3-D radar imaging unlocks the untapped behavioral and biomechanical archive of Pleistocene ghost tracks

Abstract

Footprint evidence of human-megafauna interactions remains extremely rare in the archaeological and palaeontological records. Recent work suggests ancient playa environments may hold such evidence, though the prints may not be visible. These so-called “ghost tracks” comprise a rich archive of biomechanical and behavioral data that remains mostly unexplored. Here we present evidence for the successful detection and 3-D imaging of such footprints via ground-penetrating radar (GPR), including co-associated mammoth and human prints. Using GPR we have found that track density and faunal diversity may be much greater than realized by the unaided human eye. Our data further suggests that detectable subsurface consolidation below mammoth tracks correlates with typical plantar pressure patterns from extant elephants. This opens future potential for more sophisticated biomechanical studies on the footprints of other extinct land vertebrates. Our approach allows rapid detection and documentation of footprints while enhancing the data available from these fossil archives.

Full text

1
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
3-D radar imaging unlocks
the untapped behavioral and
biomechanical archive of
Pleistocene ghost tracks
Thomas M. Urban1*, Matthew R. Bennett
2*, David Bustos3, Sturt W. Manning1,
Sally C. Reynolds2, Matteo Belvedere
2, Daniel Odess4 & Vincent L. Santucci5
Footprint evidence of human-megafauna interactions remains extremely rare in the archaeological
and palaeontological records. Recent work suggests ancient playa environments may hold such
evidence, though the prints may not be visible. These so-called “ghost tracks” comprise a rich archive of
biomechanical and behavioral data that remains mostly unexplored. Here we present evidence for the
successful detection and 3-D imaging of such footprints via ground-penetrating radar (GPR), including
co-associated mammoth and human prints. Using GPR we have found that track density and faunal
diversity may be much greater than realized by the unaided human eye. Our data further suggests that
detectable subsurface consolidation below mammoth tracks correlates with typical plantar pressure
patterns from extant elephants. This opens future potential for more sophisticated biomechanical
studies on the footprints of other extinct land vertebrates. Our approach allows rapid detection and
documentation of footprints while enhancing the data available from these fossil archives.
Trace fossils in the form of footprints (tracks) occur more frequently in the palaeontological and archaeological
records than is commonly assumed. Holocene and Plio-Pleistocene examples have been described at an increas-
ing number of sites and are found primarily in unlithied, erodible substrates1–5. Footprints provide evidence of
an animal’s presence, pedal anatomy, abundance, co-association with other animals and behavioral ecology, and
have been used to infer not only body size and mass, but also pedal anatomy and biomechanics6. At some loca-
tions, especially in the American southwest, these important yet delicate fossils may hold the key to unanswered
questions about human behavior during the upper Pleistocene, particularly those related to hunting activity, with
footprints oering access to predator-prey interactions outside the more typical “site” locus of a kill or camp7.
How researchers detect and record fossil footprints is a burgeoning area of method development in contemporary
ichnology8–11, and crucial to both maximizing the information yielded and also preserving these fragile traces of
the past12. Former lake beds and playa sites, which occur extensively across the Americas and in parts of Africa,
have the potential to hold these latent ichnological archives.
