Investigation Report on the Cut Yellow Spur Rope

Figure 3: Close-up photo of the damaged area of the camming device that popped out of the wall after the fall.

         Tests
       RMRG Test Tower Equipment
       Most of the tests described in this report were conducted on RMRG’s 35-foot-high (10.7 m) steel test tower (Figure 4; see the link http://wwwrockymountainrescueorg/randdphp for details on the test tower — original note).

Figure 4. RMRG Test Equipment

         The test tower is equipped with a mechanical hoist, 30 steel plates each weighing 33 pounds (15 kg), totaling 1000 pounds (455 kg), which provide a sufficiently wide range of test loads. The fall impact is triggered by a release mechanism (the “Sea Catch” designed by McMillan), which can be activated manually or via computer. Data is collected using a National Instruments data acquisition card (Model 6251) installed on a laptop and LabView 8.2 software. Multiple sensors record measurements of load, temperature, and distance. The load sensor can measure up to 10000 pounds (44.5 kN), and the data acquisition system records over 2500 samples per second. This system effectively recorded key data for the fall impact tests designed in this report. Additional details regarding the test tower and test equipment can be found in Holden et al. 2009 (Holden, T., May, B., and Farnham, R. (2009), “Use of the ‘Rescue Randy’ Anthropomorphic Test Device in Fall Impact Testing of Rescue Systems,” International Technical Rescue Symposium, Pueblo County. wwwitrsonlineorg/PapersFolder/ ... 2009_ITRSPaper.pdf. — original note)
       At any given time of the year, several research tests are conducted on the RMRG test tower equipment, including safety tests for rescue systems and new equipment. In addition, testing services are provided to other rescue organizations in nearby states whenever time permits. Each scenario-based test on the tower is pre-designed and takes several hours to set up. This investigation involved seven days of testing on the tower.
       Several items in the following tests involve setups where the rope contacts rock. We collected various rocks from the Yellow Spur route with similar density, crystal structure, and sharpness and piled them on the test tower, primarily consisting of thin-layered sandstone (limestone conglomerate) slabs commonly found on the route. The ropes used in the tests were commercially available climbing ropes with diameters ranging from 9.8 mm to 11 mm. In the fall tests, the load sensor was placed on the rope end connecting the climber and the belayer to measure all differences in load during the process. All tests were sequentially recorded on digital video.
       Fall Severity
       The estimated distance of a climber’s “leader fall” (excluding the distance fallen after the rope was cut) was 20–30 feet (6.1–9.1 meters), producing a fall severity of approximately 800 pounds (3.6 kN). This severity figure is based on previous tests and aligns with similar real-world falls. However, the belayer reported feeling significantly less force than expected during the fall. Therefore, we designed two sets of tests to measure the potential force exerted on the belayer during falls under different conditions.
       First, a 165-pound (74.8 kg) rescue anthropomorphic test device was used to simulate a leader fall from approximately 25 feet (7.6 m). The test scenario largely replicated the terrain at the Yellow Spur accident site: a Bluewater Enduro 11mm dynamic climbing rope passed through a carabiner, connected to the belayer via a standard belay device (ATC) fixed to the anchor, with the other end (connected to the device) falling (with no obstacles during the fall). Figure 5 shows the specific setup of this experiment on the tower.

Figure 5. Left: Fall severity test between the climber and the belayer (text box from top to bottom: climber, carabiner, belayer. — translator’s note); Right: Force experienced by the climber and belayer during the leader fall (blue line: climber, red line: belayer, green line: ratio between the two. — translator’s note).

