Step 2 result
Current disposition after the final public-record check: LF-5 — horizon assignment unresolved. The investigation is parked pending new primary evidence; read the footnoted assessment and access limits. The source-specific interpretation below remains provisional.
The detailed 2025 section supports TH3 as the working interpretation for LF-5, but the published record does not yet establish a reliable cross-publication horizon assignment. The discrepancy also appears within the 2025 paper: its overview plan and detailed photographs do not appear to give the corresponding north-flank patch the same label.
This audit follows the photographic match into the published stratigraphic evidence. It distinguishes an explicit source description from our inference about correspondence between drawings. Review date: October 28, 2026. Record: WS-STRAT-001, attached to unresolved issue WS-SURFACE-001.
The strongest available connection
In 2025 Fig. 13A, a section marker along the trench edge links the groove-bearing patch to Fig. 13B. Panel B identifies LF-5, places the adjacent upper material at TH3, and labels TH2 and TH1 below it; TH4 is exposed farther right. Section 3.2 explicitly describes LF-5 on TH3 above the hardened TH2 surface. [S6 · Bennett et al. (2025)]
That is a useful advance: there is a published plan-to-section link for this particular feature, rather than only a general statement that the site contains several horizons. The section also shows deformation beneath LF-5. The depth reached by that deformation must not be substituted for the identity of the surface on which the groove formed.
The nearby overview photograph, 2025 Fig. 11A, also places LF-5 in the area labeled TH3 and distinguishes the lower TH2 area carrying LF-1–LF-4. These descriptions are related products of the same investigation, not independent replications. [S6 · Bennett et al. (2025)]
The contradiction is broader than one reused photograph
| Evidence | What is actually labeled or stated | Weight in this audit |
|---|---|---|
| 2025 section 3.2; Figs. 11A and 13A–B | LF-5 on TH3; TH2 below | Strongest feature-specific support for the working TH3 interpretation |
| 2025 Fig. 2A overview plan | An irregular north-side patch beside the trench junction is labeled TH2, with an adjoining TH3 polygon | Apparent conflict with the detailed views; patch correspondence is our topological inference |
| 2026 Fig. S1 overview plan | Retains that north-side TH2/TH3 arrangement | Does not resolve the plan/detail conflict; it is a closely related drawing |
| 2026 Fig. S12 | Matched central patch labeled TH2; neighboring patch labeled TH3 | Confirmed photographic disagreement with 2025 Fig. 13A |
| 2026 methods | Numbering is described as continuous with earlier studies | We found no explicit TH2/TH3 renumbering explanation |
For the plan comparison, the identifying landmarks are the main trench, its perpendicular extension on the opposite flank, the irregular north-side exposure outline, and the neighboring TH5 strip. Those features locate a plausible corresponding polygon. This is not a surveyed registration of the schematic plan, so the internal discrepancy is recorded as apparent, not declared a proven drafting error.
A blanket “subtract one from every 2025 label” rule is unsupported. It would also change the TH5 strip that retains its label in the matched photographs. Neither publication year nor a majority vote among related figures decides the correct assignment.
What the other stratigraphic evidence can establish
The 2021 Fig. 2A and 2026 Fig. S6 retain TH2 below TH3 below TH4. The latter’s original-trench column marks linear features at all three horizons, so a generic LF marker cannot uniquely identify LF-5. [S1 · Bennett et al. (2021); S5 · Springer et al. (October 14, 2026)]
The 2026 Fig. S5 middle-reach photograph documents the north-facing wall. The 2025 Fig. 13 caption locates the disputed exposure on the north flank, whose trench-facing wall faces south. Thus the relevant published views are of opposite walls. The lower reach in S5 switches wall orientation, but does not identify LF-5. A continuous, feature-specific connection from Fig. 13B to the 2026 sampled column was not recovered. The wall-orientation comparison is our geometric inference from the captions. [S5 · Springer et al. (October 14, 2026); S6 · Bennett et al. (2025)]
Both papers describe TH2 as notably dolomite-hardened; the 2025 account contrasts LF-5’s excavatable infill. This supports the written distinction, but texture or hardness is not a unique layer identifier. In particular, the 2025 mineralogical percentage reported for Area A is not a chemical measurement of the LF-5 patch. The 2026 Table S1 refers both TH2 and TH3 to Fig. S12, so that table-to-figure link also fails to select a unique patch. [S5 · Springer et al. (October 14, 2026); S6 · Bennett et al. (2025)]
The result is therefore narrower than a corrected chronology: a feature-specific source description has been recovered, while an independently registered connection to the 2026 dated horizon remains missing. Matching elevations or borrowing a modeled midpoint would not repair that missing connection.
Research decision
Use “LF-5 — horizon assignment unresolved” in cross-publication comparisons. Retain “TH3 in the 2025 detailed description and section” as source attribution: it is the strongest feature-specific interpretation within that source, but remains provisional in light of the conflicting plan and 2026 photograph labels.
The canonical 2026 horizon and age remain unassigned. Neither source labels nor published counts have been edited. The photographic-registration result remains valid, and the preservation-adjusted TH2/TH3 comparison remains blocked.
The next discriminating evidence is specific:
- A registered north-flank orthomosaic connecting the LF-5 patch and Fig. 13B section to a named field layer or sampling column, with the relevant contacts traced.
- A documented reconciliation of the north-side polygons in 2025 Fig. 2A / 2026 Fig. S1 against 2025 Fig. 13 / 2026 Fig. S12.
- An explicit statement of whether any corrected assignment affects only figure annotations or also the track inventories and modeled-horizon links.
This completes the audit of the recovered published stratigraphy. It does not claim to have completed field correlation. No author or institution has been contacted.
Step 3 now assesses the consequences: label scope and downstream uses. It extends the review to the neighboring TH4 label and separates unchanged numerical records from unresolved physical-surface assignments.
Source locations and reproducibility
- S6: Bennett et al. 2025, published PDF: Fig. 2A (p. 3), section 3.2 (p. 5), Fig. 11A (p. 12), Fig. 13A–B and caption (p. 14). These pages were rendered and visually inspected.
- S5: Springer et al. 2026, main-text stratigraphy and ichnology methods; supplement: Figs. S1 (PDF p. 18), S5 (p. 22), S6 (p. 23), S12 (p. 28), and Table S1’s TH2/TH3 rows (p. 47), visually inspected.
- S1: Bennett et al. 2021 author manuscript, Fig. 2A, PDF p. 11, visually inspected. The original supplementary PDF remains unavailable in this review.
stratigraphic-audit.json records source hashes, observation locations, inferred links, competing explanations and the unresolved join. Raw source PDFs and figure images are not redistributed in the repository. This audit provides no new sediment assay, surveyed layer trace or direct footprint age.