New Discovery of Middle Jurassic Dinosaur Tracks in Chaya County, Eastern Tibet, China

1. Introduction

The terrestrial red-bed basins of southwestern China are important regions for global Jurassic dinosaur research. The Mesozoic strata of Sichuan and Yunnan provinces have yielded abundant dinosaur remains and tracks, forming a complete and continuous evolutionary succession of dinosaur faunas. This succession includes the Early Jurassic Lufengosaurus [1] and Zizhongosaurus faunas [2], the Middle Jurassic Chuanjiesaurus [3,4] and Shunosaurus-Omeisaurus faunas [5,6], and the Late Jurassic Mamenchisaurus fauna [7,8,9], representing characteristic dinosaur faunas of East Asia. In the Jurassic deposits, theropod and sauropod tracks are abundant, providing a solid basis for reconstructing regional dinosaur evolution and paleobiogeographic exchange [1,10,11,12]. The Changdu Basin in eastern Tibet is adjacent to the two major fossil-rich areas of Sichuan and Yunnan. Mesozoic continental clastic rocks are widely exposed in the basin, covering approximately 11,000 km² [13,14,15]. These deposits provide favorable sedimentary conditions for preserving dinosaur fossils. However, fieldwork constraints associated with the high-altitude, cold climate, and complex terrain of the Qinghai–Tibet Plateau have limited studies of dinosaurs from eastern Tibet. Published dinosaur remains and tracks are scarce and scattered, in stark contrast to the more extensive research conducted in Sichuan and Yunnan.

Chao et al. [16,17] were the first to report Jurassic dinosaur remains in the Changdu area. However, the Early and Middle Jurassic species established from these specimens were later regarded as invalid because the original materials were lost and the taxonomic assignments lacked adequate supporting evidence [18,19,20]. In addition, Monkonosaurus lawulacus [22] from the Upper Jurassic–Lower Cretaceous [21] is represented only by sacrals [23] and lacks sufficient diagnostic features. Therefore, it has been regarded as an invalid species [24]. In recent years, researchers have discovered fragmentary sauropod and theropod remains in Chaya County. These fossils are considered potentially related to the contemporaneous dinosaur fauna of southwestern China [18,19]. Dinosaur tracks from the Middle Jurassic are the most abundant. Xing et al. [20,25,26] reported at least eight sauropod tracksites dominated by Brontopodus and one tracksite with a Deltapodus trackway. Yao et al. [27] and Li et al. [28,29,30] focused on theropod footprints and suggested that this assemblage can be assigned to the Grallator–Eubrontes plexus. Chen et al. [31] reported the world’s smallest sauropod footprint, Parabrontopodus isp., as well as a theropod trackway attributable to Eubrontidae. Weise et al. [32] also reported a newly discovered theropod trackway, although it remains unidentified at the genus and species levels and has not been studied in detail. In addition, Early Cretaceous sauropod, ornithopod, and theropod footprints [33], Late Cretaceous ornithopod and sauropod footprints [34], and bird tracks [35] have been reported. Recent studies indicate that eastern Tibet was an important Jurassic dinosaur habitat, yet its record remains poorly documented. Dinosaur remains are rare. Although dinosaur tracks are more abundant, they have not been thoroughly studied. Limited comparisons with coeval dinosaur faunas from adjacent regions limit our understanding of this assemblage. The Duolongwa tracksite described in this study contains one sauropod tracklayer and one newly identified theropod tracklayer from eastern Tibet. Here, we describe the tracksite, provide an ichnological analysis of the tracks, discuss the paleogeographic distribution of tracks in eastern Asia, and infer the trackmakers, their gait, and the depositional paleoenvironment.

