Sauropod dinosaur tracks from the Purbeck Group (Early Cretaceous) of Spyway Quarry, Dorset, UK

Headnote

Dinosaur tracks have a long history of discovery and study in the UK, but track sites for sauropodomorph dinosaurs- the group that included the giant, graviportal herbivorous sauropods-are comparatively rare. Here, we provide a description of a sauropod dinosaur track site at Spyway Quarry in Dorset, southern England. The tracks at Spyway were discovered in the late 1990s and occur in the Stair Hole Member of the Durlston Formation in the Purbeck Limestone Group, of earliest Cretaceous age. More than 130 individual tracks of large sauropod dinosaurs are present at the site, but they are generally poorly preserved and do not form clear trackways, although it is likely that they represent multiple individuals. They provide further evidence for sauropods living in or passing through coastal lagoonal environments. Although poorly preserved, Spyway represents the largest in situ dinosaur track site currently accessible within the Purbeck Group, with considerable potential for further discoveries through ongoing quarrying in the surrounding area.

Received: 11 April 2024 Accepted: 10 June 2024

Subject Category: Earth and environmental science

Subject Areas: palaeontology

Keywords: sauropod, Cretaceous, Spyway Quarry, Dorset, dinosaur, track

1. Introduction

Dinosaur tracks have a long history of study in the UK, beginning in the mid-nineteenth century (e.g. [1]) but with new sites still being documented (e.g. [2-4]) as part of a wider 'dinosaur track renaissance' (e.g. [5]). Close to 300 distinct dinosaur track occurrences have been documented from the UK,each representing a track morphotype at a particular horizon and locality [6], although in situ dinosaur trackways (as opposed to isolated and often ex situ tracks) are rare in the UK, with approximately 14 key localities currently accessible [6].

Sauropodomorph dinosaurs, the group that includes giant, graviportal herbivorous sauropod dinosaurs such as Brachiosaurus and Diplodocus, originated in the Middle-Late Triassic and survived until the end of the Cretaceous [7], but based on the body fossil record, they were most diverse (at least in the Northern Hemisphere) during the Middle and Late Jurassic (e.g. [8]). A diversity decline of sauropodomorphs is proposed to have occurred during the Early Cretaceous, although interpretations of diversity changes through the Jurassic to Cretaceous transition are complicated by sampling biases [8,9]. Lhe body fossil record of sauropodomorphs in the UK is relatively poor with limited taxonomic diversity (e.g. [10-12]), probably as a result of the limited availability of fossilifer-ous terrestrial sediments in the intervals of peak sauropodomorph diversity during the Jurassic [6]. Likewise, sauropodomorph tracks are rare in the UK, comprising only 16% of occurrences. Lhe UK dinosaur track record is instead dominated by tridactyl (three-toed) tracks produced by theropods or ornithopods, comprising approximately 75% of all known occurrences [6].

Key UK localities preserving sauropodomorph tracks in situ today include Late Lriassic sites at Bendrick Rock and Penarth in South Wales [4,13], Mddle Jurassic sites at Ardley Quarry in Oxfordshire [14-16], Rubha nam Barthairean and Cairidh Ghlumaig in Scotland [2,3] and the Early Cretaceous of Spyway Quarry, near Acton in Dorset [17].

Lhe fossil footprints at Spyway were first identified in 1997 by local quarryman Kevin Keates when removing the Upper Freestone of the Durlston Formation of the Purbeck Limestone Group. Lhe top surface of the underlying Bottom Freestone was covered with numerous large, shallow bowl-shaped depressions. Lhese were subsequently identified by Keates, together with another local quarryman, Lrev Haysom, as potential dinosaur footprints. Lhe land on which the tracks occur is owned by the National Lrust and was leased to Keates Quarries Ltd. Lhe tracksite was known for many years as Keates Quarry, but the name was later changed to Spyway Quarry after the opening of the site for public access in 2016 to minimize confusion with the name of the quarry firm. Following the discovery, quarrying temporarily ceased for an initial scientific investigation commissioned by the National Lrust and led by Joanna Wright [18]. Following this work, the tracks were covered using an anti-weed, geosynthetic fabric covered with crushed stone (K Keates 2021, personal communication) to facilitate quarrying of the immediately surrounding area before being re-exposed in 2013.

While a comprehensive internal National Lrust report was developed documenting the site as initially exposed [18], only a brief half-page report (without figures) was ever formally published [17]. Ferraby & Powesland [19] described a publicly available photogrammetric model of the site produced in 2014 and Edgar et al. [20] described the impacts on the site of it being opened to public access and the elements. However, aside from these geoconservation-focused contributions, the site has been largely invisible in the scientific literature, receiving only brief occasional mentions [21,22]. Lhis, combined with changes in the exposed quarry surface through time yielding multiple additional tracks [20], technological advances in the methods used to record and report tracks (e.g. [5]), and advances in understanding of the UK dinosaur track record [6] mean that the time is ripe for a revised description of the site. Lhus, this study provides an up-to-date description of the Spyway Quarry dinosaur tracks, detailing the nature of the individual tracks and current inferences for the nature of the trackmakers and their behaviour.