       The load sensors attached to the belayer and the falling climber recorded the force experienced. The right graph in Figure 5 shows how the force varied over time (horizontal axis) (left vertical axis) and the force ratio (right vertical axis). The force ratio is derived from the ratio of the force on the belayer to that on the climber. The force on the climber peaked at approximately 800 pounds (3.6 kN) around 1.2 seconds. The peak force on the belayer was around 600 pounds (2.7 kN). The force ratio was close to 0.7, consistent with the rope moving directly through the carabiner.
       In this test, when the rope pulled the falling dummy through the belay device, a force of 600 pounds (2.7 kN) lifted the belayer off the ground. If the belayer in the Yellow Spur accident had experienced a similar force, he would have been jerked sharply to the side (the second pitch begins with a traverse). Therefore, we conclude that the force on the belayer from the stretched rope in the accident was far less than 600 pounds (2.7 kN).
       Fall Severity on the Edge
       Figure 2 shows Photo 04 of the Camalot camming device and the connected sling and carabiner that we found immediately after the accident. Before the fall, the climber’s rope should have passed from the belayer’s side, through a carabiner, and connected to the climber’s harness. Because the climber fell past this point, the rope would have made a significant bend at the carabiner, and another bend at the rock edge beneath the carabiner connected to the bottom of the sling. We tested the force experienced by the belayer in this scenario on another test day. Figure 6 left shows the setup of the reenactment.

Figure 6. The rope changes direction on the rock and through the carabiner. (The rope on the left indicated by the arrow is the loaded end connected to the climber): Left: Test setup; Right: Force experienced during the change in direction (blue line: climber side, red line: belayer side; green line: ratio between the two; text box: the rope passing through the carabiner changes direction at the rock edge. The carabiner on the wall causes some compression of the rope.)

       We first conducted a quasi-static test simulating the moment of peak load. The lower right end of the rope was attached to the belayer, then, mimicking the accident scene, the rope passed through a carabiner and bent over the rock edge. The climber side was attached with a separate rope and a 1000-pound (455 kg) weight was slowly lowered to simulate the peak stretching force on the rope. The forces on both sides of the carabiner were measured separately.
The main structure of the test environment remained unchanged, but we made some modifications to observe whether and how strongly the carabiner squeezes the rope against the rock, a situation that would reduce the rope tension on the belayer’s side. Test results showed that the crushing caused by the rope bending over the rock edge could reduce the force on the belayer’s side. Figure 6 (right) shows the results of force under different weights in this system. In this table, the force on the climber’s side increased from zero up to a maximum of 1000 pounds (4.4 kN). At the same time, the force on the belayer’s side peaked at about 150 pounds (0.7 kN), with a ratio of approximately 0.15. This set of data illustrates a sharp decrease in force from the climber’s side to the belayer’s side. When the rope on the climber’s side was subjected to 800 pounds (3.6 kN), the belayer’s side only experienced a pull of 100 pounds (0.4 kN).
These tests did not include variations in the angle at which the rope bends over the rock. However, the angle of rope bending exponentially affects the aforementioned reduction ratio of force, meaning that increasing the bend angle significantly reduces the force on the belayer. Since it was difficult to determine the exact angle of rope bending in the Yellow Spur incident, we set the rope bending angle in the tests based on the photo in Figure 2 to approximate the real conditions of the accident. Therefore, the experimental results suggest that in falls on similar terrain, the belayer is very likely to experience minimal pulling force.
         Comparing the damage to the climber’s rope caused by different cutting methods
Figure 7 shows photos of both ends of the broken climber’s rope. The break covered approximately 2 inches of rope. A small section of the rope core was pulled out of the sheath at the time of the accident.

Figure 7. Photos of both ends of the climber’s rope where it was cut.

The tests conducted during the investigation included cutting the rope under various conditions to infer possible causes of rope failure. These tests revealed that ropes cut under different conditions exhibit distinctly different damage characteristics. Figure 8 shows photos of rope breaks under two different test scenarios. In both tests, the rope was loaded with 800 pounds (3.6 kN) and cut by a sharp object. The left image shows the state of the rope after being cut by a sharp blade lightly pressed against the loaded rope. The damage occurred very quickly, with almost no stretching or deformation of the sheath. The right image shows the state of the rope after being cut by a sharp rock. It can be seen that the break in this case is more jagged. In the blade-cut rope, the core and sheath were nearly cut simultaneously, while in the rock-cut rope, the core was stretched and protruded.

