Colorado State University lightning fatalities of 24 July 2008:
 2 graduate students fatally wounded on the CSU Fort Collins campus

Stephen Hodanish
National Weather Service
Pueblo, Colorado

This research is part of the Lightning Casualty Case Study section of the Colorado Lightning Resource Page, NWS Pueblo, Colorado. 

"The primary reason for these lightning casualty case studies is to observe where victims were located relative to thunderstorm activity when they were struck".

UPDATE: 01 Oct 2008; Corrected the ratio of In-Cloud flashes vs Cloud to Ground flashes at the end of the document. The document originally stated a ratio of 10:1...it is more like 5:1.

          Introduction

During the early evening of 24 July 2008, two graduate students, William J. Szlemko, 35, and Marc V. Richard, 33, were struck by lightning on the Colorado State University (CSU) campus in Fort Collins, Colorado. According to CSU news media, they were struck in the “Sherwood Forest” area of the campus which is a heavily treed area just south of the Warner College of Natural Resources Building (Fig. 1 and Appendix). Mr. Szlemko was pronounced dead shortly after the incident. Mr. Richard survived for 2 days before succumbing to his injuries. 

 

 

Figure 1. Satellite image showing the location where the 2 men were struck (marked by yellow pin in center of image). Click on image for a larger view. 

          Data analysis

In order to observe which lightning flash caused the fatality, two pieces of information need to be known. The first piece of information is knowing the exact time when the lightning flash hit the victim, and the second is the location of where the victim was struck. Typically, the victims' location is well documented. According to CSU police, the incident occurred just south of the Forestry Building on the main CSU campus. The GPS coordinates for this location were 40.573867N, 105.081617W.

The exact time of when the flash occurred is not well known in this case. The reason for this is no one actually observed the two men being struck. Although it cannot be confirmed, it is believed that they were not found until ~10 minutes after they were hit by the flash. The first 911 call to the CSU authorities was made at 7:27 pm MDT (0127 UTC 25 July 2008).

In order to find out which flash struck the two men, Cloud to Ground (CG) lightning data from the North American Lightning Detection Network (NALDN) was analyzed during a one hour time period prior to the 911 call being made to the CSU authorities. All lightning flash data which occurred within 10 miles of the fatality location was analyzed. Figure 2 shows this data.  

 

Figure 2. Lightning flash data which occurred over a 1 hour time period within a 10 mile radius of the location of where the 2 men were struck. The “0.00” represents the location of where the fatalities occurred. Data was analyzed between 0027 UTC (627 pm MDT) and 0127 UTC (727 pm MDT). The 911 call was received to the CSU Police at 727 pm MDT.  Click on image for a larger view.  

An examination of figure 2 shows one flash in particular which occurred very close to where the men were struck. This flash occurred at 7:18:15 pm MDT and was located 0.17 miles (~900 feet) away from the fatality location (the median accuracy of the NALDN is about 1600  feet). This flash occurred about 9 minutes prior to the 911 call being received by the CSU authorities. Two other flashes were also noted to occur relatively close to the fatality location, one at 7:20:24 and the other at 7:20:29 MDT. These 2 flashes were located 0.62 and 0.47 miles away from the fatality location respectively. Figure 3 shows the location of the 3 closest flashes relative to where the men were struck. 

 

Figure 3. Location of the 3 closest flashes as detected by the NALDN which occurred during the 1 hour time period shown in Fig. 2. The flash that occurred at 7:18:15 MDT was the flash that likely caused the 2 fatalities. The location where the 2 men were struck is also shown.  

Although any of these 3 flashes may have caused the fatalities, it will be assumed that the flash which occurred at 7:18:15 pm MDT was the flash that caused the fatalities. This assumption is based on the fact that the closest flash likely caused the fatalities.

Figure 2 showed all of the CG lightning which occurred within a 10 mile radius up to 1 hour prior to the 911 call. Based on the information which was shown in this figure, we now (likely) know which flash caused the fatalities. We will now examine this data to see how much lightning was occurring closer in (both spatially; 6 miles, and temporally; 10 minutes) relative to the time of when the fatalities occurred. Figure 4 shows this data.