2. Geological setting

Spyway Quarry is located between Langton Matravers and Worth Matravers, East Dorset (50.60216° N, 2.02020° W; figure 1), and is locally designated as a regionally important geological site (RIGS; under its previous name of 'Keates Quarry') by Dorset's Important Geological Sites Group. Lhe dinosaur tracks occur on the top surface of a bed locally known as the Bottom Freestone, part of the Freestone Vein within the Stair Hole Member (which incorporates the frequently cited Intermarine Member of Clements [23] within which the Freestone occurs) of the Durlston Formation in the Purbeck Limestone Group [24]. Lhe Stair Hole Member comprises interbedded shelly limestones (biomicrites and biosparites) and mudstones deposited in a marginal marine environment. Lhe Stair Hole Member is dated to the Berriasian of the Early Cretaceous [24]. Approximately 6 m section of the overlying beds of the Stair Hole Member is exposed on the eastern side of the quarry above the track surface (figure 2).

A second discovery of sauropod tracks on the top surface of the Bottom Freestone was made at Lewis Quarries, approximately 200 m east of Spyway Quarry in 2018 [25].

The Freestone Vein includes coarse bioclastic limestones within the Stair Hole Member that have been extensively quarried for building stone for the past few hundred years. The Bottom Freestone is approximately 30 cm thick locally with a pale yellow/white colour on a fresh surface and contains bivalve shells (less than 1 cm) with some larger oyster fragments and ostracods. These sediments are thought to have been deposited on the edge of a freshwater lagoon separated from the open sea [26]. The overlying Upper Freestone, which covers the footprint horizon, is a similar shelly limestone, but shells are more chaotically arranged and inferred to be rapidly deposited, possibly during a storm event [18].

Dinosaur tracks have been discovered at multiple stratigraphic levels within the Tulworth and Durlston formations of the Purbeck Timestone Group [17,27,28]. The greatest abundance of tracks comes from the Intermarine beds within the Stair Hole Member [17,27].

3. Methods

Spyway Quarry was studied during multiple visits between summer 2020 and early 2023, including a detailed examination of the track-bearing surface, sedimentary logging in nearby quarries and study of the remaining Tewis Quarries tracks, which are currently stored at Keates Quarries Ttd. We generated a new photogrammetric model of the whole of Spyway Quarry in August 2021 (figure 3). Our methods largely follow those used by Ferraby & Powesland [19] to generate the publicly available April 2014 model (https://skfb.ly/oDvzt) and are described in detail in Edgar et al. [20]. The 2021 Spyway model is publicly available on Figshare (https://figshare.com/articles/media/ 3D_model-2021_Spyway_Quarry_Dorset_UK/21256107). In this study, we present new photogrammetric models of 109 individual tracks (or pairs of adjacent tracks) from Spyway and two blocks containing tracks from Tewis Quarries using Agisoft Metashape Standard (v. 1.7.5) and Professional (v. 1.7.4) editions, following methods in Edgar et al. [20]. Photos were taken on a Nikon D750 digital STR camera with an 18-55 mm f/3.5-5.6G lens. Fndividual photogrammetric models are available at https://doi.org/ 10.5281/zenodo.11509629. Height maps for selected tracks were generated in the software CloudCom-pare (v. 2.12.3).

To compare the Spyway track surface across the 2014 and 2021 three-dimensional models with the original line map by Wright et al. [18], we created a line map highlighting track positions and other significant features on the quarry surface (figure 4). Without flattening the quarry surface, screenshots of the quarry surface from both three-dimensional models were captured from the default view in Metashape and imported into Fnkscape software. In Inkscape, a new layer was generated, and tracks were drawn free-hand onto this layer. Subsequently, the track layers from 2014 and 2021 were superimposed and juxtaposed with the original line map from Wright et al. [18]. This was done to synchronize track numbering with that of Wright et al. [18] and ensure consistency in track placement, shape and surface exposure over time. The numbering system in figure 4 corresponds to that used in the individual photogrammetric models reposited on Zenodo.

We calculated the maximum diameter and circularity of individual tracks from the final summary scaled line model (figure 4) using the automated image measurement function in ImageJ [29] (figure 5 and electronic supplementary material, table SI). We note that the outline of the track is used incorporating the displacement rim in part because of difficulties consistently determining the exact boundaries of tracks owing to poor preservation. Thus, these outlines are assumed to be minor overestimates of the actual track size. Tracks that are not completely exposed because they are truncated by the edge of the quarry were excluded from the analysis.

Six blocks containing sauropod tracks, each weighing approximately 200-300 kg, were removed from Tewis Quarries as part of the initial conservation plan for the site developed by the National Trust. These blocks are stored outside at Keates Quarries Ttd and are currently accessible with the permission of Kevin Keates. The original and more extensive Tewis Quarries track surface was digitized in 2018 [25], but the model data generated during this work is not publicly accessible and was not available for examination as part of the current study. The site has now been quarried and is no longer accessible. Given the variable quality and completeness of the tracks exposed on the accessible blocks, we present selected three-dimensional models representing two of the blocks as per the photogrammetric methods described for Spyway Quarry.