2. Geological Background

The Changdu Basin lies within the eastern Tethys orogenic domain, sandwiched between the Jinshajiang tectonic belt and the Lancangjiang tectonic belt. In the Late Triassic, as the Jinshajiang and Lancangjiang Oceans closed sequentially, the Changdu region transitions from marine to continental conditions during the Early Jurassic. The paleogeographic isolation between it and the central Yunnan Basin of Yunnan Province, as well as the Sichuan Basin of Sichuan Province, was removed. This coastal inland basin became a key habitat for the evolution and survival of Jurassic dinosaurs in East Asia [28,29,30,36].

The Duolongwa tracksite is located south of the Changdu Basin (Figure 1A), approximately 4 km southeast of Duolongwa Village, Chaya County, Changdu City (Figure 1B; coordinates: 97°54′35″ E, 30°31′16″ N). The Duolongwa Section is exposed along Provincial Road S203 and comprises the Middle Jurassic Dabuka Formation (Figure 1C-D), which is subdivided into three units. Unit-1 is dominated by purple-red, very thick-bedded massive medium- to fine-grained sandstone, locally interbedded with purple-red thin- to very thin-bedded argillaceous siltstone or silty mudstone (Figure 1E). Unit-2 consists of gray-green, thick-bedded massive medium- to fine-grained sandstone, with occasional interbeds of purple-red or gray-green thin-bedded muddy siltstone or argillaceous siltstone, and rare thin conglomerate layers (Figure 1F). Unit-3 is composed of purple-red, thick- to very thick-bedded medium- to fine-grained sandstone, locally interbedded with purple-red thin- to medium-bedded argillaceous siltstone (Figure 1G).

According to He et al. [37], the sporopollen taxa Classopollis and Psophosphaer reported in the Dabuka Formation are characteristic taxa of the Early and Middle Jurassic. In addition, the Dabuka Formation has yielded several common Middle Jurassic freshwater bivalves from western Yunnan, including Lamprotula (Eolamprotula) cremeri, Pseudocardinia kweichouensis, Pse. angulate, Pse. cf. khadjakalanensis, Unio yunnanensis, Psilunio ovalis, Psi. lufengensis, Psi. thailandicus, Psi. cf. Yuanmouensis, Cuneopsis johannisboehmi, and C. sichuanensis. Therefore, the geological age of the Dabuka Formation is generally considered to be Middle Jurassic [17,38].

3. Materials and Methods

The Duolongwa Section was exposed by road-cut slope excavation. The tracklayers dip at approximately 80°, and two tracklayers are present (Figure 1C-D). The tracks were discovered following erosion of the overlying muddy siltstone, and all tracks are preserved in situ as concave epireliefs. The DLW1 tracklayer covers approximately 14 m² and is located 8.3 m above the base of Unit-3; 12 theropod tracks are visible (Figure 2). The DLW2 tracklayer covers approximately 13 m² and is located 1.5 m below the top of Unit-3; two sauropod tracks are visible (Figure 3).

After numbering all tracks, approximately 40 photographs were taken of each track from different directions. These photographs were imported into Agisoft Metashape Professional (64-bit) to generate a 3D height image. Interpretive outline drawings were then drawn in CorelDRAW 2017.

Following Zhen et al. [40], we created a schematic diagram for measuring track data. According to the Thulborn’s [41] criteria, the measured parameters include pedal track length (FL) and pedal track width (FW). For theropod tracks, additional measurements include the interdigit divarication angles between digits Ⅱ and Ⅲ (α), the interdigit divarication angles between digits Ⅲ and Ⅳ (β), and the interdigit divarication angles between digits Ⅱ and Ⅳ (α + β). Trackway measurements for theropods include pace length (PL), stride length (SL), pedal track pace angulation (ANG), and rotation of pedal track relative to the mid-line of trackway (γ; “-” means the imprint is rotated inward; otherwise it is rotated outward). According to the methods proposed by Olsen et al. [42] and Lockley et al. [43], we calculated the mesaxony of digit Ⅲ (a/b) and the projection ratio (X/Y) of the theropod tracks. Following Xing et al. [44] and Henderson et al. [45], the hip height (H) and body length (L) of the theropod trackmakers were estimated using the equations H = 4.9FL (Eq. 1) and L = 2.63H (Eq. 2), respectively. The body size of sauropod trackmakers was estimated using the equations proposed by Alexander et al. [46] and Farlow et al. [47], where H = 5.3FL (Eq. 3) and L = 3.7H (Eq. 4).

