Introduction

Gaume and Puzrin (2021) proposed that a delayed slab avalanche at the tent caused "severe but non-lethal" injuries, and compelled the group to descend toward the forest¹. According to the 1959 case files, four of its members (Dubinina, Thibeaux-Brignolle, Zolotaryov and Kolevatov), exhibiting varying injury severity², were found in a ravine, nearby a cedar tree³ roughly 1.5 km downslope from the tent⁴. Here we re-evaluate the proposed avalanche descent sequence using the 1959 autopsy reports and sworn testimony of the forensic examiner B.A. Vozrozhdenniy⁵ as the primary source for all medical and functional constraints, supplemented only by standard physiological frameworks and an empirical model to quantify walking time.

The survival windows and the functional capacities documented in the forensic record are incompatible with the baseline model-derived time required to traverse the ~1.5 km descent. This discrepancy is methodological rather than interpretative: it follows directly from the functional limits documented in the forensic records and persists across the full range of plausible walking-time estimates, irrespective of the modelling approach used to compute them. Forensic survival and functional capacity constitute hard empirical constraints that any physical model must satisfy; they reflect documented physiological limits and cannot be overridden by mechanical force-tolerance estimates alone.

Forensic Analysis

All medical and functional constraints are drawn directly from the 1959 autopsy reports and the sworn testimony of forensic examiner B.A. Vozrozhdenniy.

Standard trauma physiology (e.g. ATLS) is used solely as a descriptive framework to characterize the physiological implications of these documented injuries. It introduces no new assumptions and does not extend beyond what is explicit in the 1959 record.

Dubinina

Autopsy: bilateral rib fractures (R2-R5 right, R2-R7 left), massive ~1.5 L haemothorax, myocardial contusion2.

Vozrozhdenniy's testimony: Dubinina "died 10-20 minutes after the trauma", "she could have been conscious" and her condition represented "complicated traumatic shock resulting from bilateral rib fractures followed by internal haemorrhaging into the pleural cavity"⁵;

Standard trauma physiology: a haemothorax of this magnitude corresponds to Class III haemorrhagic shock, arising directly from the bilateral rib fractures, which simultaneously produce ventilatory compromise. The fractures and the resulting intrathoracic haemorrhage constitute a single traumatic complex; the autopsy provides no support for delayed or staged progression. Sustained exertion over 20-21 minutes is physiologically incompatible with Class III shock, whose cardiopulmonary failure is accelerated by the cold exposure.

Thibeaux-Brignolle

Autopsy: depressed multi-fragment basilar skull fracture with extensive comminution and massive cerebral oedema².

Vozrozhdenniy's testimony: after such a trauma Thibeaux-Brignolle "would have had a severe concussion, that is, he would have been in an unconscious state", that "moving him would have been difficult and, close to the end, movement would not have been possible even if he had been helped. He could only have been carried or dragged", while possibly showing "signs of life for 2-3 hours".

Standard trauma physiology: basilar skull fractures with extensive comminution can preserve residual cardio respiratory activity while eliminating voluntary motor capacity, consistent with Vozrozhdenniy's description.

Transporting an unconscious adult under such conditions would produce widened, irregular or overlapping track patterns, none of which were reported in the 1959 search record.

Zolotaryov

Autopsy: right-sided flail chest (ribs II-VI fractured along mid-sternal and mid-axillary lines) with ~1 L haemothorax².

Standard trauma physiology: this injury pattern constitutes a unified thoracic trauma complex: chest wall dissociation produces paradoxical respiration, pulmonary contusion and hypoxemia, while the haemothorax produces Class II haemorrhagic shock. Flail chest physiology does not permit sustained exertion; the ventilatory deficit results from mechanical loss of chest wall integrity, not from the pain or individual tolerance. Cold exposure accelerates the lethal triad of hypothermia, coagulopathy and acidosis. Even brief exertion aggravates hypoxemia; no form of sustained human locomotion can physiologically compensate for these combined deficits.

Kolevatov

Vozrozhdenniy attributed death to hypothermia. No ante-mortem trauma impairing locomotion documented.