4. Results

4.1. Spyway Quarry

Wright et al. [18] noted 111 tracks scattered across the bedding surface, with an additional three deemed indeterminate (tracks 1-114 in figure 4). Using the 2014 and 2021 photogrammetric models alongside field observations in 2021, a further 24 tracks were identified, bringing the total number of tracks at the site to 138 (tracks 115-138 in figure 4 [20]). These new tracks are primarily situated along the perimeter of the quarry in the northern, northwestern and eastern sectors, where a larger surface area was exposed in 2014 compared with the present day or that exposed in 1998. The reduction in track surface size since 2014 results from vegetation encroachment and collapse/creep of the embankments. Additionally, several tracks identified in 1998 were no longer visibly distinct on the surface in 2021, while others appeared faint, probably owing to continued erosion [20].

Tracks are generally preserved as subcircular to oval depressions with low raised rims that rise 10-30 mm above the main trackway surface (figure 6). None of the tracks preserve good morphological details such as the impressions of claws or toes. As such, they should be considered poorly preserved (scoring 0 on the preservation scheme of Belvedere & Farlow [30]). We interpret the outlines of the tracks as probably not fully representative of the original morphology in many cases owing to visible erosion of original track rims. In general, there is no clear grouping of individual tracks into trackways, although there are three examples of aligned series of tracks of similar sizes that could potentially represent parts of trackways (tracks 18, 19 and 20; tracks 12, 13 and 14; tracks 103, 104, 111 and 135). The latter two of these sets of aligned tracks run parallel to one another, and it is possible that they could represent left and right tracks within a wide-gauge trackway.

Tracks range from 24 to 130 cm in maximum diameter (electronic supplementary material, table SI) with a mean diameter of approximately 67 ± 22 (la) cm, with most between 50 and 90 cm (figure 5a). Tracks overall have a high mean circularity score of 0.83 ± 0.07 (la) (where 1 equals a perfect circle) appearing largely roughly circular to slightly elliptical in shape (figure 5b). The majority of tracks are most likely pes (i.e. rear foot) based on their shape. However, a small number of manus tracks are probably present, based primarily on an inferred D-shape outline (noting, however, that sauropod tracks interpreted as manus impressions can also appear quite circular in outline, e.g [31]). Several of these manus tracks are closely associated with pes prints. In Wright et al. [18], 14 manus prints were tentatively identified (tracks 13, 17, 21, 23, 27, 29, 34, 36, 37, 39, 42, 49, 59 and 81), of which 13 were described as D-shaped. From an evaluation of the current quarry surface and the 2014 images, we can only identify three of these as manus with reasonable confidence (track 49 paired with pes print 15;tracks 36 and 42; figure 7), with five further tracks possibly identifiable as manus (tracks 13, 17, 21, 27 and 59). The preservation of the site and highly fractured surface make identification of D-shaped tracks challenging.

While the majority of the tracks appear on the surface of a single bedding plane, in tracks 4, 60, 61 and 117, there is evidence of overlying layers present across parts of the track evidencing that the same print and displacement rim occurs in multiple layers (figure 8). Sediment laminations drape the shape of the track and are roughly parallel to one another, including with well-defined displacement rims in places.

Other notable features observed on the bedding surface are two elongate grooves that taper at each end and are independent of the pervasive fracture network: the largest of these (approx. 3 m long and 9 cm wide) is adjacent to tracks 27 and 26 (figure 9) and the second is approximately 19 cm long and approximately 4 cm wide. In both cases, the impression extends below and is slightly overhung by a lip of the bedding surface. The smaller impression has roughly linear fine grooves exposed in its base. A number of hypotheses have been proposed to explain the large elongate groove on the quarry bedding surface including that it is a tail groove or foot drag mark [18]. Given the similarities of the two impressions plus the positions of the imprints relative to the tracks, we suggest that it is more likely that these represent impressions of pieces of wood, which are commonly preserved within the Purbeck Limestone Group (e.g. [32]).

4.2. Lewis Quarries

The six blocks each contain a complete or partial sauropod track (figure 10). Unfortunately, no track labels or connection to the original position of these on the bedding surface is available. The tracks are relatively similar to those at Spyway but more poorly preserved. Of those that appeared relatively complete, maximum diameter ranges from 54 to 82 cm including extensive displacement rims, with a smaller internal measurement of 38-47 cm. They are preserved as shallow sub-circular depressions (approx. 2 cm at the deepest point) on the top surface of the Bottom Freestone and lack any distinguishing features.