This study also discusses the gait and speed (V) of the trackmakers. Thulborn et al. [41] modified the empirical formula of Alexander et al. [46] and suggested that when SL/H ≤ 2 (Eq. 5), the trackmaker was walking slowly, and V = 0.25g^0.5SL^1.67H^-1.17 (Eq. 6). When 2 < SL/H < 2.9 (Eq. 7), the trackmaker was moving at a trotting gait, and when SL/H ≥ 2.9 (Eq. 8), the trackmaker was running quickly, with V = [gH(SL/1.8H)^2.56]^0.5 (Eq. 9). Here, g = 9.8 m/s².

4. Morphological Paleontology

4.1. Eubrontidae igen. et isp. indet. Morphotype-1

Description. A total of 11 pedal tracks are preserved in situ as concave epireliefs on the DLW1 tracklayer. They were left by a bipedal, tridactyl, digitigrade trackmaker; no manual or tail tracks are observed (Figure 2A). T1 Trackway (T1) consists of three tracks (right pedal track 1-1-1, left pedal track 1-1-2, and right pedal track 1-1-3). SL1-1-1 is 206.1 cm, PL1-1-1 is 107.5 cm, PL1-1-2 is 101.0 cm, ANG1-1-2 is 165°, and γ1-1-2 is -2°. 1-1-1 and 1-1-3 are poorly preserved, and the digit impressions are difficult to discern. 1-1-2 (Figure 2B) is the best preserved. It has an FL of 29.6 cm and an FW of 29.8 cm, with α (40°) > β (36°). Digit III is the longest (DLIII = 19.2 cm, DWIII = 9.1 cm), followed by digit II (DLII = 11.2 cm, DWII = 6.1 cm) and digit IV (DLIV = 11.1 cm, DWIV = 6.9 cm). The digits are robust, but the phalangeal pads are poorly preserved. The distal phalangeal pads of digit III are rounded, and claw marks are visible at the distal ends of the digits, with the claw mark on digit III being the largest. A rounded metatarsophalangeal pad is present at the posterior end of the track, with a distinct concavity between it and digit II. T2 Trackway (T2) consists of three tracks (right pedal track 1-2-1, left pedal track 1-2-2, and right pedal track 1-2-3). SL1-2-1 is 182.6 cm, the mean PL is 91.4 cm, and ANG1-2-2 and γ1-2-2 are 175° and -3°, respectively. All three tracks are poorly preserved, and the digit impressions are difficult to discern; therefore, only FL and FW were measured, with mean values of 20.1 cm and 21.2 cm, respectively. Left pedal track 1-3-1 and right pedal track 1-3-2 are two successive tracks (PL1-3-1 = 80.3 cm), but 1-3-1 is poorly preserved, with indistinct phalangeal pads and digit impressions. 1-3-2 (Figure 2C) is relatively well preserved, with FL and FW of 36.2 cm and 27.1 cm, respectively, and α and β of 24° and 27°. Digit III is the longest (DLIII = 24.1 cm, DWIII = 6.3 cm), with three well-preserved rounded phalangeal pads. Digit II is relatively robust (DLII = 19.2 cm, DWII = 9.0 cm), and digit IV shows a distal rounded phalangeal pad (DLIV = 10.1 cm, DWIV = 5.2 cm). Sharp claw marks are visible at the distal ends of the digits. A distinctly protruding, small metatarsophalangeal pad is present at the posterior end of the track. Tracks 1-4-1 and 1-4-2 are successive (PL1-4-1 = 114.3 cm), and both are poorly preserved; their mean FL and FW are 32.9 cm and 26.4 cm, respectively. 1-5-1 and 1-6-1 are also poorly preserved. Track 1-5-1 is infilled with purplish-red silty mudstone, with FL and FW of 33.2 cm and 21.3 cm, respectively, whereas 1-6-1 cannot be measured accurately (Table 1).