Figure 8. Left: Loaded rope cut by a sharp blade. Right: Loaded rope cut by a sharp rock.

The characteristics of the break in the accident rope (Figure 7) match the test result of the rock-cut rope shown in the right image of Figure 8. These findings suggest that at the time of the accident, the rope passed over a sharp object and broke when stretched close to its maximum elongation.
Dynamic Fall Tests
      After completing the relatively static tests mentioned above, the investigation team conducted a series of dynamic fall tests on the RMRG test tower. The primary goal of these tests was to analyze which combinations of force, angle, and rock structure could lead to the rope breakage observed in the accident. During the tests, we attempted to recreate the dynamic falls that might have occurred in the Yellow Spur incident to reproduce the cutting condition on the accident rope.
Rope Failure Test – Direct Contact with a Sharp Edge
As mentioned earlier, photos taken immediately after the accident led investigators to hypothesize that the climber’s rope had passed near a rock edge close to the carabiner at the end of the sling in Figure 2. It is possible that the rope was cut as it passed over this edge. We spent two days investigating the impact of a sharp rock edge on rope breakage during a similar leader fall.
In each test, we placed a rock with a sharp edge on the crossbeam of the test tower. Figure 9 shows a typical pre-fall setup, where the belayer’s side of the rope was connected to a load cell (indicated by the upper arrow in the image). The rope passed through a carabiner, then over the sharp edge of the rock, and below, on the climber’s side, a 200-pound (90.7 kg) weight was suspended (the rope end indicated by the lower arrow in the image). The ropes used in these tests were commercially available 9.8mm dynamic climbing ropes.

Figure 9. Left: Example of a pre-fall setup. Right: Common appearance of partial rope failure.

We conducted multiple fall tests, varying the position of the carabiner near the edge, using different rock edges, and adjusting fall angles so that the rope slid 1 to 3 inches along the sharp rock edge during the simulated fall. Each test caused significant damage to the rope. However, in some cases, the rope was not completely severed. The right image in Figure 9 shows a rope after a fall test, where the sheath was torn but the core largely remained intact.
The left image in Figure 10 shows the data from a test where the rope was completely severed. The blue variable shows the fluctuating forces on the rope as it was temporarily “caught” and released by the friction against the rock edge over time. The rope force sharply dropped to zero after reaching its peak, indicating the point of complete rope failure. In this test, the maximum force on the climber’s side of the rope was just slightly above 1200 pounds (5.3 kN).

Figure 10. Left: Typical load variation values during a rope failure test. Right: Typical damage caused by direct contact with a sharp edge.

The right image in Figure 10 shows the cut area of the rope after complete failure in the test. The sheath tore rapidly, exposing the core directly to the rock edge. However, the core did not break at the same location. We inferred that because the rope was under tension and elongated, it flattened out as it passed over the rock edge, thus keeping the protected part of the core slightly offset for at least a short period. Additionally, the elongated rope had a length in contact with the rock edge, resulting in damage spanning several inches of the rope.
Although several of the above tests led to complete rope failure, their damage characteristics did not match those of the accident rope (Figure 7). The damaged area on the accident rope was very short, suggesting that the rope in contact with the sharp rock edge did not shift during the damage event. Therefore, the test setups or mechanisms in this phase could not produce a rope cut similar to that in the accident.
       Pendulum Rope Failure Test
      Another potential cause of rope failure is a fully loaded rope sliding laterally over a sharp rock edge. Conceptually, this is similar to the previous rope cutting tests where a sharp rock cut a loaded rope: the rock surface in contact with the rope did not change (see right image of Figure 8). The pendulum rope failure test we conducted aimed to simulate a scenario where the climber is not directly above the last protection point, resulting in a pendulum swing after a fall. Based on prior tests on the RMRG test tower, the rope undergoes pendulum motion while highly tensioned. That is, there is first a vertical fall until the rope has enough load to start swinging the falling object toward the last protection point. If there is a sharp edge between the climber and the last protection point, the rope will slide across it.
We spent two days testing the proposed pendulum failure theory. The left image in Figure 11 shows one of several setups used for the pendulum tests. The rope used was a commercially available 10.2mm dynamic climbing rope. (The test rope was thicker than the accident rope, but it can be assumed that a thinner rope would also fail under similar conditions — original note.) We placed a rock with a sharp edge on the test tower. The test rope was hung about 3 feet (1 meter) above the rock, with a weight attached to the lower end. In these tests, the weight was not dropped but was instead swung laterally, pulling the rope across the sharp rock edge in the direction of the yellow arrow in the left image of Figure 11.
The weight was pulled to the right side of the photo, connected to a release mechanism on the other side of the test tower. The blue device clamped to the wooden beam is a smooth metal angle used to prevent the load rope from rubbing against the rock edge before the weight is released. If the rope slides over the wooden board, this device reduces friction between the rope and the board.

