 March and lasted several days. A nearly stationary frontal boundary from Texas to the northeast resulted in strong low-level flow from the south bringing copious amounts of moisture northward. There were hundreds of thousands of strokes and rain up to 15 inches in Louisiana and Arkansas during this period. Several additional but less intense multi-day scenarios were established later in the month across the southeast United States that produced very large lightning frequencies.
Globally, the GLD30 network doubled the number of detections beginning on 18 August 2015. This major change in network performance took place somewhat after the summer lightning maximum in the Northern Hemisphere activity. The increased detection is becoming especially apparent as Northern Hemisphere lightning increases again, and such record counts compared with the same months last year can be expected to continue through 2016.
Since both the NLDN and GLD360 network have been substantially upgraded, the global and United States lightning frequencies reached levels for the month of March not seen in the past as spring storms erupted over the southeast United States.
Hurricane Patricia began as a tropical cyclone, forming in the eastern
Pacific. It rapidly intensified into a Category 5 hurricane, making landfall on Oct. 23 at 6:15 p.m. CDT near Cuixmala in Jalisco state of southwest Mexico. As the storm approached Mexico it was interesting to watch the continuous lightning pattern.
As you can see from the images, lightning was present in a tight circle in the eyewall. This is somewhat uncommon because about 90% of named tropical systems do not have lightning in the eyewall region. Lightning is typially in the outer bands.
The important lesson about this type of continuous lightning pattern is that when it does appear, the storms are almost always very strong. This was the case for Hurricane Patricia, with some of the highest winds recorded and lowest pressure readings. On Oct. 23, wind speeds reached an intensity estimate of 184 mph (160 kt), which is tied with eastern north Pacific Hurricane Linda of 1997 for the strongest on record. A Vaisala dropsonde released into the eye measured a sea-level pressure of 892 mb (adjusted for surface winds), which breaks the record for the lowest pressure of an east Pacific hurricane.More about the images:
The first Patricia map shows activity during the hours of the hurricane hunter reconnaissance on October 22, 2015. The eyewall lightning consolidated around the track during the hours of that recon flight.
The second map shows 17 hours on October 23, 2015, where the tight circle of lightning is continuing. Impressive
For the first few months of 2015, the cumulative total of lightning fatalities was somewhat above normal compared with the previous ten years (Fig. 1). After a period of several days this month, the cumulative total has now returned to normal compared with recent years. This graph was developed by Mr. William Roeder and distributed to members of the National Weather Service Lightning Safety Awareness team. This team provides the lightning safety information found at www.lightningsafety.noaa.gov.
FIGURE 1. Cumulative number of U.S. lightning fatalities by day of the year for 2015 compared with the most recent ten years.
early start to the lightning fatality totals for the year kept the 2015
curve in Fig. 1 above normal until yesterday. As a result, several
ideas were proposed regarding the cause. Among them were a very much
higher number of 2015 cloud-to-ground (CG) flashes and/or more positive
In order to examine this possibility, data from
Vaisala’s National Lightning Detection Network (NLDN) were examined.
Figure 2 shows the results for all CG flashes over the continental
United States land area, and Figure 3 shows positive counts. The 2015 CG
flash counts are currently 13% ahead of the ten–year average, while the
positive CG flash count for 2015 is 1% larger than the average of
recent years, as of early July.
Note that both 2015 NLDN curves
are below normal earlier in the year when the fatality curve in Fig.1
was above normal. It can be concluded that no apparent relationship
exists between the number of fatalities and number of CG flashes, or
FIGURE 2. Cumulative number of U.S. cloud-to-ground lightning flashes by day of the year for 2015 compared with the most recent ten years. Data from National Lightning Detection Network (NLDN).
FIGURE 3. Same as Figure 2 for positive cloud-to-ground flashes.
Around 21:00 UTC on 22 April 2015, the Chilean volcano, Calbuco, began its first eruption in over 40 years, sending up a massive ash cloud that generated spectacular displays of lightning. Vaisala’s GLD360, the world’s most accurate and truly global lightning detection system, located a large number of discharges during the event. Starting at 22:28 UTC on 22 April, shortly after the initial eruption was reported, and extending over a period of about 10 hours, GLD360 recorded a significant number of flashes within 50 km of the volcano. Most of the discharge positions were clustered close to the volcano, but some were spread out to the northeast, in exact agreement with the propagation of the ash cloud as seen by satellite images recorded throughout the night and with the 12:00 UTC (22 April) sounding taken at nearby Puerto Montt. The map shows that while there was a lot of lightning activity across the north of the continent and into the Atlantic ocean, there was little to no lightning elsewhere within 2,500 km of the volcano.
Vaisala Radiosonde Verifies Observations
At 12:00 UTC (22 April), the Dirección Meteorológica de Chile released a weather balloon with a Vaisala Radiosonde from a nearby sounding station at Puerto Montt. The radiosonde data confirmed southerly to southwesterly winds near and above the tropopause, where the highest part of the ash cloud was reported by the Volcanic Ash Advisory Center, Buenos Aires, as the plume crossed the Chilean-Argentine border.