Remarks. Based on their ichnomorphological features, these theropod tracks can be divided into three types. The type represented byT2 is medium-sized theropod tracks, with FL ranging from 18.2 to 22.1 cm. Their FLs are slightly smaller than the FW, with a mean FL/FW ratio of 0.96, and the phalangeal pads are relatively robust. Because T2 is poorly preserved, the digit impressions are indistinct, and the phalangeal pads, metatarso-phalangeal pads, and claw traces are unclear. Only the 1-2-1 and 1-2-2 provide rough estimates of α+β, X/Y, and a/b values (Table 1); therefore, its ichnotaxonomic assignment remains uncertain.

The type represented by T1 shows some ichnomorphological features of the theropod ichnogenus Corpulentapus, first reported by Li et al. [48] from the Lower Cretaceous of Zhucheng, Shandong Province. According to Li et al. [48], Corpulentapus is characterized as a bipedal, digitigrade, tridactyl track with robust phalangeal pads, subequal track length and width (FL = 9.5–28.5 cm), broad-digit imprints, and a lily-shaped ichnomorphology. These ichnomorphological features are broadly comparable to those of the tracks of T1 (Figure 2B). However, compared with Corpulentapus described by Li et al. [48], the tracks of T1 have a weak mesaxony of digit Ⅲ and a larger projection ratio. For example, the a/b values calculated in this study for the holotype of Corpulentapus (track no. 214.174) and the 1-1-2 track of T1 are 2.33 and 1.40, respectively, whereas their X/Y ratios are 0.32 and 0.39, respectively. Therefore, T1 is more consistent with the ichnofamily Eubrontidae. According to Lull [49,50], Eubrontidae make large, bipedal, tridactyl tracks with FL generally exceeding 25 cm. The tracks show broad digits, sharply pointed or blunt claw marks, and there is no trace of the tail or manus. The ichnofamily includes six ichnogenera: Asianopodus (Matsukawa et al. [51]), Changpeipus (Young [52]), Chapus (Li Jianjun et al. [53]), Eubrontes (Hitchcock et al. [54]), Hunanpus (Zeng Xiangyuan [55]), and Lufengopus (Lü et al. [56]). These ichnogenera are characterized by distinctly rounded phalangeal pads. Asianopodus differs from T1 in having a clear boundary between the metatarso-phalangeal pads and the digital impressions (Matsukawa et al. [51]). Changpeipus differs in exhibiting a distinctly rhomboid ichnomorphology (Young [52]). Chapus differs by its relatively large digit Ⅲ claw mark (Li et al. [53]). Hunanpus differs by the presence of distal metatarso-phalangeal pads (Zeng [55]), and Lufengopus differs by its broad, rounded heel region (Lü et al. [56]). T1, with its well-defined phalangeal pads, broad digit imprints, and blunt claw traces, is most consistent with Eubrontes. However, the digits of T1 are more robust than those of typical Eubrontes, possibly reflecting a higher moisture content in the substrate at the time of track formation, although not sufficiently high to cause lateral collapse of the sediment due to gravity. A similar preservational condition was reported by Li et al. [57] for Eubrontes glenrosensis from the Jurassic of Wulatezhongqi, Neimenggu Province, which likewise shows notably robust digit impressions.

The type represented by 1-3-1 and 1-3-2 are poorly preserved in the metatarso-phalangeal pads, and their digit imprints are indistinct. Nevertheless, they retain some ichnomorphological features of Eubrontes, such as sharply pointed claw marks and broad, rounded phalangeal pads. Compared with T1, they have larger FL values than FW values and relatively slender digit imprints. This difference may reflect higher substrate moisture content during T1 formation, leading to deeper foot penetration, more robust impressions, and a larger FW.