Here we present ndings from White Sands National Monument (WHSA), New Mexico, USA. Our work
demonstrates the eectiveness and eciency of non-destructive GPR for detecting and documenting fossil foot-
prints in so sediments, including human tracks. Ichnofossils of extinct Rancholabrean fauna occur widely at
WHSA and include tracks of Proboscidea (mammoth), Folivora (ground sloth), Carnivora (canid and felid),
and Artiodactyla (bovid and camelid), as well as humans. ey occur on an extensive gypsum playa (Alkali
Flat, Fig.1), the erosional relict of ancient Lake Otero, dating from the Upper Pleistocene. e sheer number
of tracks, tens of thousands extending over large areas, allows animal and human-animal interactions via true
‘paleo-tracking’ to be deduced7. is valuable resource however is only intermittently and partially visible at the
surface during specic moisture/salt conditions, and when visible may be covered quickly by driing sand. e
1Department of Classics and Cornell Tree Ring Laboratory, Cornell University, Ithaca, NY, 14853-3201, USA. 2Institute
for Studies in Landscapes and Human Evolution, Bournemouth University, Poole, BH12 5BB, UK. 3National Park
Service, White Sands National Monument, P.O. Box 1086, Holloman AFB, NM, 88330, USA. 4National Park Service,
Cultural Resources Directorate, Washington, DC, 20240, USA. 5National Park Service, Geologic Resources Division,
Washington, DC, 20240, USA. *email: tmu3@cornell.edu; mbennett@bournemouth.ac.uk
open
Content courtesy of Springer Nature, terms of use apply. Rights reserved

2
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
occasionally visible tracks are therefore known colloquially as ‘ghost tracks’. Given the scale of the site the resource
management challenges are considerable.
e aims of the GPR survey were: (1) to test whether animal tracks, including those of humans, could be
detected and usefully resolved with this method, including evaluating the potential of GPR to unpack multiple
superimposed track-making events at dierent horizons; and (2) to explore the 3-D subsurface below larger ani-
mal tracks, including how these may be altered by the track-making event. In both cases the geophysical survey
was undertaken prior to any excavation or surface preparation with the exception of the removal of loose surface
pebbles in order to ensure a smooth traverse for the antenna to improve data quality. We found that GPR (1)
allows rapid detection and 3-D recording of multiple species, including humans; and (2) provides non-destructive
information on conditions beneath the track of larger fauna, from which we suggest biomechanical inferences
may be drawn.
Site. On the eastern side of Alkali Flat there is a double trackway of human prints that extends for over 800 m.
is trackway is currently under investigation and appears to represent a single individual walking rst north and
then returning south aer an unknown interval. e two trackways are parallel but o-set by a distance varying
between one and two metres. Individual tracks are visible at the surface under appropriate moisture conditions
Figure 1. Map showing the White Sands National Monument, Alkali Flat, and the study site. Digital elevation
model is from Shuttle Radar Topography Mission 1 arc-second data with the surcial geology taken from U.S.
Geographical Survey maps (relief in .). Note the precise location of the study site is withheld in accordance
with the requirements of the National Parks Service (NPS) compliance with U.S. law. Interested parties may
apply to the NPS for the specic site location. Data from https://www.usgs.gov/centers/eros and map made with
ArcMap 10.1 (http://desktop.arcgis.com/en/arcmap/).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

3
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
and sections of the trackway have been excavated (Fig.2). e GPR survey reported here was undertaken along
this trackway at a location where a series of proboscidean tracks, presumably Mammuthus columbi13, cross in a
westerly direction perpendicular to the two human trackways (Fig.3a). Deformation in front of one of the mam-
moth’s manus tracks partially closed two tracks of the northbound human trackway, showing that the mammoth
crossed that human trackway aer it was made. In turn, a single human footprint of the southbound human
trackway is superimposed on a mammoth manus track, showing that the human crossed the mammoth track on
Figure 2. Photographs of the study site at WHSA. (a) “ghost tracks”: the surface expression of the tracks is poor
as can be seen from the image and they can only be seen under specic moisture and salt conditions. Scale bar
500 mm from target to target. (b) Tracks at the study site excavated to reveal both human and mammoth tracks.
(c) GPR equipment used in this study. (d) Gridded foam mats used to protect the surface during the GPR survey
following Jacob et al.30. (e) Excavated human tracks at the study site.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

4
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
the return. is provides a clear sequence for the tracks, demonstrating co-association. ough the time-lapse
between each of the three track-making events is unknown, the mammoth track is temporally book-ended by
the two human trackways (believed to be the same individual). Current N. American dates for an initial human
presence along with dates for mammoth extinction on the continent support a late Pleistocene biostratigraphical
dating of the tracks at WHSA7,14.