 
 

Figure 4. Lightning flash data which occurred during a 10 minute time period within 6 mile radius of the location of where the 2 men were struck. The “0.00” represents the location of where the fatalities occurred. Data was analyzed between 0108:15 UTC (7:08:15 pm MDT) and 0118:15 UTC (7:18:15 pm MDT). A total of 79 cloud to ground lightning flashes occurred during this time period. Click on image for a larger view.

From figure 4, we can see this storm was producing a significant amount of cloud to ground lightning activity within a 6 mile radius up to the time of the fatal flash. A total of 79 CG lightning flashes occurred during this 10 minute time period. Twelve of these flashes occurred within 2 miles of where the two men were struck. An examination of the data during the last one minute time period shows 14 CG lightning flashes occurred; four of these flashes occurred within 2 miles.

Figures 5 and 6 show combined radar and lightning data. Figure 5 shows the lightning data which occurred during a 1 minute time period between 0118:00 UTC and 0119:59 UTC. This figure shows the actual flash which caused the casualty. Note that this image shows heavy rain occurring at the location of the flash (>50 dBz). 

 

 

 Figure 5. Radar image and lightning flash data over the greater Fort Collins area.  The lightning data in the figure above shows all of the lightning flashes (indicated by the white “-“) which occurred during a 1 minute time period between 0118:00 and 0119:59 UTC. The flash which struck the 2 men is circled.  The distance scale is located in the lower left of the image. Click on the figure for a high resolution image.

A time lapse of the lightning and radar data is shown in Figure 6. The duration of this time lapse is 35 minutes. The time lapse begins at 0044 UTC (644 pm MDT) and ends immediately after the fatal flash (0019 UTC/719 pm MDT). The thunderstorm which produced this flash was moving to the east-southeast (119 degrees) at 23 mph.
 

 

Figure 6. Loop of Radar and Lightning data over the Fort Collins area. Loop of "1 minute lightning plot" runs from 0043 to 0119 UTC (35 minutes). Radar data is composite reflectivity data from National Weather Service Doppler weather radar KCYS (Cheyenne, WY). Note that in the above loop, the "1 minute" lightning data shows all of the cloud to ground lightning that occurred during that one minute time period; i.e., "1 minute lightning plot Fri 00:44...." indicates all of the lighting which occurred between 0043:00 and 0043:59). In the above loop, the 1 minute lighting plots change every minute while the radar data changes every 4-5 minutes. The reason for this is it takes 4-5 minutes to generate a single composite radar reflectivity image. The "time date stamp" of the radar data denotes the time the radar image began. Click on the figure for a high resolution loop. 

          Discussion

During the early evening of 24 July 2008, two graduate students were struck by lightning on the campus of Colorado State University (CSU) in Fort Collins, Colorado. One student passed away shortly after the flash while the other passed away 2 days later. According to CSU news media, they were struck in the “Sherwood Forest” area of the campus which is a heavily treed area just south of the Warner College of Natural Resources Building (see fig 1).

In this case study, we know that no one actually eye witnessed the men being struck. From the radar time lapse (see fig 6), heavy rain was occurring prior to the flash and this likely explains why nobody else was outside to witness the event. It is also possible that visibility was restricted due to the heavy rain. In addition, the location of where the men were struck was heavily treed (see appendix). As was mentioned earlier in this report, it is believed approximately 10 minutes had gone by before the first 911 report was received. Data shown in this report supports this conclusion; i.e,  as per Figure 2, the flash which likely caused the fatalities occurred at 0118:15 UTC while the first 911 report to the CSU authorities arrived at 0127:30 UTC (a time difference of 9 minutes and 15 seconds). 

From a lightning safety perspective, what could have been done to prevent this tragic event? The radar time lapse (Figure 6) clearly showed a thunderstorm had developed well to the west-northwest of the Fort Collins area and steadily moved towards the campus.  Rain from this storm likely began to fall on the main campus around 650 pm MDT, with the heavier rain arriving around 7 pm. This heavy rain continued up to the time of the fatal lightning strike.