T2 cannot be confidently interpreted as any ichnogenus. Although T1, 1-3-1, and 1-3-2 show some ichnomorphological features of Eubrontes, their poor preservation precludes a direct assignment to Eubrontes. Accordingly, they are here assigned to Eubrontidae igen. et isp. indet.

4.2. Brontopodus isp. (Indeterminate Species), Morphotype-2

Description. Two tracks are preserved in situ on the DLW2 tracklayer (Figure 3A). 2-1 (Figure 3B) is a concave epirelief track. It is elliptical in outline and filled with purplish-red fine sandstone. An expulsion rim is visible. Its FL and FW are 55.8 cm and 45.5 cm, respectively. The digit impressions are indistinct, making it impossible to determine whether it represents a left or right pedal track. 2-2 is a convex relief. It is elliptical in outline, with indistinct digit impressions, and it is difficult to distinguish whether it is a manus or pedal track. Only FL and FW were measured for this track (Table 1).

Remarks. The Jurassic sauropod ichnogenera from eastern Asia include Liujianpus [58], Parabrontopodus [59], and Brontopodus [60]. Brontopodus is readily distinguished from Liujianpus and Parabrontopodus by its wide-gauge trackway, whereas the latter two typically exhibit narrow-gauge trackways [58,59]. Reports of Middle Jurassic sauropod tracks from East Asia remain scarce and are concentrated in southwestern Yunnan and eastern Tibet, China.

Xing et al. [61] described a sauropod right pedal track in the Middle Jurassic Chuanjie Formation in Lufeng County, Yunnan Province. It is elliptical, 71.5 cm long, and 54.5 cm wide. They noted indistinct digit impressions, incomplete metatarso-phalangeal pads, and a rounded heel region. Because the trackway information is unclear, they tentatively assigned it to Brontopodus. Similarly, Xing et al. [62] reported large sauropod tracks from the nearby Hemenkou area of Shuangbai County, Yunnan Province. These tracks are characterized by elliptical pedal tracks, elliptical to reniform manus tracks, and medium- to wide-gauge trackways. They were assigned to Brontopodus.

A newly discovered and important Middle Jurassic sauropod track-bearing region was reported in eastern Tibet. Xing et al. [21] first reported Brontopodus from Morong Village, Karuo County, where the manus and pedal tracks are elliptical, and the trackway is wide. Later, Xing et al. [25] documented three additional sauropod tracksites in Karuo County. According to their description, these specimens were grouped into large, medium, and small types, but their ichnomorphological features are consistent with those from the Morong tracksite. Therefore, they were also assigned to Brontopodus. More recently, Xing et al. [26] reported five tracksites in Chaya County. The Dolongwa Tracksite I appears to correspond to the DLW2 tracklayer investigated in this study. However, during our re-examination, the tracklayer had been damaged, and the tracks illustrated here as 2-1 and 2-2 represent newly exposed specimens. According to the original description, the manus and pedal tracks at Dolongwa Tracksite I are both elliptical; the pedal tracks have a length of 32.4 cm to 71.5 cm, and the trackway is wide. We therefore follow the interpretation of Xing et al. [26] and tentatively assign Morphotype-2 to Brontopodus isp.

5. Discussion

5.1. Trackmaker and Gait

5.1.1. Morphotype-1

Based on Equations 1 and 2, the T1 trackmaker is estimated to have had an H of 1.49 m and an L of 3.39 m. Its mean PL and SL are 104.3 cm and 206.1 cm, respectively. The results from Equations 5 and 6 indicate that the trackmaker moved at a slow walking speed (SL/H ≈ 1.37 < 2), with an estimated V of 0.53 m/s. The pedal tracks are slightly rotated inward relative to the midline of the trackway (γ = -2°), and the ANG is 165°.