Figure 3. (a) e principal tracks and trackways observed at the study site which are split into Location-1
and Location-2 (shown in true spatial relationship). (b) GPR amplitude slice (2.0 to 4.0 ns). Human prints that
were excavated and used for analysis are indicated with (+) while an unexcavated sloth trackway (identied in
subsequent eldwork) is indicated with (x).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

5
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
Results
Location-1: e results of our GPR survey (Fig.3; see Material and Methods section for parameters) demonstrate
that human footprints can be identied with GPR with horizontal resolution sucient to estimate stride-length
(Fig.4). Table1 compares the stride and step estimates measured aer excavation to those estimated from GPR
data at Location-1. e results are broadly comparable and signicantly show similar variance. e GPR survey
only failed to identity one human track out of the 27 present at Location-1. e survey line interval of 12.5 cm
meant that occasionally a human track could be entirely missed or insuciently sampled, suggesting that 100%
successful detection could be achieved thorough minor modications to the survey design. Importantly, the
survey was able to resolve targets for excavation that are not visible at the surface (Figs3, 4). Such areas warrant
further investigation in the future.
Further, certain 3-D properties of the human footprints also appear to be resolvable within the GPR data, with
the prints exhibiting detectable variation with depth (Fig.4). Detailed topological information for the plantar
surface of human tracks is beyond the resolution of the present data set, but for the majority of tracks visible, both
the presence and relative depth could be inferred from the GPR data. Precise GPR velocity estimates would be
needed to provide accurate absolute depth estimates from two-way travel times. Comparison of excavation depths
of tracks with GPR data, however, suggests a velocity of 0.05–0.06 m/ns, a normal range for a moist, ne-grained
substrate. e potential here for future development is clear and with higher antenna frequencies, closer survey
line spacing, and transillumination (multi-directional data collection), improved 3D resolution should be achiev-
able, including potentially topological information for the plantar surface of smaller tracks, (e.g., human ones).
Location-2: Both mammoth and human tracks have been identied and excavated at Location-2, along with
a range of other potential tracks which have yet to be excavated. For example the anomalies at the western side
Figure 4. 3-D perspective-view GPR results at various depths revealing hidden tracks and volumetric
variations including sub-track consolidation. (a) Just beneath the surface (0–5 cm). (b) 5–10 cm of surface
clipped. (c) 10–15 cm of surface clipped. (d) A photograph of an excavated human print (le) shown beside a
close-up view of the corresponding GPR anomaly (right) at a depth of 5–10 cm, collected prior to excavation.
GPR Derived (mm) Field Derived (mm)
Mean Standard
Deviation NMean Standard
Deviation N
S Bound Foot Size 281.23 2.38 8 266.78 8.31 15
N Bound Foot Size 281.8 2.13 12 257.02 8.02 13
North Step Length 804.7 1.53 12 712.52 2.89 16
South Step Length 798.02 2.62 10 693.14 2.22 14
North Stride Length 1463.33 2.43 12 1422.02 4.21 16
South Stride Length 1451 4.35 9 1404.78 4.18 14
Table 1. Human foot size, stride and step length calculations derived via the GPR survey for Location 1. e
eld derived track data are based on eld survey of excavated tracks using a grid and o-set method, while the
GPR derived track data are based on GPR anomalies marked for Location 1 (Fig.3b).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

6
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
of Location-2 (Figs3, 4) provided future excavation targets that were not visible in any form at the time of the
survey. ese were later determined to be sloth tracks aer becoming partially visible following a period of rain
(though interpretation is preliminary as these remain unexcavated) (Fig.5). e largest of the mammoth foot-
prints (837 × 714 mm; Fig.3, T1) has been excavated (Fig.6c). Deformation in front of this track partially closed
the northbound human tracks 200–300 mm to the west. e track was subsequently overprinted by one of the
south bound human tracks (Fig.6a,c). e exact outline of the mammoth track, believed to be made by a manus,
is not well-dened but by analogy with extant Loxodonta africana would most likely correspond to a mature (>50
years) bull with a potential shoulder height of over 3 metres15–17.
