Crestone, Colorado Lightning Fatality of 27 July 2003
National Weather Service
This research is part of the Colorado Lightning Resource Page at NWS Pueblo, Colorado
"The primary reason for these lightning casualty case studies is
to observe where victims are
This document contains large image files. A high speed internet connection (or a lot of patience) is recommended
At approximately 4:37 pm mdt (2237 UTC) on 27 July 2003, a 25 year old female was struck and killed by a lightning flash while hiking along the Willow Creek Trail just east of Crestone, Colorado. The female, along with her 33 year old husband, were coming down the trail after hiking Kit Carson and Challenger Peaks, and were approximately 2.5 miles from the trail head near Crestone when the flash occurred (Figures 1 and 2). According to rescue officials, the couple were located at mid slope in heavy timber when the flash occurred. Although the male was thrown to the ground by the flash, he received no injuries related to the flash.
Figure 1. Map showing hiking trails in the Crestone,
Colorado region. The red "X" marks the location of the lightning strike
victim. Trail #865 is the Willow Creek Trail. The distance between Crestone
and the location of the lightning incident is approximately 2.5 miles.
Map from Delorme. Used with permission.
Figure 2. General topography of the southern Colorado region. The victim was struck just east of Crestone, Colorado (center of image). The gray line running along the Sangre De Cristo Mountains (mountains immediately east of Crestone) is the county line seperating Saguache and Custer counties. The Wet Mountain Valley and San Luis Valley are labeled for reference. Height (scale upper left) is in thousands of feet.
Estimating the Time and Location of When the Fatal Lightning Flash Occurred
In order to observe which lightning flash from the National Lightning Data Network (NLDN) 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 victim(s) location is well documented. In this case study, the victim was not moved after being struck, and rescue authorities recorded the location of the incident with a GPS unit (in this case, latitude 37.990'N; longitude 105.625'W).
Knowing the exact time of when a lightning flash incident occurs can be difficult at times, especially if it is only one victim and no other people were nearby when the lightning flash hit. However, in this case study, the confidence of when the time the flash occurred is high. The time of the incident was best estimated from the husbands watch, which as rendered inactive by the lightning flash. According to authorities, the mans watch stopped at 4:35 pm mdt (22:35 UTC).
Once the temporal and spatial information is known, then
it is a matter of reviewing the NLDN data set and observe which flash occurred
at the location and time found above. The NLDN data set revealed a negative
24 kA (24,000 Ampheres) flash, with a multiplicity of 2, occurred at 4:36:12
pm mdt (22:36:12 UTC) at a latitude of 37.99513 N, longitude 105.611725'w,
very close to location of where the authorities recorded the victim with
the GPS unit. The NLDN flash data between 22:36:00 and 22:37:00 is shown
in figure 3.
Figure 3. Cloud to ground lightning data (white dashes "-") which occurred between 22:36:00 and 22:37:00 UTC 27 July 2003. Three negative cloud to ground flashes occurred during this one minute time period over the region shown above. The flash just east of Crestone, Colorado is the flash which is believed to have killed the 25 year old female. NOTE: The "+" signs in the image mark the location of the towns in this image, and do not represent positive CG flash locations.
Large scale pattern
Figure 4 is a four panel display of meteorological data forecast to occur at 2300 UTC 27 July 2003. The data displayed is based of the Rapid Update Cycle model run at 2100 UTC. The upper left pane shows winds and wind speeds (image) at 250 mb (jet stream level). The upper right panel shows winds, temperatures, height fields and vorticity fields (image) at the mid levels of the atmosphere (500 mb). The lower left pane shows winds, heights temperatures at 700 mb, along with the averaged relative humidity between the 500 mb and 700 mb layer. The lower right panel shows surface pressure, winds and surface relative humidity.