5. Discussion

5.1. True tracks versus transmitted tracks

Determining whether tracks are true tracks or transmitted undertracks is potentially important for interpretations of the trackmaker's body size, behaviour, taphonomy and stratigraphic position. Transmitted tracks are often considered of lower scientific value than true tracks, i.e. those forming within the bed on which the dinosaurs directly walked, and the interpretation of the Spyway tracks as transmitted was part of the rationale for the opening of the site to the public without restricting access to the track surface [33]. The initial investigation of the site suggested that the tracks were true tracks [18]. Evidence for this was based on the identification of infilling sediment layers that drape the track and onlap to the edges of the track and sedimentary evidence from the raised rims (unknown track in fig. 7 of Wright et al. [18]). The lack of morphological detail preserved within the tracks was attributed to the sediment type and environmental conditions rather than track transmission, which would result in a progressive loss of detail with increasing depth in the sediment [34]. Subsequent unpublished interpretations [33] suggested a transmitted origin and that the original track surface was higher in the sequence. This question is difficult to answer with certainty for the Spyway tracks. We noted that evidence of tracks in layers overlying the main track surface occurs in several parts of the site, which provides suggestive but not conclusive evidence that the tracks on the main surface are transmitted. Regardless of the interpretation, we noted that the identification of the Spyway tracks as either true or transmitted does not necessarily limit the scientific significance of the site, depending on the questions posed.

5.2. Trackmaker identity

The very large size and sub-circular, bowl-like appearance of the pedal prints, the apparently D-shaped manus print (where preserved) and the earliest Cretaceous stratigraphic occurrence indicate that the tracks at Spyway and nearby Lewis Quarries were made by sauropods, the dominant group of large-bodied herbivorous dinosaurs in the Late Jurassic-earliest Cretaceous interval [8]. Fish-feeding traces can be preserved with a similar appearance to sauropod tracks (e.g. [35]); however, the preservation of well-developed rims and the placement of D-shaped manus tracks in close relationship to more bowl-shaped pes tracks makes this interpretation unlikely for the Spyway tracks. The apparent overall limited preservation of manus tracks is not necessarily unusual, and the occurrence of pes-dominated sauropod track sites has been suggested to relate to the position of the centre of mass of the trackmaker (e.g. [36]).

A more refined taxonomic identification than Sauropoda is hampered by the lack of unambiguous trackways, the limited number of manus-pes sets, and the generally poor preservation of the tracks. As a result, few of the synapomorphies of taxonomic groups within Sauropoda identified by Carrano & Wilson [37] as potentially preservable in tracks can be adequately assessed. One feature that would potentially be significant would be whether trackways have footprints that approach the midline (narrow gauge) or are widely separated from the midline (wide gauge), or are intermediate between the two (intermediate gauge) [37-39]. Narrow gauge trackways have been inferred to be ancestral for sauropods as a whole, with a transition to wide gauge trackways via intermediate gauge trackways proposed to occur within the clade Titanosauriformes (Brachiosauridae + Titanosauria) [37-39]. Early members of this group, such as brachiosaurids, are proposed to have the intermediate gauge track morphology. In two areas of the quarry, there are groups of tracks that might be interpreted as parts of trackways-if the two parallel aligned sets of tracks 12, 13 and 14 and tracks 103, 104, 111 and 135 were interpreted as left and right sides of a single trackway, then this would appear to be wide gauge. However, the identity of the aligned prints as trackways is difficult to confirm. Moreover, it has been documented that the gauge of sauropod tracks can be variable along a trackway dependent on variations in behaviour and substrate (e.g. [40]). As a result, the usefulness of this character for identifying the trackmaker at Spyway is limited.

Sauropod body fossils are extremely rare in the Purbeck Limestone Group [22]. Only isolated teeth and an incomplete metacarpal bone have been discovered, and both are currently identified as Sauropoda indet. [22]. The underlying Portland Group also yields only indeterminate sauropod material [22]. Elsewhere within the UK and Europe, Tithonian-Berriasian sauropod body fossils are also relatively infrequent, but most occurrences have been assigned to either turiasaurs or macro-narians (including titanosauriforms) (e.g. [41-43]), although records of diplodocoids have also been reported (e.g. [44,45]). As such, there are three major groups of sauropods known from the Tithonian-Berriasian interval that could have produced the tracks at Spyway, and the poor preservation of the Spyway tracks does not help in distinguishing between these clades.

Wright et al. [18] identified brachiosaurids (early titanosauriforms within Macronaria) as the most likely trackmaker for the Spyway tracks based on their relatively common body fossil occurrences in this interval and the large size of the tracks. Although this identification is plausible (figure 11), the limited information available for the Spyway tracks means that the trackmaker cannot be confirmed beyond Sauropoda.

5.3. Palaeobiological inferences

Assessment of the number of individuals represented at the site is complicated by the lack of definite trackways and poor preservation. Variation in the maximum dimensions of the prints appears to be normally distributed around a mean value of approximately 67 cm (figure 5), and there is no clear indication of a bimodal distribution or multiple peaks in the data. While the large range in maximum track dimensions may indicate more than one size class of trackmakers, some of this may be variations in size that can be present in transmitted tracks and/or erosion of the boundaries of the tracks (e.g. [34]). The large number of tracks at the site is consistent with the possibility of multiple individuals, although this inference is complicated by the lack of definite trackways. Sauropod trackways recovered elsewhere, including within strata of approximately equivalent age, have provided clear evidence of gregarious behaviour in the clade (e.g. [46,47]). The tracks at Spyway are consistent with such behaviour. They also provide evidence for sauropods inhabiting or at least passing through a coastal lagoonal setting, as also indicated by many previous sauropod track occurrences (e.g. [2,48,49]).