Deformation structures below and around mammoth tracks elsewhere at WHSA have been investigated
using a series of hand-dug trenches. Deformation occurs initially through loading below the anterior part of the
track-maker’s foot, followed by a posterior-directed shear component as toe-o occurs, before nally pressure
release induces uid escape. e specic structures present vary with sub-surface stratigraphy but are always
driven by anterior loading. ese observations compare well to mean plantar pressure data obtained for African
elephants (Loxodonta africana) trained to walk across force plates18. For African elephants mean plantar pressure
peaks in the lateral and distal parts of the foot19,20 which corresponds to areas of maximum sub-track deforma-
tion18. Figure6 shows two depth slices through location-2; the bottom or near bottom of the tracks appear in
Fig.6a (4 to 6 ns), also showing a human print in track 1 (T1), which is further illustrated in Fig.6c. At depths
just below (6 to 8 ns) shown in Fig.6b, GPR amplitude variations are visible across the mammoth tracks, with
higher amplitudes in the anterior and lateral sides of individual tracks. is layer corresponds to that immediately
below the surface of the true track. is is broadly similar to the plantar pressure observed by Panagiotopoulou
et al.20 for extant African elephants (Loxodonta africana; Fig.6d). In vertical cross section (Fig.7a) the amplitude
pattern continuing below the tracks has a hook-shape with the apex at the anterior side of the track; that is in
the direction of travel. Bennett et al.18 have shown that this corresponds to an area of peak deformation below
mammoth tracks at WHSA due to the biomechanics of the locomotion. e patterns reported here are consist-
ent across all the mammoth tracks in the surveyed area (Fig.7a,b), with a consistent 3-D structure (Fig.7b). We
suggest that variations in amplitude with depth are detecting sediment compression consistent with the plantar
pressure records for extant Proboscidea. Deformation by the weight-load of the mammoth would cause compres-
sion of the substrate and in theory increase electrical permittivity in the compacted sediment, by reducing the
low-permittivity air-fraction per unit volume. In turn this should increase the surface area per unit volume, along
with associated soil moisture, and the net result is a slight reduction in radar velocity resulting in higher ampli-
tude GPR reections where the compacted substrate occurs. We suggest therefore that the GPR data is picking up
some semblance of a plantar pressure record for the extinct mammoth.
Discussion
We have recently reported on the use of magnetic sensors to detect mammoth foot prints at WHSA11 and the lim-
itations reported in that work are overcome by the use of GPR in the work reported here. First, the magnetometer
is less reliable in detecting smaller tracks such as those made by humans. Human tracks are only detected if the
base of the true track (plantar contact surface) is deeply impressed relative to the surface (>200 mm). Track depth
and therefore ll-volume appears critical. Mammoth tracks are always detected, however. e instrument is also
subject to intermittent external electromagnetic noise from nearby military activity which is specic to WHSA
but none the less important. Lastly, magnetic data are not well-suited to providing 3D data, which are especially
valuable in instances where tracks comprise a palimpsest of overwritten track-making events7. It does, however,
Figure 5. Close up 3-D GPR perspective of mammoth track 1 (T 1) along with human and sloth tracks. e
presence of sloth tracks was not known prior to the GPR survey and was later conrmed when the prints
became partially visible aer a period of precipitation. e sloth prints remain unexcavated at this location.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

7
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
Figure 6. Location- 2 GPR slices at given travel-times with comparative data. Each 2 ns slice represents
an estimated thickness of 5–6 cm. (a) Bottom or near bottom of true tracks, with a superimposed human
footprint seen in the perimeter of T1. (b) Slice immediately beneath the tracks yield amplitude patterns similar
to observed plantar pressure data. (c) Depths derived from excavation of the largest mammoth track (T1)
with human print. (d) Mean plantar pressure data from African elephants (Loxodonta africana), curtesy of
Panagiotopoulou et al.20. is based on the mean records of ve elephants and the methods are described in
Panagiotopoulou et al.20. Means for each of the ve animals are based on between 1 and 24 pressure records.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

8
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
oer a method of broader prospection that could be applied where there are surface-evident tracks, or for gen-
eral reconnaissance of suspected track locations, particularly for larger fauna. For example, shore normal survey
transects at playas and former lake beds should in theory identify tracks by analogy with the types of distribution
described in recent eld observations21,22.