In addition to the heavy rain activity, the lightning plots indicated there was a considerable amount of cloud to ground lightning activity prior to the men being struck. As was shown in figure 4, a total of 79 CG lightning flashes occurred 10 minute prior to the incident. Twelve of these flashes occurred within 2 miles of where the two men were struck. An examination of the data during the last one minute time period showed 14 CG lightning flashes occurred; four of these flashes occurred within 2 miles. It should also be noted that the lightning data shown in this report was only cloud to ground lightning data; no in-cloud or cloud to cloud data was measured. Typically, for every 1 CG flash there is 5 in-cloud lightning flashes. All lightning, be it cloud to cloud or cloud to ground, produces thunder.

Based on this lightning data, the 2 students very likely knew lightning was in the immediate area. It is possible they may have not actually seen the nearby CG flashes due to the trees and buildings in the immediate vicinity, but it is likely they heard the continuous rumbles of thunder. 

It should also be noted, however, that the intervening buildings, the heavily treed area, and the heavy rain may have muffled the sound of thunder. In addition, construction was occurring nearby and compressors running construction equipment may have been running during the time leading up to the event. This construction activity (if it were occurring at the time of the flash) may have also muffled the sound of the thunder.
 
Information acquired from CSU Student Affairs indicated that the two men enjoyed martial arts, and met in this area to practice their sport (martial arts equipment was found on the scene). Although it is not known for sure, it is believed they were practicing their sport at the time of this tragic event.  

So why did the 2 men not seek safe shelter? Although we will never know for sure, information gleaned from people who knew the first victim (Mr. Szlemko) indicated that he enjoyed the outdoors very much and spent a lot of time there. For example, as a young boy, Mr. Szlemko worked on a cattle ranch and would drive cattle in open country in northern Wyoming. During these cattle drives, Mr. Szlemko would be "out in the open" for long periods of time, and there would be no place to seek shelter when thunderstorms developed over the wide open prairie. It is likely that Mr Szlemko was accustomed to being outdoors when thunder and lightning was occurring, and therefore, he may not have felt that he was in iminent danger.

This meteorological lightning casualty forensic case is quite unique when compared to other cases that the author has completed in the past decade. In nearly all of the other lightning strike forensic cases, the lightning activity which produced the casualty(ies) was infrequent, that is, the storm was only producing one flash every 2 to 3 minutes. In this case, the thunderstorm was producing over 8 flashes per minute! This is a significant amount of cloud to ground lighting activity. (Click here to read the other individual case studiesclick here to read a summary of many cases).

It is well documented that you should not be under a tree when cloud to ground lightning activity is occurring. A large majority of lightning deaths which occur outside occur either in open fields or near/under trees. For that matter, you should not be outside at all when lightning is in the area. The best thing these two men should have done was to wait inside one of the buildings that were immediately nearby, or to seek shelter in an automobile.

Remember, NO PLACE OUTSIDE IS SAFE WHEN LIGHTNING AND THUNDER IS OCCURRING IN THE AREA. IF YOU CAN SEE LIGHTNING OR HEAR THUNDER, GET INTO A WELL BUILT BUILDING OR AN AUTOMOBILE IMMEDIATELY.

More information about lightning safety can be found here:

Additional Outdoor Lightning risk reduction recommendations can be found either here... here ... here or here.

         Acknowledgements

The author would like to thank Scott Anthony (CSU Police Department) and CSU Student Affairs for information regarding this case. We also thank Rich Blakeslee and Sherry Harrison (NASA) for additional lightning information. We appreciate information from Ron Holle and Bill Roeder regarding correcting an error about the ratio of In-Cloud lightning vs CG lightning. The author also thanks John Weaver (CIRA/CSU) who added invaluable insight into this case as he knew Mr. Szlemko personally. Mr Weaver also shot the photos which were shown in the appendix.

 file=18359


USA.gov is the U.S. government's official web portal to all federal, state and local government web resources and services.