Sauropod hip height has been traditionally estimated from footprints via formulae that use pes length [50-52]. Using a mean track diameter of 67 cm, the equation of Alexander [50] would suggest a minimum hip height of approximately 2.7 m, although this might be a substantial underestimate given the considerably larger size of some of the tracks, which might suggest animals with hip heights up to 4-5 m. The minor overestimation of track dimensions introduced by the production of automated measurements from the track site map (figure 4) is unlikely to have a substantial impact on these size estimates.

5.4. Significance of the site

There are very few examples of sites in the UK that preserve multiple exposed sauropod tracks within a single horizon. Late Triassic trackways from Bendrick Rock in South Wales are of basal sauropodo-morphs rather than true sauropods and are much smaller in size with a different morphology from those at Spyway [13,53]. Tracks within a nearby but stratigraphically younger Late Triassic horizon at Penarth probably represent sauropodomorph, and potentially sauropod, track makers, but their interpretation is somewhat controversial [4] and they are poorly preserved and variably exposed over time ([4]; RJB and KME 2023, personal observation). The originally described trackway surface in

Middle Jurassic deposits at Ardley in Oxfordshire, which preserves multiple sauropod trackways [14-16], has been buried and is no longer accessible, although new exposures of tracks have occurred recently as a result of ongoing quarrying ([54]; RJB and KME 2023, 2024, personal observation). Unfortunately, three-dimensional models of the Ardley site are also not available as it was discovered and documented prior to the widespread use of three-dimensional scanning and photogrammetry approaches. Two sauropod track sites have been recently described in the Middle Jurassic of the Isle of Skye, Scotland [2,3].

Other sauropod trackways and isolated tracks have been reported previously from the Purbeck Limestone Group (e.g. [17,22,27,28]), although these are generally either isolated examples (e.g. [28]) or may in some cases represent other groups such as ankylosaurs [55]. As described above, the sauropod tracks from Lewis Quarries [25] are poorly preserved and the original associations of individual tracks relative to one another have been lost.

As such, Spyway represents the most important record of sauropod tracks from the Purbeck Limestone Group and the largest dinosaur track site within the Group that is currently accessible in situ. In terms of the number of individual tracks, it is also the largest readily accessible dinosaur track site in the UK (given that new exposures of tracks at Ardley Quarry are temporary and not publicly accessible) and thus of considerable scientific significance and popular interest. Although the site is deteriorating owing to exposure to the elements [20], it is likely that the rate of deterioration is slower than at most other UK dinosaur track sites, which are generally exposed on coastal sections. The stratigraphic position of the site in the earliest Cretaceous (Berriasian) is also important, as it documents the presence of large sauropods shortly after the Jurassic/Cretaceous boundary, which has been suggested by some as a possible interval of heightened extinction rates, particularly among sauropodomorphs [8].

6. Conclusions

Here, we have provided the first published scientific description of Spyway Quarry, which preserves more than 130 individual tracks of large sauropod dinosaurs from the earliest Cretaceous of Dorset. Morphological preservation of the tracks is poor and it is not possible to group tracks clearly into trackways, limiting the useful information that can be obtained from them. Despite their poor preservation, the site remains significant as sauropod tracks are rare in the UK, and Spyway is the largest in situ dinosaur track site currently accessible within the Purbeck Group. There is potential for future discoveries through ongoing quarrying in the surrounding area, which could yield specimens that are more informative in terms of trackmaker identification and behavioural inferences.

Ethics. Fieldwork was conducted at Spyway Quarry, near Acton in Dorset, England. This is a publicly accessible site owned and managed by the National Trust with advice from the Jurassic Coast Trust. The site is recognized as a Regionally Important Geological Site (RIGS ID SY97/15). No specific permits were required for access to the site or for our study, which was non-destructive and did not involve the collection of samples. National Trust were aware of and supportive of our research (via National Trust Purbeck Rangers Ben Cooke and Rebecca Eddy) and two staff members from Jurassic Coast Trust (Christopher Reedman and Sam Scriven) were involved in the work and are coauthors on the paper. Further information on the site is included in our previous study on the geoconservation of the site: https://www.sciencedirect.com/science/article/pii/S0016787823000019.

Data accessibility. Data for the 2021 model of Spyway quarry is available at [56]. All newly generated photogrammetric models have been uploaded to Zenodo [57].

Supplementary material is available online [58]. Declaration Of Al use. We have not used Al-assisted technologies in creating this article.

Authors' contributions. R.J.B.: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, writing-original draft, writing-review and editing; K.M.E.: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, visualization, writing-original draft, writing-review and editing; L.H.: formal analysis, investigation; L.E.M.: data curation, formal analysis, investigation, methodology, supervision, visualization, writing-review and editing; H.T.J.: formal analysis, investigation, writing-review and editing; O.H.: formal analysis, investigation; S.S.: conceptualization, investigation, writing-review and editing; C.R.: conceptualization, investigation, writing-review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest. We declare we have no competing interests.

