Upward lightning from tall objects is a unique type of ground flash because it originates from a tall object attached to the Earth’s surface and has as a unidirectional leader propagate upward toward the overlying storm. Typically the upward propagating leader is positive, but upward negative leaders have also been observed. A thunderstorm charge region overlaying the tall object and induced opposite polarity charge in the object is a necessary requirement for this type of lightning. However, it is the tall and narrow shape of the object that enhances the electric field locally at its tip that creates conditions favorable for the initiation and development of an upward propagating lightning leader. Without the presence of the tall object, an upward lightning flash would not occur. The high-speed video below shows an upward lightning flash from a television broadcast tower in Rapid City, South Dakota.
Upward flashes can be primarily identified in still images by the upward branching if branching occurs. Frequently, upward leaders will propagate to the cloud base and then turn horizontally and travel along the cloud base. Sometimes, the upward leader does not branch until reaching cloud base. This behavior is likely due to a negative screening charge layer along the cloud base which acts as a potential well for the positive leader, causing it to propagate horizontally through the well. The digital still image below shows this behavior.
Our research has shown that upward flashes can either be triggered by preceding nearby lightning flashes or initiate without any preceding triggering activity. Upward flashes during the summer months are most frequently Lightning-Triggered Upward Lightning (LTUL) and triggered during certain types of storms and specific types of preceding flashes. Horizontally extensive storms such as Mesoscale Convective Systems (MCSs) with large stratiform precipitation areas are the most favorable storm type for producing upward lightning. The horizontally extensive positive ground flash is the most common triggering flash type for LTUL. When horizontally extensive positive charge is over tall objects, the passage of negative leaders through the positive charge can result in an electric field change at the object tip which initiates an upward positive leader (UPL). The triggering negative leader activity must pass close enough and/or be strong enough to cause a fast electric field change that exceeds the threshold for upward leader initiation. The triggering negative leader activity can develop during a cloud flash, or prior to or after the return stroke of a positive ground flash.
Additionally, the return stroke of a +CG flash which traverses a leader network that has formed near the objects can cause an extremely fast electric field change that initiates upward leaders immediately following the return stroke (within 2 milliseconds based on our observations). The following animations and high-speed videos show the different processes for triggering of upward positive leaders from tall objects.
Negative leaders pass near or over the tall objects during a cloud flash.
The negative end of the bidirectional leader that forms during a cloud flash can pass near or over the tall objects and initiate upward positive leaders.
Negative leaders pass near or over the tall objects prior to a +CG return stroke.
The negative end of the bidirectional leader network that develops prior to a +CG return stroke can pass near or over the tall objects and initiate an UPL before the return stroke. When the return stroke occurs, the rapid electric field change causes intensification in the already growing UPLs.
Negative leaders pass near or over the tall objects following a +CG return stroke.
For this case, the return stroke of a positive ground flash does not initiate an UPL, however, the negative leader development, that occurs well after the return stroke, passes near or over the tall objects and initiates an UPL. In this case, the initiation of the UPLs can take place 10s to 100s of milliseconds after the return stroke.
Positive return stroke traverses previously formed leader network that is near or over tall objects
During the initial stage of a positive ground flash, the negative end of the bidirectional leader network passes near or over the tall objects, but does not initiate UPLs. A subsequent +CG return stroke traverses the previously formed leader network and initiates UPLs due to the rapid and intense electric field change associated with the return stroke. UPLs typically initiate within 2 ms of the return stroke.
When conditions are favorable, usually when a large MCS and well defined stratiform precipitation area moves slowly over a location with tall buildings or towers, a large number of LTUL flashes can occur in a relatively short period of time. These “perfect storm” scenarios have been documented in Toronto, Canada and Oklahoma City, Oklahoma. On 24 August 2011, 52 upward flashes from the CN Tower in Toronto occurred the span of 84 min, and on the night of 21 October 2017, I personally recorded 31 of 39 upward flashes from a group of tall towers in north Oklahoma City. The 39 upward flashes took place within a 112 min span.
