the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Midlatitude cooling across the Oligocene-Miocene transition
Abstract. The transition from the Oligocene into the Miocene is marked by a significant positive benthic δ18O excursion, the Mi-1, which is interpreted to reflect a transient period of Antarctic ice sheet growth. Despite an increasing number of orbitally resolved benthic records reflecting deep ocean conditions, few comparable high-resolution records exist to evaluate whether this event also entailed changes in surface ocean temperatures either regionally in the Southern Ocean or globally. Here, we aim to evaluate the timing and amplitude of changes in surface ocean temperatures. We present new alkenone undersaturation temperatures from a site in the Southern Ocean on the Tasman Rise (ODP Site 1168), as well as from a site in the North Atlantic on the Newfoundland Margin (IODP Site U1406). Our results show a nearly 4 °C cooling in the Southern Ocean between 24 and 23 Ma in two steps. When benthic δ18O recovers to the lower values typical of the pre- Mi-1 excursion, there is only limited warming in average temperature between 23 and 22 Ma. In the North Atlantic our sampling captures only the second step of 2 °C cooling between 23.4 and 23.0 Ma, and average temperatures recover by nearly 2 °C by 22 Ma. These results suggest a consistent cooling of SST in both hemispheres during the Mi-1 glaciation. Bulk carbonate δ18O records from Site U1406 shows similar magnitude and timing of change as benthic δ18O from the site, and bulk carbonate from Site 1168 shows similar magnitude and timing of change as the deep South Atlantic Site 1264. Additionally, we generate surface ocean δ18Osw estimates from alkenone SST and coccolith dominated bulk carbonate δ18O for both sites. Comparison of these estimates with benthic δ18O is consistent with significant deep ocean cooling during both the Mi-1 and Mi-1.1 (22.5 Ma) glacial intervals. Together, these records indicate that Mi-1 was associated with coherent surface ocean cooling across both hemispheres, supporting a tightly coupled response of the surface ocean, deep ocean, and Antarctic cryosphere during this major glaciation.
- Preprint
(2570 KB) - Metadata XML
-
Supplement
(468 KB) - BibTeX
- EndNote
Status: open (until 28 Aug 2026)
- RC1: 'Comment on egusphere-2026-3415', Jared Nirenberg, 21 Jul 2026 reply
-
RC2: 'Comment on egusphere-2026-3415', Anonymous Referee #2, 13 Aug 2026
reply
This manuscript presents new high-resolution alkenone-based SST and carbonate stable-isotope records across the Oligocene-Miocene transition from ODP Site 1168 in the Tasman region and IODP Site U1406 in the North Atlantic. The dataset is valuable and fills an important gap in orbitally resolved surface-ocean records across the Mi-1 interval. I consider the manuscript suitable for eventual publication. However, several interpretations are currently unclear by the chronology and proxy uncertainties. In particular, I have concerns about the non-independence introduced by the Site 1168 age model, uncertainty in reconstructed surface-water delta O18, and the attribution of benthic delta O18 changes to deep/intermediate-water cooling. I therefore recommend a major revision.
Major comments:
- The Site 1168 age model is refined by tuning maxima in delta O18 bulk to maxima in Site 1264 delta O18 benthic, based on the assumption that the two records contain a common in-phase component ( line 147-157). This creates some circularity when the subsequent similarity in timing between these records is used to support synchronous surface, deep-ocean, and cryosphere changes. The large changes in inferred sedimentation rate across Mi-1 should also be discussed in this context.
- The manuscript interprets negative anomalies in δ18Osw- δ18Obenthic during Mi-1 and Mi-1.1 as evidence for significant deep-ocean cooling. This interpretation is plausible, but not unique. The difference also depends on local surface-water salinity and hydrography, deep-water δ18O, water-mass changes, age-model uncertainty, and the SST reconstruction itself. Indeed, Section 5.3 emphasizes substantial regional hydrographic influences from the Leeuwin Current and Gulf Stream. I suggest reframing deep-ocean cooling as one possible explanation rather than a uniquely diagnosed mechanism, unless independent deep-water temperature evidence can be provided.
- The description in Section 3.4 appears internally inconsistent. For Site U1406, the text describes subtracting benthic δ18O from surface δ18Osw, whereas for Site 1168 it states that surface δ18Osw is subtracted from Site 1264 benthic δ18O. Later discussion and Figure 8 consistently refer to δ18Osw–δ18Obenthic.
- The manuscript converts the larger Site 1168 benthic δ18O excursion into up to ~4°C greater intermediate-water cooling relative to deeper sites. This calculation implicitly attributes most of the excess δ18O change to temperature. However, Site 1168 was located at intermediate depth in a region likely affected by changing Southern Ocean circulation and water masses. Changes in local water δ18O or salinity could therefore also contribute to the larger benthic δ18O excursion. The ~4°C estimate should be presented explicitly as an upper-bound end member assuming no local change in water δ18O.
- At Site 1168, the SST reconstruction combines and , using where is saturated or affected by co-elution, and averaging the two proxies where both are considered reliable. Because these proxies have different calibrations and behavior near their upper limits, it is important to demonstrate that the inferred two-step cooling is not sensitive to proxy switching.
- The claim of “coherent” cooling across both hemispheres should also be moderated. Site U1406 does not resolve the first major cooling step seen at Site 1168, and the two sites show opposite SST trends between approximately 23.5 and 23.3 Ma. The records clearly indicate cooling at two widely separated midlatitude sites, but they do not demonstrate uniformly synchronous hemispheric-scale cooling.
Minor comments
- Line 11: Please provide the time of Mi-1.
- Lines 40-50: The motivation for high-resolution SST records is not clearly developed. The discussion of deep-ocean temperature and Antarctic ice-sheet albedo motivates surface-temperature constraints, but not high temporal resolution. I suggest linking the need for high-resolution SST more directly to orbital-scale variability in benthic δ18O and the risk of aliasing in lower-resolution records.
- Lines 284-288: The statement that a third cooling occurs between ~22 and 21 Ma appears inconsistent with the main interval presented in this manuscript. Please clarify the data supporting this statement.
- “bulk carbonate”, “fine-fraction carbonate”, and “coccolith fraction” appear to be used somewhat interchangeably. Please use consistent terminology and clearly state what material was actually analyzed.
- Figure 8: The caption refers to eccentricity as panel (e), although the figure appears to contain panels(a)-(d).
- Conclusions: The phrase “105 to 106yr timescale resolved by our sampling density” seems inconsistent with sample spacing of approximately 10–60 kyr. Please revise.
- Please use “significant” only where statistical significance has been demonstrated.
Citation: https://doi.org/10.5194/egusphere-2026-3415-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 93 | 24 | 13 | 130 | 9 | 13 | 11 |
- HTML: 93
- PDF: 24
- XML: 13
- Total: 130
- Supplement: 9
- BibTeX: 13
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
My comments are in the attached pdf.