A large and distinct skin impression on the cast of a sauropod dinosaur footprint from Early Cretaceous floodplain deposits, Korea

The occurrence and features of skin impressions in a sauropod footprint, the largest (>50 cm in diameter) reported to date for this taxon, from the Lower Cretaceous Haman Formation (Albian) in Korea are described, and its preservation and paleoenvironmental implications are interpreted. The skin impression-bearing deposits are floodplain sediments formed by sheetflood processes. The large impression is preserved in silty mudstone with microbial lenses and wisps overlying a planar- to cross-laminated and fine-grained sandstone to siltstone bed. The paleoenvironment of the skin impression-bearing deposits is interpreted as a saline sandflat to mudflat where microbial mats can form around lakes or ponds under semi-arid paleoclimatic conditions with alternating wetting and drying intervals. These paleoenvironmental conditions would have permitted the distinct preservation of skin impressions in a dinosaur footprint. The observations here suggest that some sauropod dinosaurs in the Cretaceous had a well-developed polygonal skin texture covering nearly the whole of their foot pads, as seen in modern elephants, which would increase stability when walking on muddy and wet ground.


Results
Skin impression-bearing deposits. The skin impression-bearing sediments are floodplain deposits formed by sheetflood processes. The deposits consist of interlaminated to interbedded fine-grained sandstone to siltstone and mudstone, planar to cross-laminated fine-grained sandstone, and calcareous silty mudstone (Fig.  1). Current and wave ripple marks and polygonal mud cracks are common in these beds, indicating that these deposits were formed on lake margins. These deposits are usually purple in colour, with the preservation of calcic and vertic paleosols, suggesting that deposition occurred during alternating wet and dry intervals and semi-arid paleoclimatic conditions. Sauropod, ornithopod, and theropod dinosaur footprints are observed intermittently in these deposits, and bird footprints in some areas.

Skin impressions.
In the purple-coloured planar to cross-laminated fine-grained sandstone beds with mudstone films, a large skin impression and its cast, with a diameter of approximately 50 cm and a distinct polygonal texture, are preserved in a subcircular and very shallow (approximately 1 cm deep) footprint of a sauropod dinosaur (Fig. 2). The specimen was found in float with several blocks produced during building construction. The impression and its cast were revealed by parting the impression-bearing bedding plane. The impression is preserved in silty mudstone lamina (a few mm thick) overlying a planar-to cross-laminated and fine-grained sandstone to siltstone bed. The impression-bearing mudstone is composed of mostly clay minerals and some quartz silts (Fig. 3). Thin lenses and wisps of organic material are present in the mudstone lamina (Fig. 3, Supplementary  Information Fig. 2). Casts of sulfate evaporite minerals are present in places in the underlying and overlying deposits, indicating deposition under semi-arid paleoclimatic conditions. The cast of the skin impression is composed of raised hexagonal polygons ranging from 6 mm to 18 mm in width. In general, the sizes of the polygons decrease radially. The polygonal profile has a distinct V-shape with   A polygonal skin impression with rather diffuse edges is also preserved in another subcircular footprint with a very shallow depression (Fig. 4). Partial and indistinct impressions of polygonal skin textures varying from diffuse polygons to ghosts of polygons have been observed in other blocks from these deposits.
The partial preservation of polygonal skin impressions was also observed in another shallow footprint in the interlaminated fine-grained sandstone to siltstone and mudstone bed ( Supplementary Information Fig. 1). While the skin impression described above is preserved in a nearly complete footprint, the impressions in this footprint are partially preserved in small patches. The impression-bearing footprint is approximately 63 cm in length and 46 cm in width. The patches of skin impressions are approximately 14-28 cm in diameter, and the polygons in the impressions are approximately 7-13 mm in diameter.