Funding. This work was supported by a Research England's QR Enhancing Research and Knowledge Exchange Funding Programme award and the Geologists' Association Curry Fund to K.M.E. and R.J.B. Elements of this research were conducted by H.T.J., L.H. and O.H. in their undergraduate dissertations within the Palaeontology & Geology programme at the University of Birmingham. Acknowledgements. We thank National Trust Purbeck Rangers Ben Cooke and Rebecca Eddy for their advice and support during this work, Dominic Powlesland for helpful and advice and for making all models freely available, Alan Holiday and Rhiannon Rogers of the Dorset Important Geological Sites Group and father and son Kevin Keates and Kevin Keates of Keates Quarries Ltd for access to active quarrying areas and support. We thank Jens Lallensack, an anonymous reviewer, and the editor, Laura Porro, for their helpful comments that improved the final

References

References

1. Beckles SH. 1851 On supposed casts of footprints in the Wealden. J. Geol. Soc. 7,117. (doi:10.1144/GSL.JGS.1851.007.01 -02.26)

2. Brusatte SL, Challands TJ, Ross DA, Wilkinson M. 2016 Sauropod dinosaur trackways in a Middle Jurassic lagoon on the Isle of Skye, Scotland. Scott J. Geol. 52,1-9. (doi:10.1144/sjg2015-005)

3. dePolo PE, Brusatte SL, Challands TJ, Foffa D, Ross DA, Wilkinson M, Yi H-Y. 2018 A sauropod-dominated tracksite from Rubha nam Brathairean (Brothers' Point), Isle of Skye, Scotland. 5coffJ. Geol. 54,1-12. (doi:10.1144/sjg2017-016)

4. Falkingham PL, Maidment SCR, Lallensack JN, Martin JE, Suan G, Cherns L, Howells C, Barrett PM. 2022 Late Triassic dinosaur tracks from Penarth, South Wales. Geol. Mag. 159,821-832. (doi:10.1017/S0016756821001308)

5. Falkingham PL, Marty D, Richter A. 2016 Dinosaur tracks: the next steps. Bloomington, IN: Indiana University Press.

6. Edgar KM, Haller L, Cashmore DD, Dunne EM, Butler RJ. 2023 Stratigraphic and geographic distribution of dinosaur tracks in the UK. J. Geol. Soc. 180,1-10. (doi:10.1144/jgs2023-003)

7 Sander PM 2023 Sauronods Curr Riol 33 R52-R58 (doHO 1016/i cub 2022 12 0241

/ o Jul lUt^l 1 1V1 * L\J L. j jaUIVUV/Uj. \-Ul f- UlVI* <JJ i WJZ- UJUt ^VJVIt \\J * \\J \ \JI |\UUtLVZ-Z-* \ i-*\J i-\j 8. Mannion PD, Upchurch P, Carrano MT, Barrett PM. 2011 Testing the effect of the rock record on diversity: a multidisciplinary approach to elucidating the generic richness of sauropodomorph dinosaurs through time. Biol. Rev. Camb. Philos. Soc. 86,157-181. (doi:10.1111/j.1469-185X.2010.00139.X)

9. McPhee BW, Mannion PD, de Klerk WJ, Choiniere JN. 2016 High diversity in the sauropod dinosaur fauna of the Lower Cretaceous Kirkwood Formation of South Africa: implications for the Jurassic-Cretaceous transition. Cretac. Res. 59,228-248. (doi:10.1016/j.cretres.2015.11.006)

10. Upchurch P, Martin J. 2003 The anatomy and taxonomy otCetiosauws (Saurischia, Sauropoda) from the Middle Jurassic of England. J. Vert. Paleontol. 23,208-231. (doi:10.1671/0272-4634(2003)23[208:TAATOC]2.0.CO;2)

11. Holwerda FM, Evans M, Listen JJ. 2019 Additional sauropod dinosaur material from the Callovian Oxford Clay Formation, Peterborough, UK: evidence for higher sauropod diversity. PeerJ 7, e6404.(doi:10.7717/peerj.6404)

12. Mannion PD. 2019 A turiasaurian sauropod dinosaur from the Early Cretaceous Wealden Supergroup of the United Kingdom. PeerJ 7, e6348. (doi:10.7717/peerj.6348)

13. Lockley MG, King M, Howe S, Sharp T. 1996 Dinosaur tracks and other archosaur footprints from the Triassic of South Wales. Ichnos. 5,23-41. (doi:10.1080/10420949609386404)

14. Day JJ, Norman DB, Upchurch P, Powell HP. 2002 Dinosaur locomotion from a new trackway. Mure 415,494-495. (doi:10.1038/415494a)

15. Day JJ, Norman DB, Gale AS, Upchurch P, Powell HP. 2004 A Middle Jurassic dinosaur trackway site from Oxfordshire, UK. Palaeontology 47, 319-348. (doi:10.1111/j.0031-0239.2004.00366.x)

16. Day JJ, Upchurch P, Norman DB, Gale AS, Powell HP. 2002 Sauropod trackways, evolution, and behavior. Science 296,1659. (doi:10.1126/ science.1070167)

17. Wright JL. 1998 Keates' Quarry dinosaur footprint site, Intermarine Member, Purbeck Limestone Group (Berriasian), UK. Proc. Dorset. Nat. Hist Archaeol. Soc. 119,185-186.