From figure 4, it is seen that a large area of high pressure is noted aloft, with the center of this high over Utah. Wind flow through a majority of the atmosphere over the Crestone area is from the northeast. This is also confirmed by a the 23 UTC RUC sounding taken over the Crestone area (Figure 5). At both 700 mb and the surface, a trough of lower pressure is noted extending northeast-southwest from northeast Colorado to Southwest Colorado, extending directly across the Sangre De Cristo mountains. Deep moisture is noted over the Crestone area, as seen both in figures 4 and 5.
Figure 4. 2100 UTC Ruc forecast fields valid at 23 UTC 27 July 2003. The upper left pane shows winds, temperatures and wind speeds (image) at 250 mb (jet stream level). The upper right panel shows winds, temperatures, height fields and vorticity fields (image) at the mid levels of the atmosphere (500 mb). The lower left panel shows winds, heights temperatures at 700 mb, along with the averaged relative humidity between the 500 mb and 700 mb layer (image). The lower right panel shows surface pressure, winds and surface relative humidity. The little "X" immediately to the left of the "H" in "Home" marks the location of the fatal flash.
Figure 5. 2100 UTC RUC sounding valid at 2300 UTC just east of Crestone, Colorado.
Radar and Lightning data
Composite WSR-88D radar data and lightning data where combined to observe the location and motion of the storm on this date. Based on winds blowing from northeast to southwest over southern Colorado, thunderstorms which developed over this region would be expected to move towards the southwest. Figures 6a-p show the radar and lightning data in 5 minute increments between 2125 UTC and 2245 UTC. At 2125 UTC, storms were noted in the Wet Mountain Valley over Westcliffe (click here for movie loop of fig 6a-p). These storms moved to the southwest at 20 mph and were affecting the Sangre De Cristo mountains north of Crestone by 2146 UTC.
Although it is impossible to know, thunder was likely being heard by hikers on the Willow Creek trail as early as 2150 UTC (46 minutes prior to the fatal flash), as the radar/lightning plots were indicating flashes occurring to the north of Crestone at this time. Shortly after 2205 UTC, lightning was noted striking in the immediate vicinity of Crestone, and moderate to heavy rain was falling all along the trail at this time. Cloud to ground lightning continued to strike along and near the trail up to the time of the incident, although the brunt of the lightning and heavier rain activity was south of where the fatality occurred by 2236 UTC.
From an analysis of radar and lightning data shown in Figure 6a-p, it is clear that the couple who were hiking on this day were well aware of the lighting activity prior to the flash striking and killing the 25 year old female hiker. Rain likely began to fall just prior to 2200 UTC, and thunder was likely being heard by this time. Moderate to heavy rain was occurring on the trail head from 2200 UTC up to the time of the fatal flash, and lightning was striking very close by during this time period.
From the radar and lightning data, the fatality occurred while it was raining. The heaviest rain had moved south of the area at the time of the fatal flash (Fig. 7). Numerous flashes had occurred in the immediate vicinity prior to the fatal flash.
This example clearly shows it is not a good idea
to be hiking in the Colorado mountains during the afternoon time period.
Past lightning studies indicate lighting typically begins to occur in the
Colorado High Country as early as 11 am. If you plan to hike in the Colorado
mountains, you should begin your hike early in the morning, and get below
tree line by noon.
Figure 7. Same as Figure 6, except instead of 5 minutes of lightning data displayed, only 1 minute of lighting data is shown. The fatal flash, denoted by the white dash just above the "R" in Crestone is likely the flash that caused the fatality. Lightning data shown occurred between 22:36:00 and 22:37:00. Radar data is a composite image. It takes a little over 5 minutes to complete a composite image. The time shown for the radar data is the beginning of the radar volume scan. NOTE: The blue "+" signs denote the location of the towns, and do not denote the locations of positive cloud to ground flashes
The authors would like to thank Mike Norris, Saguache
county Sheriff, and Ben Black and Danny Haynes, emergency responders for
valuable information regarding this case. Questions regarding this case
should be addressed to Steve Hodanish.