GPR surveying, on the other hand, was able to resolve 96% of the human tracks present and all of the larger
vertebrates. Results suggest, however, that with rened survey design all human tracks could be detected simply
as a matter of scaling. e tracks are detectable due to the inll which exhibits higher amplitude GPR reections
than the surrounding substrate. is contrast is likely due to stronger electrical properties within the tracks than
in the surrounding sediment matrix and the higher electrical permittivity probably reects a dierence in water
content due to textural contrasts. In this case, the GPR response suggests that the substrate lling the prints holds
more moisture than the surrounding sediment even under dry conditions, something that is evident when the
tracks are excavated.
However, it is the subsurface amplitude variations below the mammoth tracks that we consider to be of par-
ticular signicance. ese are manifest as areas of higher amplitude that likely have a dierent explanation from
that postulated above for the prints themselves. With the sub-track anomalies, it is not newly introduced sedi-
ments that explain the reections, but we theorize this is caused by compression of the existing sediments which
alters the electrical permittivity. e similarity between the plantar pressure records of extant elephants and the
areas of anomalous high amplitude immediately beneath the area the footprint is striking, and suggests GPR
is detecting a plantar pressure record below the mammoth track due, we suggest, to compression of sub-track
sediment. is proposed compression pattern was not visible with excavation, and appears at least at the tested
location to be only discernible with GPR.
Conventional biomechanical inference from footprints oen relies on a pressure to depth substitution in
which deeper areas reect higher plantar pressures. is has been found to only hold however for shallow tracks
in the case of human footprints23, but is something that is rarely questioned in ichnological studies for larger
or extinct trackmakers. Other experimental studies that have examined this relationship in modern human
footprints have also found the relationship to be tenuous24,25. ere are many reasons why strain may not be
solely accommodated by footprint volume (i.e. depth), and moreover taphonomic processes can rapidly modify
observed footprint topology obscuring any relationship26. For example, the topology of the manus track (T 1) in
Location-2 (Figs2b, 6c) does not reect the sub-surface pressure anomalies (Fig.6b). Our results suggest that
irrespective of variation in track depth, a pressure record is encoded via sub-track sediment properties (likely
compression), and in some cases this is independent of the track and its topology.
Taken to its logical extent, potentially thousands of plantar pressure records are waiting therefore to be col-
lected at sites like WHSA and elsewhere in North America1,7 and Namibia26,27 without the need for tracks to be
excavated. e potential here to enhance our understanding of the biomechanics of extinct animals may yield
important information for developing more sophisticated biomechanical models from and for extant elephants
and by analogy from anatomically similar dinosaurs such as sauropods28. It may also improve the quality of geo-
technical models applied to both elephants and mammoth tracks since it would allow estimated plantar pressures
to be used rather than as now uniform indenters29.
At sites such as White Sands, a new data archive in the form of a rare and unique ichnological record is there
to be unlocked. Accessing this record in all conditions using non-destructive geophysical methods has signi-
cant implications for the eectiveness of research, conservation and resource management at White Sands and
Figure 7. (a) GPR fence diagram showing amplitude variations below the mammoth tracks. Note the ‘hook-
shaped’ structure indicated by the white arrows consistently points in the mammoth’s travel direction. (b)
GPR amplitude isosurface from T3 rotated to show several perspectives of the 3-D structure of the sub-track
anomaly.