The T2 trackmaker is estimated to have had an H of 0.98 m and an L of 2.59 m. Calculations based on Equations 5 and 6 suggest it moved at a slow trot (SL/H ≈ 1.85 < 2), with an estimated V of 2.18 m/s, substantially faster than the T1 trackmaker. Similar to T1, the pes impressions are slightly rotated inward relative to the mid-line of trackway (γ = -3°); however, the ANG is 175°, which is considerably larger than that of T1, likely reflecting the higher speed of the T2 trackmaker.

Morphotype 1 exhibits several ichnomorphological features consistent with Eubrontes, but its classification remains uncertain. Most previous studies have supported a theropod affinity for Eubrontes trackmakers [25,42,61,63,64]. Although Eubrontes tracks have been widely reported in China, their trackmakers have rarely been discussed in detail [65]. Xing et al. [65] proposed, based on dinosaur remains, that the trackmaker of Eubrontes zigongensis was Sinosaurus triassicus [66], an averostran. However, the tetradactyl footprints of E. zigongensis are distinct from other Eubrontes specimens, including Morphotype-1 described in this study [36]. More recently, An et al. [18] reported a pair of pubic bones and associated limb remains of a small theropod from Chayu County and suggested a possible affinity with Shidaisaurus jinae or Monolophosaurus. However, Li et al. [30] argued that the specimen is substantially smaller and suggested that the most likely trackmaker of the Middle Jurassic Eubrontes from eastern Tibet is Gasosaurus constructus, which was discovered in the adjacent Sichuan Basin and originally described by Dong and Tang (1984) [67]. It had an estimated L of 3.5-4 m. Based on Equations 1 and 2, the estimated L and H of the Morphotype-1 trackmaker are 2.56-4.67 m and 0.98-1.77 m, respectively, values that broadly overlap with the inferred size range of G. constructus. Among theropods currently known from southwestern China, G. constructus remains the only taxon whose body size is consistent with the Morphotype-1 trackmaker. Although the assignment cannot be confirmed unequivocally based on the evidence, we tentatively follow Li et al. [30] in considering Gasosaurus constructus the most likely trackmaker of Morphotype-1.

5.1.2. Morphotype-2

Based on the FL of 2-1, the trackmaker is estimated to have had an H of 2.96 m and an L of 10.94 m, according to Equations 3 and 4. This inferred L is generally smaller than that of the common Middle Jurassic sauropods from southwestern China, including Shunosaurus [68] (a maximum L of 12 m), Omeisaurus [69] (a maximum L of 20 m), and Mamenchisaurus [70] (a maximum L of 26 m).

We follow the view of Xing et al. [26] and tentatively assign Morphotype-2 to Brontopodus. Although the Middle Jurassic Cetiosauridae and Brachiosauridae remains from eastern Tibet reported by Chao [17] have been lost and are now regarded as nomina invalida, An et al. [18,71] noted that these specimens may have some affinity with sauropods from the Sichuan and Yunnan provinces. Xing et al. [25] suggested that the trackmakers of Early to Middle Jurassic Brontopodus in southwestern China may be taxa associated with the Shunosaurus-Mamenchisaurus fauna, whereas Late Jurassic Brontopodus trackmakers are more likely to be Titanosauriformes. Given the discovery of Brontopodus in the Middle Jurassic Chaya Group of eastern Tibet, together with the ichnomorphological similarity between Morphotype-2 and previously described sauropod tracks from eastern Tibet, we follow Xing et al. [26] in interpreting the trackmaker of Morphotype-2 as a member of the Shunosaurus-Mamenchisaurus fauna. In addition, subadult Shunosaurus individuals from the Sichuan Basin reported by Zhang et al. [72] and Peng et al. [73], with an L of approximately 6 m provide an osteological reference for this inference, suggesting that the Morphotype-2 trackmaker may have been a juvenile or subadult of the Shunosaurus-Mamenchisaurus fauna.