Upward Negative Leaders
So for we have discussed the development of upward propagating positive leaders from tall objects since this polarity is dominantly observed. However, upward negative leaders (UNLs) can also be triggered by preceding lightning activity. We have observed only one UNL case in Rapid City over the last 14 years. For this one case, our analysis shows that a negative cloud-to-ground (-CG) return stroke passing near a tower caused the initiation of the UNL. The high-speed video from this case is shown below.
Other researchers have observed UNL cases, and it appears that the processes leading to UNL initiation are similar to that for UPL initiation, however, the polarities of the triggering leaders and cloud charge regions involved are opposite. The fact that UNLs develop much less frequently is likely due to the following reasons.
1) The charge arrangement and dominate flash type (positive ground flash) for MCSs favors UPLs.
2) Electric field change threshold for initiation of UPL is significantly less than that required for UNL development.
Therefore, since favorable conditions for UNL initiation occur less often, and when they do, appear to require a higher electric field change threshold for initiation, they are rare compared to UPLs. The analyzed data for the one UNL case showed an extremely large electric field change compared to the electric field change magnitudes associated with UPL cases.
Self-Initiated Upward Lightning (SIUL)
Upward lightning can initiate without any preceding nearby flash activity. This most often occurs during winter months when robust snow events such as blizzards or heavy snow envelop the tall objects. Convection within these stronger winter systems causes electrification to take place, but unlike the warmer summer conditions, the cloud base tends to be much lower bringing the tall objects into the charge regions or exposing the objects to charged snowfall. Because SIUL usually occurs with the tall object shrouded in clouds with heavy frozen precipitation, visually observing this type of upward flash is difficult. The same towers that experience LTUL in summer in Rapid City, South Dakota have experienced SIUL flashes during two major snow events. Most notably, the blizzard of 4 Oct, 2013 created favorable conditions for 25 SIUL flashes from the towers. Although the towers were obscured for all of the upward flashes, a 2-dimensional interferometer system recorded five of these flashes. Below is a figure showing the upward development of the flash from that tower as the radiated energy from the leader tips was triangulated by the interferometer. In all cases, there was no preceding nearby flash activity prior to the upward leader initiating.
Below is a video animation of the data recorded by the digital interferometer. You can visualize that you are standing facing the towers. Each of the individual data points represents the azimuth and elevation to electromagnetic radiation generated by the lightning leader (and received by the sensor) as the leader propagated. The system records data in sequential 4 microsecond windows and determines the direction to the strongest signal in each time window. Since lightning tends to branch as it grows, you see the source points plot the spreading branched leaders as they grow. The leader clearly initiates from a single point and then spreads upward as it branches. Occasionally, you can see a rapid succession of source points that travel back along a branch toward the tower. These are recoil leaders which form on decayed branches in an attempt to reionize the branch.
Our research has shown that SIUL flashes during heavy snow events can be long duration events (up to 2 seconds) with the leaders traveling large horizontal distances. For one case on 4 Oct 2013, a horizontally propagating upward positive leader branch produced multiple recoil leaders after decaying. Early recoil leader activity had the negative end traveling back to the tower. However, in one case, the branch path decayed to the point that the negative end of the recoil leader deviated from the path, traveled toward ground as a stepped negative leader, and initiated a -CG return stroke away from the tower. Technically, this could be classified as a ground-to-ground flash given that the leader initiated from a grounded object and resulted in a subsequent, spatially separate return stroke. Below is the interferometer record from this event.
Our analysis of the lightning location data associated with a major blizzard that affected nearly half the country on 2 Feb 2011 suggested that a large majority (70-90%) of the lightning that occurred in the snow sector was self-initiated upward lightning from tall objects. The object types identified from colocation with lightning location data where tall broadcast towers, cell towers, wind turbines, tall buildings, power station smoke stacks and powerline towers. One of the fundamental questions that these findings raise is how much lightning is being produced by structures built for human activities. This lightning would not occur in the absence of these structures, so is anthropogenic activity creating more lightning, and if so, are there any negative implications?
Next section: Upper Atmosphere Lightning
Previous section: Ground Flashes