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
beyond. Knowing where unexcavated tracks are located is a key step in management, monitoring and prioritiza-
tion for resource conservation, especially since excavation leads ultimately to loss. With higher resolution surveys
possible, we anticipate in the future that excavation may not always be necessary on an extensive scale for study of
such tracks. Unpublished experimentation with dierent antenna frequencies has shown that the most consistent
results are obtained with a 250 Hz antenna, at least at WHSA where the substrate exhibits high electrical conduc-
tivity, and that enhanced resolution can be obtained by increasing the grid resolution and combining perpen-
dicular lines. Our unpublished tests with higher antenna frequencies (500 MHz and 1 GHz) show some promise
in obtaining ner detail on smaller tracks in some cases, such as those of humans, which will be the subject of a
future paper. Irrespective of this, the power of the approach lies not just in imaging buried track topology but in
the additional information obtained on sub-track compression. So beyond the immediate and obvious benets
of locating and imaging the tracks themselves, GPR oers ancillary information on pressure and momentum due
to detectable eects on sediments below and around the tracks. Resolvable and consistent trends in the GPR data
suggest that each footprint has an associated sub-structure caused by compaction of surrounding sediment. Initial
ndings suggest that this is related both to compression from the weight of the track-maker, along with shearing
forces from the momentum of the trackmaker. erefore, information about the size and direction of the track-
maker are likely exhibited in the broader GPR patterns, oering a previously uninvestigated avenue into the bio-
mechanics of extinct species such as the mammoth. is has implications for the study of fossil tracks well beyond
White Sands, including the possibility that under suitable conditions these sub-track compaction contrasts could
be retained aer lithication and therefore be present below the tracks of dinosaurs or other extinct vertebrates.
Materials and Methods
e study site was selected from the total length of the double trackway (>800 m) on the basis of having a com-
bination of dierent animal tracks. e area to be surveyed was staked out and photographed before and aer
excavation using a monopole. Ortho-rectied photo mosaics were made using Agiso Photoscan Pro (v. 1.5.0;
www.agiso.com). Tracks were excavated with brushes and dental picks. Once excavated 3D models of the tracks
were made using DigTrace (www.digtrace.com) which uses OpenMVG an open-source structure from motion
photogrammetry engine. Plantar pressure records were proved by Olga Panagiotopoulou19. is consisted of
multiple plantar pressure records for each foot of 5 African elephants collected using a Zebris Medical plantar
pressure platform. Means for each foot were co-registered and combined to produce the data.
GPR equipment included a Noggin series 250 MHz unit by Sensors and Soware Inc. (https://www.senso.
ca/products/noggin/overview/). Instrument parameters included a 120 nanosecond time-window, 8 stacks, and
5 cm trace interval. GPR data processing was done with EKKO_View; EKKO_mapper (Sensors and Soware Inc.)
and included dewow, gain, and envelope. GPR images were produced with VOXLER 4 and Surfer 14 by Golden
Soware Inc. www.goldensoware.com/products
10 mm thick interlocking foam mats were used to cover the site in sections for the geophysical survey. ese
were pre-marked with lines at 12.5 cm intervals. e GPR line spacing of 12.5 cm gave coverage of 18 m²/hr, for
parallel transects, inclusive of setting up and moving the protective foam pads. Although intensive, this provided
a fairly rapid and non-destructive option to gather information where tracks are suspected. Once excavated, these
tracks begin to degrade immediately and need to be digitized in 3D (via photogrammetry, or optical laser scan-
ner), or physically cast for conservation purposes. By comparison, excavation and 3D capture, via photogramme-
try, of Location-1 took approximately 26 hours or 0.15 m²/hr.
Data availability
Data related to this paper may be requested from the authors.
Received: 27 February 2019; Accepted: 25 October 2019;
Published: xx xx xxxx
References
1. etallac, G. J. et al. Late Pleistocene mammoth tracway from Fossil Lae, Oregon. Palaeogeogr. Palaeoclimatol. Palaeoecol. 496,
192–204 (2018).