5.2. Sedimentological and Paleoenvironmental Analysis

Unit-1 consists of purplish-red, very thick- to thick-bedded medium- to fine-grained sandstone, which commonly pinches out in a wedge (Figure 4A). Climbing-ripple cross-lamination occurs locally within the sandstone, and trace fossils are occasionally observed on oblique bedding planes. The sandstone bedding surfaces are characterized by abundant combined-ripple marks, expressed as sinuous asymmetrical ridge crests (Figure 4B). These features indicate that the depositional interface was influenced by unidirectional currents and oscillatory wave action, suggesting deposition in a fluvial point-bar environment [30,74,75,76,77] Locally, purplish-red thin-bedded silty mudstone interbeds are present. These interbeds are poorly laminated, and polygonal mud cracks occur on bedding surfaces, indicating formation in a low-energy stagnant-water setting subjected to periodic subaerial exposure. They are interpreted as fine-grained suspended sediment deposited by floodwaters in low-lying floodplain depressions and subsequently exposed, evaporated, and desiccated, representing floodplain microfacies [75,78].

Unit-2 is composed of grayish-green, thick-bedded, massive medium sandstone. Parallel bedding and trough cross-bedding are observed, and it locally occurs in lenticular bodies. Thin beds of conglomerate are intercalated within the sandstone. The clasts show moderate to good sorting. They are subround to round, have a preferred orientation, and an erosional scour surface is visible at the base of the conglomerate layer (Figure 4C). This assemblage of sedimentary features, especially the basal scour surface and the upward-fining grain-size trend, indicates deposition in a fluvial channel subfacies [79,80,81].

Unit-3 consists of unevenly interbedded purplish-red thick- to very thick-bedded medium- to fine-grained sandstone and thin- to medium-bedded argillaceous siltstone and silty mudstone. Large trough cross-bedding is observed in the sandstone beds (Figure 4D), and individual beds commonly pinch out laterally in a wedge (Figure 4E). Climbing-ripple lamination (Figure 4F) and parallel bedding (Figure 4G) are well developed. Mud cracks occur in the argillaceous interbeds (Figure 4H), and trace fossils are present on oblique bedding planes (Figure 4I). The sedimentary structures in this assemblage broadly match those of Layer 1. Together, they form a typical vertical alternation of point-bar (lateral accretion bar) and floodplain deposits in a meandering river system, reflecting the sedimentary processes of channel lateral migration and periodic flooding of the floodplain [30,75,80,81].

At the bedding surface of the DLW1 tracklayer, mud cracks cut across dinosaur tracks (Figure 2B), indicating that the area was exposed and desicated after early dinosaurs walked on a moist floodplain surface. The mud cracks formed later and modified the footprints. In addition to the track-cutting mud cracks, the bedding surface of the WDL2 tracklayer also exhibits straight asymmetrical current ripples (Figure 3A). The ripple crest lines extend in the same direction, reflecting a relatively deep-water, low-velocity unidirectional flow regime. The ripples are overlain by later tracks, indicating they formed before the mud cracks. Differences in sedimentary-structure assemblages between these two tracklayers reflect dynamic changes in water depth and exposure within the same floodplain environment at different times, but both are generally products of a floodplain microfacies setting.

5.3. Paleogeographic Analysis of the Track Sites

Chinese reports of Middle Jurassic dinosaur tracks are abundant, with most track localities concentrated in Yunnan, Sichuan, and Shanxi Provinces. Eastern Tibet has also yielded a wealth of research results in recent years (Figure 5; Table A1). Sauropod track reports are relatively scarce for two reasons. First, dinosaur trackmaking [82]: sauropods were less active than theropods, resulting in fewer tracks. Second, sauropod tracks are less rounded than the tridactyl tracks of theropods and ornithopods. Therefore, they have attracted less attention. Among Middle Jurassic sauropod tracks from southwestern China, reports of Brontopodus mainly come from Lufeng [61] and Shuangbai [62] in Yunnan Province, as well as Karuo County [20,25,29] and Chaya County [26] in eastern Tibet. The theropod track assemblage is broadly consistent with the global pattern. Li et al. [36] suggested that this theropod track assemblage can be assigned to the Grallator–Anchisauripus–Eubrontes plexus; Kayentapus is also a common ichnogenus. Among these, reports related to Middle Jurassic Eubrontes are the most abundant, with records from Karuo County in eastern Tibet [30,31], Lufeng [61] and Shuangbai regions [62] in Yunnan Province, Zizhong County in Sichuan Province [83,84], and Zizhou County in Shanxi [85].