Figure 11. Left: Setup for the pendulum test on the test tower. Right: Damage to the rope during the pendulum test.

Figure 12. Recreation of the accident scene on the Yellow Spur wall. Left: Before the fall. Right: After the fall.

We used various weights to simulate the impact of a pendulum fall. In each test, the area of the rope in contact with the sharp rock edge suffered severe damage during the swing. With a 300-pound (1.3 kN) weight, the rope eventually broke, but only after sliding back and forth over the edge several times. However, according to eyewitnesses at the accident site, no such pendulum swings occurred before the rope was cut. In the remaining pendulum failure tests, we used a 760-pound (3.4 kN) weight. In these tests, the rope broke on the first pass over the rock edge. Slow-motion video footage showed that the rope was cut approximately 2 inches after passing over the edge.
The right image in Figure 11 shows the damage to the rope during the 760-pound (3.4 kN) pendulum test. The damage was confined to a very short section of the rope, with the sheath and core being cut almost at the same location. Compared to the break in the accident rope (Figure 7) and the rock-cut loaded rope (Figure 8, right), this test result shows a slightly cleaner cut. In these tests, more of the core was exposed. Therefore, it can be inferred that there was some degree of elongation at the point of rope breakage in the Yellow Spur accident. Variations in the sharpness of different rock edges may also lead to different types of damage. However, it is clear that this failure setup is similar to the rope failure in the Yellow Spur accident.
        Recreating the Accident on Yellow Spur       
As part of this investigation, RMRG attempted to recreate the fall on the Yellow Spur to evaluate the interaction between the climber’s rope and the rock during the fall. Considering the time and resources required to recreate the accident, the popularity of the Yellow Spur route, and the potential damage to the route itself, we did not attempt to replicate the fall with the same load as the accident. However, based on the known positions of the belayer, climber, and protection points, we set up a fall test with a weight of about 30 pounds (0.1 kN) near the highest point reached by the climber to assess the pendulum characteristics of the accident.
Figure 12 shows the starting point of the second pitch as seen from the belayer’s side (marked by the tree in Figure 1). The investigator in the first photo (left image in Figure 12) was positioned approximately where the fall began. The investigator climbed this route using an RMRG rope (A). (B) is the simulated climber’s rope. Near the end of the climber’s rope, it passed through the investigator’s protection device near the tree. The carabiner (C) was attached to the same 0.4 cam and sling that bore the fall impact during the accident, and we positioned it according to the photographic documentation from the previous investigation (Figure 2). The rope (D) was connected to the climber’s side of the rope to provide weight for the fall.
       During interviews with the rangers, we learned that Miller had fallen vertically from the investigator’s position (left image in Figure 12). In this accident recreation test, the investigator in the photo dropped a simulated climber’s rope vertically without adding any outward or lateral factors. Although the fall forces in the test were much lower than those generated in an actual fall, this setup was sufficient to estimate the general characteristics of the rope movement during the incident.
       The right image in Figure 12 shows the result of a simulated fall of the climber’s rope from the position shown in the left image of Figure 12. In the photo: (A) is the rope attached to the investigator, (B) is the climber’s rope, and (C) is the carabiner. The fall trajectory was approximately 2 or 3 feet (1 meter) to the left of where the rope was at rest, on the climber’s left side. The rope fell vertically, then swung toward the climber’s right (toward the belayer) along the edge below the carabiner, before coming to a stop and getting caught in a rock notch, as shown. The section of rope between the belayer and the carabinet did not come into contact with any rock surface, and there were no obvious objects along the fall trajectory that could have caused the rope to be obstructed. The end of the rope on the climber’s side was hanging freely in the lower part of the right image in Figure 12. Apart from the edge where the rope was caught near the carabiner, there were no other significant edges. Other movements of the rope may have occurred during the fall in the incident, but the resulting position in the test aligns with the information available.
       The left image in Figure 13 shows the final setup above the #0.4 mechanical cam. The climber’s rope swung leftward, slid down over the edge, and eventually came to rest in the notch. The state and position of the carabiner match those in the photos taken immediately after the incident. The right image in Figure 13 shows a close-up of the rope sliding over the edge just before it came to rest in the notch. The notch itself is sharp, and the edge over which the rope slid on the right side is even sharper.