Walking time estimation

According to the 1959 case files, the cedar and the nearby ravine lie ~1.5 km downslope from the tent³. Walking speed was estimated using a validated empirical model for off-road locomotion, solely to quantify the descent duration implied by the 1959 scenario. No dated photographic evidence documents the state of the snow surface near the tent during the night of the incident. The only contemporaneous primary description is Chernyshev's 1959 testimony, noting tracks appearing as "elevations in the form of bars" with "toes imprinted"⁸. This morphology indicates a deformable, non-consolidated snowpack rather than a hard-packed surface. Within the locomotion model such deformability corresponds to heavy-obstruction and represents the most conservative classification supported by the record.

For a mean slope of 9.4° (see Supplementary Information) and heavy-obstruction terrain, the empirical walking-speed model predicts a velocity of 4.2 km/h. Accordingly, the ~1.5 km traverse requires approximately 20-21 minutes for an uninjured group. A five-segment sensitivity analysis is provided as Supplementary Information, yielding a traversal time consistent with this baseline.

No adjustment is applied for injuries, hypothermia or lack of footwear, as no empirical locomotion data exist for such conditions. This does not introduce uncertainty: any impairment would be expected to prolong the descent rather than shorten it.

Integration of Forensic and Locomotion Constraints

The 1959 track record describes 8-9 distinct sets of walking footprints without indications of widening, irregular gaits or dragging. No trace evidence supports the transport of injured adults. Taken together, the forensic constraints (loss of consciousness, haemorrhagic shock, flail chest physiology), survival windows and locomotion requirements are internally incompatible with a tent-site origin of the thoracic and cranial trauma. This discrepancy arises from integrating two independent components:

1. the functional limitations documented explicitly in the 1959 forensic records, interpreted within standard, non-controversial trauma physiology; and
2. a baseline model-derived estimate of the time required for an uninjured group to traverse the ~1.5 km descent, quantified using a contemporary empirical walking speed model.

The physiological constraints themselves derive solely from the autopsies and testimony; the locomotion model is used only to estimate the duration of the descent, not to interfere or modify any medical conclusion.

Across all plausible parameter choices consistent with the 1959 description of a deformable, non-consolidated snow surface, the traversal time remains comparable to (or exceeds) the upper bound of Dubinina's documented survival windows, and remains incompatible with the immediate unconsciousness of Thibeaux-Brignolle and the ventilatory/circulatory impairment described in Zolotaryov.

Thus the incompatibility does not depend on a specific model but on the consistency of forensic constraints with the minimum feasible descent duration.

Conclusion

The avalanche-initiated sequence proposed by Gaume & Puzrin does not satisfy the survival time and functional capacity constraints established in the 1959 forensic records. Under the documented physiological limitations, the thoracic and cranial trauma cannot originate at the tent while still permitting completion of a 1.5 km descent. Conversely if the descent occurred before the trauma, the tent-site slab-release model does not account for these injuries. The scenario therefore contains an internal methodological inconsistency: its physical trigger cannot be reconciled with the physiological capacities of the victims as documented in 1959. This Matters Arising does not propose an alternative mechanism; its scope is limited to assessing the internal consistency of the tent-site impact hypothesis under the constraints of the contemporaneous forensic evidence.

Introduction

This Supplementary Information provides supporting methodological details and extended data for the accompanying Matters Arising.

Forensic assessments derive exclusively from the 1959 investigation record, with modern trauma physiology (e.g. ATLS) used only to contextualise and confirm the physiological implications already inherent in the 1959 expert's statements.

Walking time estimation relies on a validated modern empirical locomotion model, stated explicitly below.

Time-to-descent components

Digital Elevation Data

All elevation values were extracted from the 2-m ArcticDEM. The DEM was reprojected to UTM Zone 40N (EPSG 32640) to ensure metric accuracy of horizontal distances.

Slope along the tent-to-cedar route was derived from the 2-m ArcticDEM.

Slope Measurement

Slope along the tent-to-cedar transect was computed using the QGIS Profile Tool applied to a straight-line polyline between the tent site and the cedar tree. Local 2-m fluctuations reflect micro-relief and DEM noise ; therefore only aggregated metrics are used to characterise the effective gradient.

The mean downhill slope of the transect is -9.4°.

Digital Elevation Data

Walking time estimation model

Walking speed was estimated using the empirical model of Wood et al. (2023), derived from large scale GPS observations of human walking across diverse terrains. The model incorporates terrain obstruction as a categorical variable. This allows assignment to the "heavy-obstruction" class when the primary-source evidence indicates a deformable surface.