The newly discovered sauropod tracks from the DLW2 tracklayer reported in this study are newly exposed tracks from the Brontopodus tracklayer reported by Xing et al [26]. They suggest that these tracks are associated with the Shunosaurus–Mamenchisaurus fauna in the Sichuan and Yunnan provinces.

As for theropod tracks, Li et al. [28] proposed that the Grallator anietangensis trackmaker from eastern Tibet may have been related to Shidaisaurus jinae, a coeval theropod from the Sichuan Basin. In a subsequent study, Li et al. [30,36] suggested that the trackmaker of the Eubrontes tracks from eastern Tibet may have been related to Gasosaurus constructus from the Sichuan Basin.

Chen [86] was the first to propose that dinosaur faunas in southwestern China may have interacted, arguing that the Paleo-Dianzhong Lake in Yunnan and the Paleo-Bashu Lake in Sichuan provided hydrological conditions to facilitate interactions. Fang et al. [15] suggested that paleo-river systems such as the Paleo-Lancang River served as important corridors for dinosaur faunal exchange between Yunnan and eastern Tibet. Xing et al. [26] proposed that intermontane lowlands and tidal flats between the Sichuan Basin and the Qiangtang Block may have provided opportunities for the migration and exchange of dinosaur fauna. The newly discovered theropod and sauropod tracks from this study provide additional evidence for potential dinosaur faunal exchange between eastern Tibet and adjacent regions. However, it must be emphasized that track assemblage similarities can only serve as an important reference for reconstructing the paleogeographic distribution of dinosaur faunas. Similarities may also result from parallel evolution and do not necessarily indicate that the same trackmakers made the tracks.

6. Conclusions

(1) The Duolongwa tracksite in the Middle Jurassic Dabuka Formation in Chaya County, Changdu Basin, eastern Tibet, comprises two tracklayers. DLW1 tracklayer has twelve theropod tracks (Morphotype-1), which were assigned to Eubrontidae ichnogen. et ichnosp. indet. DLW2 tracklayer has two sauropod tracks (Morphotype-2), which were likely made by Brontopodus isp. This finding provides new specimens for the Middle Jurassic dinosaur track assemblage in eastern Tibet.

(2) The trackmaker of Morphotype-1 had a hip height of 0.98-1.77 m and a body length of 2.56-4.67 m. The trackmakers walked (V = 0.53 m/s), while some individuals approached a trot (V = 2.18 m/s). Combined with coeval dinosaur remains from West China, Gasosaurus constructus is inferred as the most probable trackmaker of Morphotype-1. The trackmaker of Morphotype-2 possessed an estimated hip height of 2.96 m and a body length of 10.94 m. It was interpreted as a juvenile belonging to the Shunosaurus-Mamenchisaurus fauna.

(3) The paleoenvironment of the Duolongwa tracksite was a fluvial depositional system, consisting of channel and overbank subfacies. Dinosaurs inhabited the exposed overbank surfaces following flood events. After track formation, the surfaces underwent subaerial desiccation, and mud cracks cut the tracks, indicating subaerial exposure within a fluvial setting.

(4) The Dolongwa tracksite suggests a close faunal connection between the Middle Jurassic dinosaur fauna of eastern Tibet and contemporaneous dinosaur faunas in the Sichuan and Yunnan provinces, supporting the existence of migration corridors for dinosaur dispersal between these regions. However, because similarities in ichnomorphology cannot rule out parallel evolution, more dinosaur remains are needed to substantiate exchange and interaction among Jurassic dinosaur faunas in southwest China.