Vaisala Radiosondes, combined with Vaisala DigiCORA Sounding System MW41 are used to conduct reliable and accurate upper-air weather observations.
According to Martin Murphy, Senior Scientist at Vaisala, “Information about the polarity and estimated peak currents of lightning generated by volcanic ash clouds could provide additional insights into the mechanism that produces these discharges. Detailed studies of volcanic lightning have revealed two phases of activity, one of which produces large, discrete discharges such as those shown here. These discharges emit radio-frequency energy that can be observed remotely by the Vaisala GLD360.
Typhoon Maysak developed and grew last week in the southwestern Pacific Ocean. It became strong enough to be classified as a super typhoon; gusts were forecast to reach 190 mph/352 km per hour. This is one of the earliest storms of this intensity in the North Pacific Basin. It caused five fatalities and major devastation on the islands of Chuuk and Yap. It is forecast to weaken significantly before reaching the Philippines.
Figure 1 shows the location of the eye during its strong tropical phase from formation to rapid intensification from 27 March through 01 April. Due to the uniformly high detection of GLD360 over these regions, lightning can be plotted over this entire oceanic region. The 29,575 lightning events detected in this region by GLD360 are shown during these six days with a 250-km buffer around the track.
Figure 1. Map of track and GLD360 strokes for supertyphoon Maysak in North Pacific Ocean from 27 March through 01 April 2015. Strokes are color coded by day, and daily totals are in parentheses. The islands of Yap and Chuuk are indicated.
Three features of well-organized tropical cyclones have been observed and documented in publications by Demetriades and Holle of Vaisala, and others. First, the eye in the center has a small amount of intermittent lightning. Second, there is an annulus of minimal lightning outward from the eye. Third, there is usually an outer rainband area farther out, often to the south through east in the Northern Hemisphere. These rainbands are areas of heavy convective rainfall that are often a significant threat for widespread flooding.
During the first few days, Figure 1 shows widespread lightning activity to the south and east of the storm center during its development phase. Once it was organized, however, there is a near complete absence of lightning except in occasional bursts in the eyewall next to the track, and in a few large outer bands.
Figure 2 shows a satellite view early on 01 April when Maysak was at its peak in terms of organization. Note the clearly-defined eye surrounded by very smooth clouds. The storm is not especially large in area but tightly organized. At this time, that uniform circular annulus indicates the region which is often an area without much lightning, as indicated well by GLD360. There are no large outer rain bands in the satellite image, indicating that no widespread heavy rain was likely outside of the main circulation. Such information is very helpful in diagnosing the potential for heavy rains over a large area away from the center, which is not the case in Maysak.
Figure 2. Satellite image of supertyphoon Maysak at 0423 UTC on 01 April 2015.
The month of March 2015 set
a U.S. record for the longest period to start March without any tornado reports,
through the 24th of the month. However, that streak came to an
abrupt end when a major severe thunderstorm event began on the 25th.
Figure 1 shows the severe
weather reports to be concentrated in Oklahoma and surrounding states.
Unfortunately, one person was killed by a tornado that struck a mobile home
park in NE Oklahoma. This storm began in the Oklahoma area and extended to the
ENE and WSW along a cold front. Strong straight-line winds and large hail were
more common than tornadoes in this event.
Figure 1. Map of 177 severe storm reports collected by the Storm Prediction Center during the 24 hours ending at 1159 UTC on 26 March 2015.
Lightning Detection Network (NLDN) provides Advanced Total Lightning data, and shows
very extensive lightning activity for a 24-hour period during this storm.
Figure 2 shows more than one-third million detected cloud-to-ground (CG) strokes
and cloud pulses. This high-resolution map shows long streaks that represent
individual thunderstorms better than any single radar or satellite depiction
can indicate. Storms begin east of a north-south line though western Oklahoma,
due to the location of the dry line where storm initiation is common. Roughly
70% of the detected lightning events were cloud lightning pulses. The NLDN
underwent a major upgrade in summer 2013 that resulted in excellent cloud pulse
detection, as is shown in this example. The time series of CG and cloud events
is shown in Figure 3 for these 24 hours. A more rapid ramp-up in the number of
lightning events occurred at the start of the storm, than the slower decline.
Typically, thunderstorms have quicker increases in lightning toward the
beginning than at the end.
Although severe weather in the event was not as
extreme as it tends to be later in the spring, the lightning threat itself is significant.
Such a large number of lightning events caused power outages at homes and
manufacturing sites, set fire to several structures, affected emergency
communications, and downed power lines that contributed to widespread disruption
to more people than those affected by the reported severe weather events.
Figure 2. Map of more than one-third million lightning events detected by the NLDN from 1200 UTC on 25 March to 1200 UTC on 26 March 2015.
Figure 3. Hourly time series of more than one-third million events detected by the NLDN in the area of Figure 2 from 1200 UTC on 25 March to 1200 UTC on 26 March 2015.
(Hidden) Blog Tools 
About News and Insights
Use this page to stay updated on Vaisala's latest news, research, and opinions on current energy industry trends and challenges. Get guidance and insights directly from our experts in weather measurement, energy assessment, power forecasting, and asset management.