Abstract
Major climatic and ecological changes are documented in terrestrial ecosystems during the Miocene epoch. The Rhinocerotidae are a very interesting clade to investigate the impact of these changes on ecology, as they are abundant and diverse in the fossil record throughout the Miocene. Here, we explored the spatio-temporal evolution of rhinocerotids’ paleoecology during the early and middle Miocene of Europe and Pakistan. We studied the dental texture microwear (proxy for diet) and enamel hypoplasia (stress indicator) of 19 species belonging to four sub-tribes and an unnamed clade of Rhinocerotidae, and coming from nine Eurasian localities ranging from Mammal Neogene zone (MN) 2 to MN7/8. Our results suggest clear differences in the feeding ecology and thus niche partitioning at Kumbi 4 (MN2, Pakistan), Sansan (MN6, France), and Villefranche d’Astarac (MN7/8, France), while overlap of the interpreted diets and subtle variations are discussed for Béon 1 (MN4, France) and Gračanica (MN5/6, Bosnia-Herzegovina). All rhinocerotids studied were interpreted as browsers or mixed-feeders, and none had a grazer nor frugivore diet. The prevalence of hypoplasia was moderate (∼ 10 %) to high (> 20 %) at all localities but Kumbi 4 (∼ 6%), and documented quite well the local conditions. For instance, the high prevalence at the close to Miocene Climatic Optimum locality of Béon 1 (∼26 %) has been correlated with periodical droughts, while the moderate ones (∼ 10 %) at Sansan and Devínska Nová Ves Spalte (Slovakia) both dated from the MN6 (i.e., by the middle Miocene Climatic Transition, ca. 13.9 Mya) were linked to the persistence of sub-tropical local conditions. Besides locality, species and tooth locus were also important factors of variation for the prevalence of hypoplasia. The very large hippo-like Brachypotherium brachypus was one of the most affected species at all concerned localities (but Sansan), whereas early-diverging elasmotheriines were very little affected, suggesting an influence of phylogeny and/or diet in stress susceptibility.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
This version provides corrections according to the comments of the PCI recommender A. Houssaye. It differs from the previous in the following aspects: - minor changes to correct typos, make precisions (notably for stats and to stress the use of extant datasets and thresholds for our paleoecological inferences), and to ensure consistency of abbreviations (HAsfc, supplementary Sx) throughout the manuscript - more examples of values for HAsfc cited in the results - Table 1 modified to combined the taxonomic details from v1 and the effectives by method from v2 - Fig 5 and 6 modified (black outlines) to make colors easier to differenciate - Table 6 taxonomic data (tribes) added