The Wood et al. model predicts walking speed as following:

v=exp( a+b ϕ+c θ+d θ2 )

where
v = walking speed (km/h)
ϕ = hill slope angle (degrees)
θ = walking slope angle (degrees)
.

Coefficients for an off-road, heavy-obstruction terrain are:
a = 1.443
b = -0.00731
c = -0.00965
d = -0.00187

Surface classification (heavy-obstruction) followed the criteria and justification provided in the main text, only the numerical implementation is shown here.

Based on average slope (-9.4°), the predicted speed for an uninjured individual is 4.2 km/h, yielding a descent time of 20-21 min (20 min 12 s).

Segment-Level Sensitivity Analysis

To test sensitivity to local variations in slope, the transect was partitioned into four 300-m segments and a final 218-m segment. For each segment mean slope was computed and converted into a predicted walking time using the Wood et al. model.

Table 1 - Predicted Time by Segment

The total predicted descent time is 20 min 37 s, consistent with the results based on the mean slope.

This analysis shows that plausible segment-to-segment variations do not materially affect the traversal time.

Conclusion

These computations provide the quantitative locomotion baseline needed to assess the feasibility of the 1959 tent-to-cedar descent. All input data (2-m elevation samples) are supplied in Supplementary_Data_1_Descent_Profile_Coordinates.csv to ensure full reproducibility.

Forensic constraints table

Extended Data Table 2 (below) compiles for the 3 incapacitated victims found in the ravine:

1. the autopsy findings (1959),
2. the contemporaneous forensic conclusions recorded in the case file, and
3. the corresponding functional implications derived from established modern trauma physiology (e.g. ATLS), which are used solely to express the immediate functional consequences of the injuries described in the 1959 records.

This integrated summary supports the incompatibility between the expected traversal time (20-21 min) and the physiological capacities of Dubinina, Thibeaux-Brignolle and Zolotaryov.

Table 2 - Key autopsy findings (1959)

References

1. Gaume, J. & Puzrin, A.M., Mechanisms of slab avalanche release and impact in the Dyatlov Pass incident in 1959. Communications Earth & Environment 2, 10 (2021)
2. Sverdlovsk Regional Forensic Bureau, Autopsy records of the Dyatlov Pass incident (1959). English translation and scans available at dyatlovpass.com/case-files
3. Tempalov, Kuznetsov, Askinadzi, Ortyukov, Gilev, Fyodorov, Protocol of inspection of the scene where the bodies were found (6 May 1959). Dyatlov Pass Case Files Volume I, p.341 (1959).
   English translation and scans available at dyatlovpass.com/case-files-341-343
4. Ivanov, Lukin, Resolution to close the case (28 May 1959). Dyatlov Pass Case Files Volume I, p.384 (1959).
   English translation and scans available at dyatlovpass.com/case-files-384-387
5. Vozrozhdenniy, Ivanov, Testimony of forensic expert Vozrozhdenniy (28 May 1959). Dyatlov Pass Case Files Volume I, p.381 (1959).
   English translation and scans available at dyatlovpass.com/case-files-381-383
6. American College of Surgeons, ATLS: Advanced Trauma Life Support® Student Course Manual, 10th edn. (ACS, 2018)
7. Wood A., Mackaness W., Simpson, T.I & Armstrong, J.D., Improved prediction of hiking speeds using a data driven approach. PLoS One, 18 Dec 2023; doi.org/10.1371/journal.pone.0295848
8. Chernyshev, Ivanov, Chernyshev witness testimony (11 March 1959). Dyatlov Pass Case Files Volume I, p.88 (1959).
   English translation and scans available at dyatlovpass.com/case-files-88-93
9. Maslennikov, Romanov, Additional questioning Maslennikov (15 April 1959). Dyatlov Pass Case Files Volume I, p.295 (1959)
   English translation and scans available at dyatlovpass.com/case-files-295-297

Declarations

Data availability

All data used are from publicly available sources as cited.

The ArcticDEM dataset is available from the Polar Geospatial Center (pgc.umn.edu/data/arcticdem/)

The coordinate data generated for the descent profile reconstruction in this study are provided as Supplementary Data 1.

Competing interests

The author declares no competing interests.

Author contributions

Caroline Detry is the sole author of this work.