2. Altamura, F. et al. Archaeology and ichnology at Gombore II-2, Mela unture, Ethiopia: everyday life of a mixed-age hominin
group 700,000 years ago. Sci. Reports 8, 2815, https://doi.org/10.1038/s41598-018-21158-7 (2018).
3. McLaren, D. et al. Terminal Pleistocene epoch human footprints from the Pacic coast of Canada. PLOS One 0193522, https://doi.
org/10.1371/journal.pone.0193522 (2018)
4. Helm, C. W. et al. A New Pleistocene Hominin Tracsite from the Cape South Coast, South Africa. Sci. Reports 8, 3772, https://doi.
org/10.1038/s41598-018-22059-5 (2018).
5. Hatala, . G. et al. Hominin trac assemblages from Oote Member deposits near Ileret, enya and their implications for
understanding fossil hominin paleobiology at 1.5 Ma. Journal of Human Evolution 112, 93–104 (2017).
6. Bennett, M. . & Morse, S. A. Human Footprints: Fossilised Locomotion? 216 pp (Springer International Publishing, 2014).
7. Bustos, D. et al. Footprints preserve terminal Pleistocene hunt? Human-sloth interactions in North America. Sci. Adv., 4, p. eaar7621
(2018).
8. Belvedere, M. et al. Stat-tracs and mediotypes: powerful tools for modern ichnology based on 3D models. PeerJ 6, e4247, https://
doi.org/10.7717/peerj.4247 (2018).
9. Falingham, P. L. et al. A standard protocol for documenting modern and fossil ichnological data. Palaeontology 61, 469–480 (2018).
10. Bennett M. . & Buda M. (2019) Digital Technology for Forensic Footwear Analysis and Vertebrate Ichnology, 251 pp (Springer-
Nature, Cham, 2018).
11. Urban, T. M., Bustos, D., Jaeway, J., Manning, S. W. & Bennett, M. . Use of magnetometry for detecting and documenting multi-
species Pleistocene megafauna tracs. Quat. Sci Rev. 199, 206–213 (2018).
12. Bennett, M. ., Falingham, P., Morse, S. A., Bates, . & Crompton, . H. Preserving the Impossible: Conservation of So-Sediment
Hominin Footprint Sites and Strategies for ree-Dimensional Digital Data Capture. PloS One 8, e60755, https://doi.org/10.1371/
journal.pone.0060755 (2013).
Content courtesy of Springer Nature, terms of use apply. Rights reserved

10
Scientific RepoRtS | (2019) 9:16470 | https://doi.org/10.1038/s41598-019-52996-8
www.nature.com/scientificreports
www.nature.com/scientificreports/
13. Lucas, S. G. et al. Mammoth footprints from the upper Pleistocene of the Tularosa Basin, Doña Ana County, New Mexico. New
Mexico Mus. Nat. Hist. Sci. Bull. 42, 149–154 (2007).
14. Moreno-Mayar, J. V. et al. Early human dispersals within the Americas Science 362, 6419, https://doi.org/10.1126/science.aav2621
eaav2621 (2018).
15. Western, D., Moss, C. & Georgiadis, N. Age estimation and population age structure of elephants from footprint dimensions. Journal
Wildlife Man/ 47, 1192–1197 (1983).
16. McNeil, P., Hills, L. V., ooyman, B. & Tolman, S. M. Mammoth tracs indicate a declining Late Pleistocene population in
southwestern Alberta, Canada. Quat. Sci. Rev. 24, 1253–1259 (2005).
17. Paseno, M. . Quantitative and qualitative data of footprints produced by Asian (Elephas maximus) and African (Loxodonta
africana) elephants and with a discussion of signicance towards fossilized proboscidean footprints. Quat. Int. 443, 221–227 (2017).
