Some 2.7 Ma after the Permian-Triassic boundary mass extinction (PTME), a stepwise extinction of the nekton (ammonoids and conodonts) ended at the Smithian-Spathian boundary (SSB) during an episode of climate cooling. SSB records from continental shelves are usually affected by an unconformity, suggesting a forced regression of glacio-eustatic origin. Here, we document a new 30 m-thick SSB section from Jebel Aweri (Batain Plain, Oman) that provides an exceptionally complete and expanded record preserved in an exotic block. Most of this SSB section consists of metazoan reefal build-ups that formed in shallow water on an offshore sea mount. In Wadi Musjah (Hawasina nappes, Oman), another exotic block records the SSB in a deeper water setting represented by Hallstatt-type facies. These two sections provide a unique perspective on the early Spathian rapid re-diversification of conodonts. They led to a thorough revision of conodont taxonomy around the SSB and to the construction of the highest resolution biochronological scheme for this time interval in the Tethys. A total of five SSB sections from Oman representing both offshore sea mounts and lower slope deposits were included in a high-resolution, quantitative Unitary Associations analysis. The resulting 8 conodont biozones are intercalibrated with ammonoid zones and with the carbonate carbon isotope record ultimately placing the SSB in the interval of separation between UAZ₃ and UAZ₄. Only the association of Nv. pingdingshanensis with Ic. crassatus can be used to unambiguously characterize the base of the Spathian.

Sedimentological analysis (S1)

The rock samples for the sedimentological analysis were recovered during two individual fieldtrips in 2015 and 2017. The sedimentological analysis was achieved via a qualitative description of the of the lithology by observations in the field and via further analysis of the samples wich were sent to the University of Zurich. The figures for the microstructures were obtained by cutting the rocks into thin sections at the University of Lausanne.

Conodont samples for Unitary Association Method (S2 and S3)

A total of 60 samples was collected for conodonts, 26 from Wadi Musjah and 34 from Jebel Aweri. The spacing between the samples varies between 10–40 cm in the Wadi Musjah section and 0.5–3 m in the Jebel Aweri section. From Wadi Musjah, samples of 1–3 kg of carbonate rock material were dissolved with a 10 % buffered acetic acid solution following the procedure of Jeppsson et al. (1999). In Jebel Aweri, the same procedure was applied to 10–15 kg samples. The residues were treated by heavy liquid separation using sodium polytungstate (Jeppsson & Anehus 1999), then sieved with a 0.075-mm mesh. The heavy fraction was handpicked under a binocular microscope and selected conodont elements were coated with carbon and illustrated using Scanning Electron Microscopy (SEM) (JEOL JSM-6010). Uncoated elements were illustrated with a Keyence VHX 6000 digital microscope.

Five sections from Oman (Wadi Musjah (this study), Jebel Aweri (this study), Jebel Safra (Orchard 1995), Radio Tower (Chen et al. 2019) and Wadi Bani Khalid (Chen et al. 2019)) were included in the analysis. Taxonomic revisions were made to ensure the consistency of the data set using all published material.

Carbon isotope data (S4)

We analyzed the micritic matrix of 50 carbonate rock samples from Wadi Musjah and 61 samples from Jebel Aweri for bulk carbonate carbon isotopes (δ¹³C_carb). The samples were carefully cleaned and cut into slabs in order to recover unaltered micrite. The most homogenous part was drilled on a sawed surface with a diamond-tipped drill to produce a fine powder. All samples were processed at the Institute of Mineralogy and Geochemistry of Lausanne University and measured with a GasBench II connected to a Finnigan MAT DeltaPlus XL mass spectrometer with a He-carrier gas system (Spötl and Vennemann, 2003). Reproducibility (n=3 per run) of replicate analysis was better than 0.1 ‰ for standards and 0.15 ‰ for sediment samples. All reported carbonate carbon isotope results are in the standard δ notation (‰ deviation versus VPDB).

Conodont biochronozones across the Smithian-Spathian boundary (S5)

Conodont biochronozones from Oman and South China calibrated against U-Pb ages. Intercalibrated with ammonoid biochronozones from Oman and China and the δ¹³C_carb records from Jebel Aweri, Wadi Musjah, Qiakong, Laren, Shanggang and Lilong. Timing of Early Triassic after Bayesian age-depth model by Widmann (2019, thesis), Ammonoid zones after Brayard & Bucher (2008) and Brühwiler et al. (2009, 2010, 2012). South China conodont biochronozones after Leu et al. (subm.).

Unitary Association Method (S2 and S3)

The Unitary Associations method (UAM) is a deterministic approach based on the Graph Theory whose goal is to construct a discrete sequence of mutually exclusive maximal sets of coexisting species (Guex 1991; Angiolini & Bucher 1999; Monnet & Bucher 2002; Brosse et al. 2016). We have used the UAM as implemented in the software PAST (Hammer et al. 2001, 2012) using an older version (Version 2.17c, 2013) that allows us to detect and analyse Z4-type contradictions. This software is free and can be downloaded online. (e.g. https://past.en.lo4d.com/windows)