Discussion
Such variations in the size and distinctiveness of dinosaur footprint skin impressions suggest that certain conditions may be related to variations in preservation. The absence of skin impressions in many distinct dinosaur footprints and their casts suggests that the occurrence of footprints as underprints or overtracks is not the only reason for the very limited presence of skin impressions.
Some conditions may be related to the very rare occurrence of skin impressions in dinosaur footprints. First, if most dinosaurs did not have polygonal skin texture in the foot pads, as seen in most mammals, such skin impressions would have scarcely been formed in their footprints. However, the partial preservation of polygonal skin impressions in dinosaur footprints suggests that the very limited presence of such impressions is attributable to other conditions, such as the substrate on which dinosaurs walked and the taphonomic conditions after the footprints were formed.
It may be necessary for the mud substrate to have a quasi-solid state for footprints to be imprinted 5,19,20 . Thus, skin impressions of foot pads can be formed in the footprints, with slickensides formed at the margins of the footprints. The very rare association of slickensides with distinct dinosaur footprints suggests that many footprints were made when the substrate was saturated with water or that the footprints were flooded shortly after formation and covered with water for some period such that the impressions and slickensides were obliterated by water saturation. Even in cases where slickensides are closely associated with dinosaur footprints, the occurrence of skin impressions is very limited 21 . Therefore, mud surface bearing footprints with skin impressions would have been dry and semi-consolidated enough for footprint-bearing mud not to be saturated by subsequent flooding.
Consequently, the prerequisites for the preservation of distinct skin impressions in footprints can be assumed to be as follows. First, the interlaminated to interbedded nature of sand with mud drapes with thicknesses of a few millimetres would enable tracks to be impressed on the substrates. The mud drapes provided a suitable medium for the preservation of the tracks, with the underlying sands underpinning the impressions and thus preventing their disintegration 20 .
Second, the thickness and viscosity of mud must be suitable for the distinct imprinting of the foot pad skin texture. If the substrate mud is too thick or too thin, distinct skin impressions may not be imprinted. If the water-saturated mud layer is too thick, distinct skin impressions may not be formed due to the irregular Scientific RepoRts | 7: 16339 | DOI:10.1038/s41598-017-16576-y deformation of the mud by trampling. If the mud layer is thinner than the thickness of the skin texture, distinct skin impressions cannot be made in the mud. Therefore, mud drapes a few millimetres in thickness would be adequate for the formation of distinct skin impressions.
There have been no reports of skin impressions with shrinkage cracks in dinosaur footprints. This suggests that the skin-imprinted surface may have been cemented early enough to prevent the formation of desiccation cracks. The formation of microbial biofilms in the trampled layer could have played a role in initial stabilization [22][23][24] . The observation of modern tidal-flats indicates that moist to water-undersaturated deposits with microbial mats provide the best media for the preservation of human footprints 23,25 , supporting the preservation conditions for the distinct skin impressions in dinosaur footprints postulated above.
The variations in the distinctiveness, size, and completeness of the skin impressions in dinosaur footprints observed in this study are likely related to walking speed and the roughness of the surface; dinosaurs walking more slowly on less rough surfaces would have produced more distinct footprints with larger skin impressions. Therefore, the large and distinct skin impression observed in this study could have been formed when a sauropod walked very slowly or stopped momentarily on an organic material-bound smooth surface.
Another feature of the distinct and large skin impression described here is its preservation in a very shallow and gentle depression of a footprint. These observations indicate that the substrate was hardened to an extent, enabling the substrate to resist trampling by the dinosaur. The interbedded sand and mud of the substrate on which the dinosaur walked could have rendered this medium more suitable for these conditions.
The preservation process of the distinct and large skin impression in the dinosaur footprint observed in this study can be postulated as follows. One day, a sauropod walked very slowly or stopped on a smooth and microbially active muddy surface a few millimetres thick overlying sand, forming a distinct imprint of the foot pad skin texture in the footprint. The skin impression-bearing surface dried out during the succeeding drought season, resulting in semi-consolidation of the skin impression. The footprint with the skin impression was flooded during the subsequent rainy season, burying the footprint and preserving the skin impression by sediment deposition.
The paleoenvironment of the skin impression-bearing deposits described above is interpreted as a saline sandflat to mudflat where microbial deposits can form around lakes or ponds under semi-arid paleoclimatic conditions with alternating wet and dry periods. These paleoenvironmental conditions would have permitted the distinct preservation of the skin impression in the dinosaur footprint. Similar paleoenvironmental conditions have been documented in many dinosaur track-bearing deposits around the world 6,22 . Distinct skin impressions in footprints, however, are extremely rare in these deposits, suggesting that most dinosaurs did not have a polygonal skin texture on their foot pads.
Although polygonal skin impressions have been documented from the Triassic to the Cretaceous, most have been found in Middle Jurassic to Cretaceous deposits (Supplementary Information Table S1). Although the statistical significance of these data is low, it is likely that the development of a polygonal skin texture on the foot pads represented an evolutionary adaptation for dinosaurs in the late Mesozoic. The polygonal skin texture on the foot pads of dinosaurs could have played a supplementary role in reducing locomotor pressure and preventing slipping on muddy surfaces 26 . The expansion of the habitats of dinosaurs from forests to alluvial basins, including lakes and ponds, during the late Mesozoic as interpreted from the temporal limitation of dinosaur egg preservation in alluvial deposits to the late Mesozoic 27 could have been related to the adaptation of polygonal skin texture on the foot pads of dinosaurs.

Conclusions
The distinct skin impression and its cast observed in this study are the largest reported to date for a sauropod footprint. The interlaminated to interbedded nature of the sand with microbially active mud drapes a few millimetres thick under semi-arid paleoclimatic conditions could have enabled the polygonal skin texture to be impressed distinctly in the footprints. The observations here suggest that some sauropod dinosaurs in the Cretaceous had a well-developed polygonal skin texture covering nearly the whole of their foot pads to enhance traction (Fig. 5), as seen in the plantar surface with irregular networks of furrows of a modern elephant, and that dinosaurs would have adapted polygonal skin texture on their foot pads as they expanded their habitat into alluvial environments during the late Mesozoic.