18. Wright JL, Radley JD, Upchurch P, Wimbledon WA. 1998 Keates' quarry dinosaur footprint site, Intermarine member, Purbeck limestone group (Berriasian), UK. National Trust internal report [unpublished].

19. Ferraby R, Powlesland D. 2019 Heritage and landscape change: recording, archiving and engaging with photogrammetry on the Jurassic Coast World Heritage Site. Proc. Geol. Assoc. 130,483-492. (doi:10.1016/j.pgeola.2019.02.007)

20. Edgar KM, Meade LE, Jones HT, Haller L, Scriven S, Reedman C, Butler RJ. 2023 The condition, use and future of the UK's largest accessible dinosaur tracksite at Spyway Quarry, Dorset. Proc Geol. Assoc. 134,125-138. (doi:10.1016/j.pgeola.2023.01.001)

21. Ensom PC. 2002 Vertebrate trace fossils in the Purbeck Limestone Group of southern England. Spec. Pap. Palaeontol. 68,203-220.

22. Barrett PM, Benson RBJ, Upchurch P. 2010 Dinosaurs of Dorset: Part II, the sauropod dinosaurs (Saurischia, Sauropoda) with additional comments on the theropods. Proc. Dorset. Nat. Hist. Archaeol. Soc. 131,113-126.

23. Clements RG. 1993 Type-section of the Purbeck Limestone Group, Durlston Bay, Swanage, Dorset. Proc. Dorset. Nat. Hist. Archaeol. Soc. 114, 181-206.

24. Westhead RK, Mather AE. 1996 An updated lithostratigraphy for the Purbeck Limestone Group in the Dorset type-area. Proc. Geol. Assoc. 107, 117-128. (doi:10.1016/S0016-7878(96)80005-8)

25. Bournemouth University. 2018 See https://www.bournemouth.ac.uk/news/2018-07-30/dinosaurs-dorset-sauropod-tracks-found-purbecks.

26. Francis JE. 1983 The seasonal environment of the Purbeck (Upper Jurassic) fossil forests. Palaeogeogr. Palaeoclimatol. Palaeoecol. 48,285-307. (doi:10.1016/0031-0182(84)90049-X)

27. Ensom PC. 1995 Dinosaur footprints in the Purbeck Limestone Group (?Upper Jurassic-Lower Cretaceous) of southern England. Proc. Dorset. Nat. Hist Archaeol. Soc. 116,77-104.

28. Ensom PC, Delair JB. 2007 Dinosaur tracks from the lower Purbeck strata of Portland, Dorset, southern England. Geosci. in S.-W. Eng. 11,309- 325.

29. Schneider CA, Rasband WS, Eliceiri KW. 2012 NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9,671-675. (doi:10.1038/nmeth. 2089)

30. Belvedere M, Farlow JO. 2016 A numerical scale for quantifying the quality of preservation of vertebrate tracks. In Dinosaur tracks (life of the past): the next steps (eds PL Falkingham, D Marty, A Richter), pp. 93-98. Bloomington, IN: Indiana University Press.

31. Farlow JO, Bakker RT, Dattilo BF, Everett Deschner E, Falkingham PL, Harter C, Solis R, Temple D, Ward W. 2020 Thunder lizard handstands: Manus-only sauropod trackways from the Glen Rose Formation (Lower Cretaceous, Kendall County, Texas). Ichnos 27,167-199. (doi:10.1080/ 10420940.2019.1698424)

32. Steart DC, Needham J, Strullu-Derrien C, Philippe M, Krieger J, Stevens L, Spencer ART, Hayes PA, Kenrick P. 2023 New evidence of the architecture and affinity of fossil trees from the Jurassic Purbeck Forest of southern England. Bat. Lett. 170,165-182. (doi:10.1080/23818107. 2023.2197973)

33. Scriven S. 2021 The Spyway Dinosaur Footprints project. See https://jurassiccoast.org/support/the-jurassic-coast-trust/what-we-do/ conservation-case-studies/the-spyway-dinosaur-footprints-project/ (accessed 23 June 2022).

34. Henderson DM. 2006 Simulated weathering of dinosaur tracks and the implications for their characterization. Can. J. Earth Sci. 43,691-704. (doi:10.1139/e06-024)

35. Belvedere M, Franceschi M, Morsilli M, Zoccarato PL, Mietto P. 2011 Fish feeding traces from middle Eocene limestones (Gargano Promontory, Apulia, Southern Italy).Palaios 26,693-699. (doi:10.2110/palo.2010.p10-136r)

36. Falkingham PL, Bates KT, Mannion PD. 2012 Temporal and palaeoenvironmental distribution of manus- and pes-dominated sauropod trackways.1 Geol. Soc. London 169,365-370. (doi:10.1144/0016-76492011-019)

37. Carrano MT, Wilson JA. 2001 Taxon distributions and the tetrapod track record. Paleobiology 27,564-582. (doi:10.1666/0094-8373(2001)027< 0564:TDATTT>2.0.CO;2)

38. Wilson JA, Carrano MT. 1999 Titanosaurs and the origin of "wide-gauge" trackways: a biomechanical and systematic perspective on sauropod locomotion. MeoWo/ogy 25,252-267. (doi:10.1017/S0094837300026543)