18 . Bennett, M. . et al. So-sediment deformation below mammoth tracs at White Sands National Monument (WHSA, New Mexico):
implications for biomechanical inference from tracs. Palaeo 3(527), 25–38 (2019).
19. Panagiotopoulou, O., Patay, T. C., Hill, Z. & Hutchinson, J. . Statistical parametric mapping of the regional distribution and
ontogenetic scaling of foot pressures during waling in Asian elephants (Elephas maximus). J. Exp. Biol. 215, 1584–1593, https://doi.
org/10.1242/jeb.065862 (2012).
20. Panagiotopou l ou, O. et al. Foot pressure distributions during waling in African elephants (Loxodonta africana). Royal Society open
science 3(10), 160203 (2016).
21. Cohen, A., Locley, M., Halfpenny, J. & Michel, A. E. Modern vertebrate trac taphonomy at Lae Manyara, Tanzania. Palaios 6,
371–389 (1991).
22. Cohen, A. S., Halfpenny, J., Locley, M. & Michel, E. Modern vertebrate tracs from Lae Manyara, Tanzania and their
paleobiological implications. Paleobiology 19, 433–458 (1993).
23. Bates, . T. et al. Does footprint depth correlate with foot motion and pressure? Journal of the Royal Society Interface 10, p.20130009
(2013).
24. D’Aout, ., Meert, L., Van Gheluwe, B., De Clercq, D. & Aerts, P. Experimentally generated footprints in sand: analysis and
consequences for the interpretation of fossil and forensic footprints. Am. J. Phys. Anthropol. 141, 515–525 (2010).
25. Hatala, . G., Dingwall, H. L., Wunderlich, . E. & ichmond, B. G. e relationship between plantar pressure and footprint shape.
Journal of Human Evolution 65, 21–28 (2013).
26. inahan, J., Pallet, J., Vogel, J., Ward, J. & Lindique, M. e occurrence of elephant tracs in the silt deposits of the lower uiseb
iver, Namibia. Cimbebasia 13, (37–43 (1991).
27. Morse, S. A. et al. Holocene footprints in Namibia: the inuence of substrate on footprint variability. Am. J. Phys. Anthropol. 151,
265–279 (2013).
28. Hutchinson, J. ., Miller, C., Fritsch, G. & Hildebrandt, T. e anatomical foundation for multidisciplinary studies of animal limb
function: examples from dinosaur and elephant limb imaging studies. In Anatomical Imaging (pp. 23–38). Springer, Toyo (2008).
29. Schanz, T. et al. Quantitative interpretation of tracs for determination of body mass. Plos One 8(10), p.e77606 (2013).
30. Jacob, . W., Berna, F., Urban, T. M. & Chazan, M. Eect of two dierent protective surface material on ground penetrating radar
signal characteristics. Proceedings of the 17th International Conference on Ground Penetrating adar (GP) (2018).
Acknowledgements
is work was funding by a cooperative agreement between White Sands National Monument and Cornell
University. Views and conclusions are those of the authors and do not necessarily reect policies of the National
Park Service (NPS). MRB acknowledges Natural Environment Research Council grants (NE/H004246/1 and NE/
M021459/1). e authors would like to extend their sincere thanks to Olga Panagiotopoulou for providing access
to the African elephant plantar pressure data.
Author contributions
e eldwork was undertaken by T.U., D.B., M.R.B., V.S. and D.O. in April 2018. Geophysical processing was
undertaken by T.U. M.R.B. was responsible for the links to plantar pressure and wrote much of the manuscript
with T.U. M.B., S.M. and S.C.R. contributed to the write-up or added specic perspectives or insight relevant to
their areas of expertise.
Competing interests
e authors declare no competing interests.
Additional information
Correspondence and requests for materials should be addressed to T.M.U. or M.R.B.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Cre-
ative Commons license, and indicate if changes were made. e images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons license and your intended use is not per-
mitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the
copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
© e Author(s) 2019
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com