Reply on RC2

Hollis et al report new stable carbon isotope measurements of organic matter from sediments deposited on the continental shelf and slope of New Zealand and eastern Australia during the late Paleocene (termed the Waipawa organofacies). The authors identify unusually high δC values measured within the Waipawa organofacies, consistent with measurements made by others on contemporaneous sections in China and Argentina. The authors use a detailed suite of geochemical analyses (including bulk and compound specific stable isotope analysis) to claim the unusually high δC values are caused by a combination of lignin degradation and low CO2 levels. Associated with this event is global cooling (and growth of ice sheets and fall in sea level) that likely resulted from lower atmospheric CO2 (evidenced by the high δC values), which may have been caused by reduced volcanism and increased carbon burial.

The authors make the connection between the high δ 13 C values and low CO 2 , but a quantitative estimate of CO 2 is lacking. Using the terrestrial δ 13 C data to quantify CO 2 would allow for a more useful comparison of CO 2 and temperature, and greatly improve what is presently a very qualitative comparison (high δ 13 C = low CO 2 and cooling = low CO 2 ). This is particularly important given that "The relationship between temperature and atmospheric greenhouse gas levels through the Paleocene is very poorly resolved…" (40) and the authors state (75), "we explore the possibility that this 13C enrichment of bulk OM reflects a short-lived drawdown in atmospheric CO2, reflecting the relationship in carbon isotope discrimination between atmospheric CO2 and C3 plant biomass (Cui and Schubert, 2016, 2017, 2018Schubert andJahren, 2012, 2018)." Yet, any determination of CO 2 using this relationship is conspicuously absent. Furthermore, the authors later state (290), "Only by accounting for potential processes of 13C-enrichment during OM transportation, deposition and early diagenesis it is possible to identify any residual enrichment that may be related to a drawdown in atmospheric CO2 levels." Why do all this if CO 2 is not going to be estimated quantitatively (even if only a back of the envelope calculation to show a possible range of CO 2 drawdowns, given possible marine influences, and autogenic processes)?
Alternatively, the authors could calculate CO 2 given their interpretation that (376)(377), "the pristane CIE implies that the primary terrestrial substrate is enriched in 13 C bỹ 4‰." The authors could also calculate CO 2 for a range of CIE magnitudes, to show the magnitude of CO 2 change that would be required to get any size CIE. It would certainly help to better answer the question of whether a drawdown in CO 2 is a plausible explanation for the δ 13 C trends and the observed cooling (the current assumption is there was cooling therefore CO 2 must have decreased). Is the purported 20-30% decrease in CO 2 required for a 1 o C decrease in deep sea temperature (455) consistent with CO 2 estimated assuming a +4‰ terrestrial CIE (based on the terrestrial CIE)? If so, that would greatly support the stated conclusions linking high δ 13 C to low CO 2 (and the various processes indicated within). If not, it may suggest climate sensitivity differed from the 3 o C assumed here, which would also be an interesting result. Much of the work to assess climate sensitivity in the Paleogene has focused on the warmest periods.
Besides, the aquatic sources show a similar 2-4‰ shift to the terrestrial sources (380)(381)(382) (Barnet et al., 2019), which is consistent with a modest (20-30%) decline in CO 2 , assuming a climate sensitivity of 3°C." Given all the work that was done to quantify the various OM sources and degradation, this statement is a bit disappointing (besides, the authors do assign values, e.g., 550-553, where they identify a residual excursion of ~2.5‰, exclusive of degradation processes, or the purported 4‰ CIE measured in phytane, 376). As noted above, even a back-of the-envelope calculation given a few assumptions (or a range of CIE sizes) would be useful to see if a CO 2 decline is even a plausible interpretation from the δ 13 C data. Otherwise the entire premise of a CO 2 decline is based solely on data separate from this study (deep-sea benthic δ 18 O data and climate sensitivity estimates).
Thanks very much for these comments. To be honest this is an issue that was intensely debated by the co-authors. Some of us were very much in favour in making specific CO 2 determinations, whereas others argued that the uncertainties were too great to provide an estimate. As reviewer 1 notes, we were concerned that any specific estimate would likely be widely cited because data for this interval are so sparse and for this reason, we resiled from including an estimate. However, the reviewer has convinced us that this is a major shortcoming in the paper. Therefore, we have prepared a new section for the revised paper (see Calculating CO 2 in the attached file). Note that the requirement to differentiate between terrestrial and marine OM is simply because the method we employ is based on the δ 13 C of terrestrial OM.
We welcome further comments from the reviewer on this addition to the paper.

Specific Comments:
82: "From these analyses, we estimate the magnitudes of the δ 13 C excursion in both primary terrestrial and marine OM and use these values to infer broad changes in the concentration of atmospheric CO 2 ." Where is the calculation of CO 2 from the δ 13 C data?
This is now added as noted above.
Many of the geochemical methods are repeatedly simply cited back to Naeher et al. (2019), rather than being reported here. At least, a brief summary of the methods used here would be useful to the reader. For example, some important details on the standards used for IRMS and the analytical precision of these measurements, which may differ from the previous work? This was done for the compound specific work, but would make reading this paper easier as a stand-alone product, without needing to read back to Naeher et al. (2019) for the methods.
Agreed. We will add summary methods to this paper.
The summary paragraph of Section 5.4 " 13 C enrichment attributable to drawdown of atmospheric CO 2 " lacks any description of how 13 C enrichment relates to drawdown of CO 2 .
Yes. Text was transferred to section 6.2 but we realise that it leaves the question hanging, so will amalgamate these two sections.
Conclusions. I think a calculation of CO 2 from the δ 13 C data would go a long way towards bolstering the linkages between CO 2 , cooling, C burial, volcanism, and sea level, etc proposed in the conclusions.
Yes, agreed. Important reference but only discusses the negative shift in δ 13 C in the first phase of diagenesis.