39. Klinkhamer AJ, Mallison H, Poropat SF, Sloan T, Wroe S. 2019 Comparative three-dimensional moment arm analysis of the sauropod forelimb: implications for the transition to a wide-gauge stance in titanosaurs. Anat. Rec. 302,794-817. (doi:10.1002/ar.23977)

40. Castanera D, Pascual C, Canudo Jl, Hernandez N, Barco JL. 2012 Ethological variations in gauge in sauropod trackways from the Berriasian of Spain, ief/io/fl 45,476-489. (doi:10.1111/j.l 502-3931.2012.00304.x)

41. Royo-Torres R, Cobos A, Alcala L. 2006 A giant European dinosaur and a new sauropod clade. Science 314,1925-1927. (doi:10.1126/science. 1132885)

42. Royo-Torres R, Upchurch P, Mannion PD, Mas R, Cobos A, Gasco F, Alcala L, Sanz JL. 2014 The anatomy, phylogenetic relationships, and stratigraphic position of the Tithonian-Berriasian Spanish sauropod dinosaur Aragosaurus ischiaticus. lool. J. Linn. Soc. 171,623-655. (doi:10. 1111/ZOJ.12144)

43. Perez-Pueyo M, Moreno-Azanza M, Barco JL, Canudo Jl. 2019 New contributions to the phylogenetic position of the sauropod Galvesaurus herreroiimm the late Kimmeridgian-early Tithonian (Jurassic) of Teruel (Spain). Bol. geol. min. 130,375-392. (doi:10.21701/bolgeomin.130.3. 001)

44. Mocho P, Royo-Torres R, Malafaia E, Escaso F, Ortega F. 2016 Systematic review of late Jurassic sauropods from the Museu Geologico collections (Lisboa, Portugal).! Iber. Geol. 42,227-250. (doi:10.5209/rev_JIGE.2016.v42.n2.52177)

45. Taylor MP. IdMXenoposeidon is the earliest known rebbachisaurid sauropod dinosaur. PeerJ 6, e5212. (doi:10.7717/peerj.5212)

46. Myers TS, Fiorillo AR. 2009 Evidence for gregarious behavior and age segregation in sauropod dinosaurs. Palaeogeogr. Palaeoclimatol. Palaeoecol. 274,96-104. (doi:10.1016/j.palaeo.2009.01.002)

47. Castanera D, Barco JL, Diaz-Martinez I, Gascon JH, Perez-Lorente F, Canudo Jl. 2011 New evidence of a herd of titanosauriform sauropods from the lower Berriasian of the Iberian range (Spain). Palaeogeogr. Palaeoclimatol. Palaeoecol. 310,227-237. (doi:10.1016/j.palaeo.2011.07.015)

48. Lockley MG, Meyer CA, Hunt AP, Lucas SG. 1994 The distribution of sauropod tracks and trackmakers. GAIA10,233-248.

49. Mannion PD, Upchurch P. 2010 A quantitative analysis of environmental associations in sauropod dinosaurs. Paleobiology 36,253-282. (doi:10. 1666/08085.1)

50. Alexander RM. 1976 Estimates of speeds of dinosaurs. Wofure 261,129-130. (doi:10.1038/261129a0)

51. Thulborn RA. 1990 Dinosaur tracks. Melbourne, Australia: Chapman & Hall. (doi:10.1007/978-94-009-0409-5)

52. Gonzalez Riga BJ. 2011 Speeds and stance of titanosaur sauropods: analysis of Titanopodus tracks from the Late Cretaceous of Mendoza, Argentina. An. Acad. Bras. Cienc. 83,279-290. (doi:10.1590/s0001-37652011005000002)

53. Tucker ME, Burchette TP. 1977 Triassic dinosaur footprints from South Wales: their context and preservation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 22,195-208. (doi:10.1016/0031-0182(77)90028-1)

54. Evans D. 2023 Discovering more trackways at Ardley. Earth Heritage 58,39-41.

55. Wright JL, Barrett PM, Lockley MG, Cook E. 1998 A review of the Early Cretaceous terrestrial vertebrate track-bearing strata of England and Spain. N. M. Museum of Nat Hist. andSci. Bull. 14,143-153.

56. Edgar KM, Meade LE, Jones HT, Haller L, Scriven S, Reedman C etal. 2023 3D Model-2021 Spyway Quarry, Dorset, UK. Figshare. (doi:10.6084/ m9.figshare.21256107)

57. Butler RJ, Edgar KM, Haller L, Meade LE, Jones HT, Hill 0, Scriven S, Reedman C. 2024 Sauropod dinosaur tracks from the Purbeck Group (Early Cretaceous) of Spyway quarry, Dorset, UK. [dataset]. Zenodo. (doi:10.5281/zenodo.11509629)

58. Butler RJ, Edgar KM, Haller L, Meade LE, Jones HT, Hill 0.2024 Supplementary material from: Sauropod dinosaur tracks from the Purbeck group (Early Cretaceous) of Spyway quarry, Dorset, UK. Figshare. (doi:10.6084/m9.figshare.c.7303155)