Figure 13. Yellow Spur accident recreation. Left: After the simulated fall (the three arrows in the middle right point to the sharp rock edge, the arrow at the top right indicates the swinging direction, and the left arrow points to the belayer), viewed from above the final position after the fall; Right: Close-up near the edge

       Conclusion
      It is not possible to determine with certainty the exact cause of the rope failure that led to Miller’s death on Yellow Spur. However, a series of specific events likely occurred. First, the investigation found that the rope and associated climbing equipment showed no inherent (human-caused) defects or damage prior to use. Second, Miller encountered difficulties while climbing the route and experienced a typical leader fall, which is a very common occurrence for lead climbers. Third, evidence suggests that Miller placed a piece of protection near the highest point of his fall. Had this piece held, the fall might have been arrested after a very short distance. The pull-out of the highest piece and the distance to the next piece resulted in a long fall. Such a long fall should have been relatively safely arrested and stopped had the rope not been cut. Fourth, the accident recreation tests and observations of the site terrain suggest that the rope, under high tension, likely swung across a sharp rock edge. If Miller’s fall had not involved a swing, the rope might not have been cut while passing over the sharp wall (although it may have been damaged). A slightly smaller impact force might also have prevented the rope from being cut by the sharp edge. In summary, multiple factors contributed to Miller’s death; it was the result of a series of events.
       Lessons for Climbers
       All lead climbers should be aware of the possibility of a leader fall. Climbers assess and manage the level of risk they are willing to accept. Doing this effectively includes understanding the potential consequences of any fall. However, climbers often operate under the assumption that “the rope won’t break” — or at least that it is unlikely for the rope to be cut. As a result, incidents where a rope is cut during a climb attract considerable attention in the climbing community and provide valuable lessons for safety education. From the current investigation and conclusions, we can draw two key lessons.
       Lead climbers often place protection after traversing a protruding ledge to prevent hitting it during a fall. Protection should also be placed to prevent overshooting the ledge, especially when a fall could cause the rope to slide across a sharp edge. Clearly, any ledge with a sharp edge that a climber might pass over during a fall is an extremely hazardous factor. However, the rope failure tests conducted in this investigation highlight two additional considerations. First, lead climbers should try to observe the terrain where a fall might occur over a ledge and weigh whether the potential swinging motion could cause the stretched rope to slide laterally across an edge. Second, climbers must consider how the terrain might change if any piece of protection along the route fails, potentially bringing the rope into contact with nearby sharp edges that were not directly exposed during previous falls. In some cases, if a route has multiple hazards, it may be better to alter the original route to avoid the area or to retreat.
       Acknowledgments
       The members of the Rocky Mountain Rescue Group offer their deepest condolences to the family and friends of the deceased. We thank the belayer and the eyewitnesses for their valuable information during the interviews. We also thank the Boulder County Sheriff’s Office for allowing the investigation team to examine the accident rope and related gear, which were logged as evidence. Finally, we thank all members of the climbing community who followed this investigation and patiently awaited the results of the accident report.