Ammonites and other fossils were sampled wherever found in situ, and tuffaceous samples were collected throughout the section (full details of horizons and determinations are given in the supplementary data).
Biostratigraphic determination of the Chacay Melehue section confirms the presence of deposits of Late Pliensbachian through earliest Toarcian age (Fig. 2). This section was previously studied for geochronology 3739. Sample 2296R collected at 17.34 m in the Chacay Melehue section (see supplementary Fig. 1) and being located within the tenuicostatum zone ~6 m above the Pliensbachian–Toarcian boundary, was analysed by Riccardi & Kamo 38 This sample has a mean age of 183.11±0.12 and a Bayesian eruption age estimate 39 of 183.11 ± 0.28 Ma.
Here, we have analysed 4 additional samples CM-ASH-1, 3, 5 and 6 from within the same section. Data is corrected to the EARTHTIME tracer ET535, based on U-Pb CA-ID-TIMS analyses of individually abraded zircon crystals (see supplementary information section for details on the methodology).
CM-ASH-1 at 8.69 m, has an estimated depositional age of 184.10 ± 0.54 Ma and is in the latest Pliensbachian disciforme Andean ammonite zone, equivalent to the latest margaritatus–spinatum northwest European ammonite zones 40–42. The bivalve Kolymonectes weaveri Damborenea is also present here from 0.50–12.74 m in the section and has an established stratigraphic range from the Late Pliensbachian through the Early Toarcian 43. CM-ASH-1 occurs ~5 m below the Lowest Occurrence (LO) of the nannofossil Lotharingius hauffii Grün & Zwili (LO 13.55 m), which has an age range from the Late Pliensbachian, NJ5a subzone to the Callovian, NJ12a subzone44.
CM-ASH-3, at 19.24 m, gives a mean age of 182.84 ± 0.1 Ma and a Bayesian eruption age estimate of 183.66 ± 0.21 Ma CM-ASH 3 is located above the LO of Lotharingius barozii Noël (at 17.34 m, 2296R occurs at the same level). Lotharingius barozii Noël is characteristic of the latest Pliensbachian to earliest Toarcian disciforme–tenuicostatum Andean ammonite zones 40,45,46 as well as above the LO of Dactylioceras (Eodactylites) cf. simplex (Fucini) (LO 11.08 m) indicative of the early Toarcian tenuicostatum Zone 47. Thebase of theAndean hoelderi Zone is identified in the section at 21.66 m and is marked by the LO of Harpoceras serpentinum (Schlotheim), Cleviceras exaratum (Young & Bird) and Hildaites cf. murleyi (Moxon). The Andean hoelderi Zone is considered approximately equivalent to the serpentinum (= falciferum) ammonite Zone of northwestern Europe 42,45.
At the Global Stratotype Section and Point (GSSP) for the base of the Toarcian at Peniche (Portugal), the LO of Lotharingius barozii is in sediments of the uppermost emaciatum ammonite Zone, ~50 cm below the base of the Toarcian Stage. Furthermore, the Pliensbachian–Toarcian boundary at this locality is marked by the lowest occurrence of Dactylioceras (Eodactylites) simplex, which is considered to allow global correlation of this level, thereby providing strong support for the proposition that the geochronology in this part of the Chacay Melehue section constrains the age of the boundary. CM-ASH-5 did not yield an interpretable age and was largely comprised of inherited or reworked zircons.
The final ash dated in this study, CM-ASH-6 at 25.02 m, gives a weighted mean U-Pb date of 182.84 ± 0.13 Ma, and a Bayesian eruption age estimate of 183.66 ± 0.33 Ma 39 and is located ~4.5 m above the LO of Harpoceras serpentinum (Schlotheim), Cleviceras exaratum (Young & Bird) and Hildaites cf. murleyi (Moxon) (LO 21.66 m) within the Andean hoelderi Zone, equivalent to the serpentinum (= falciferum) ammonite Zone of northwestern Europe 42,45,47.
Leanza et al. 37 also sampled and analyzed two ash beds in the Chacay Melehue locality using U-Pb ID-TIMS: one of the ashes, at ~24 m in the section, yielded an age of 185.7± 0.40 Ma: this bed is located biostratigraphically above the Pliensbachian–Toarcian boundary, is cross-bedded, and has a very wide array of zircon ages within the zircon population. Consequently, it appears likely that the bed is largely made up of reworked volcaniclastic material, despite the tightly clustered age ranges of the youngest zircons that contribute to this precise date, but probably do not give an accurate depositional age. A second ash bed was dated by Leanza et al. 37, which produced an age of 182.3 ± 0.4 Ma; its exact stratigraphic position within the succession is, however, unknown with respect to our measured section. Field photographs in Leanza et al. 37 could not be matched to the outcrop at the times of our field investigations.
To improve constraints on the age of the Pliensbachian–Toarcian boundary and the age of the lower Toarcian tenuicostatum–hoelderi boundary we used a Bayesian Markov chain Monte Carlo (MCMC) model in which stratigraphic superposition is imposed on U-Pb zircon dates 48. The result is an age–depth model incorporating dates from all beds above and below each sample to produce an internally consistent age model (Fig. 3. B & C.). This model allowed us to extrapolate ages at specific depths, assuming relatively constant sedimentation rates of the deposits between the ash beds that provide the geochronological constraints (Fig 3C). To determine the age of the Pliensbachian–Toarcian boundary, we assessed the stratigraphic position of the boundary to be at 11.08 m in the section, concurrent with the LO of Dactylioceras (Eodactylites), and interpolated the age to be 183.71 +0.40/-0.51 Ma (Fig. 3 C). A similar exercise was performed for the tenuicostatum–hoelderi boundary (concurrent with the tenuicostatum–serpentinum zone boundary in NW Europe), using the LO of Harpoceras serpentinum (Schlotheim), Cleviceras exaratum (Young & Bird) and Hildaites cf. murleyi (Moxon) (LO 21.66 m). Thus, at 21.66 m in the section an age of 182.77 +0.11/ -0.21 Ma was interpolated from the model (Fig. 3C).
The age–depth model coupled with biostratigraphy provides a new more precise age for two of the major events in the earliest Toarcian as well as a new age for the Pliensbachian–Toarcian boundary.