the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Aug 2025
Insights into Lake Baikal Radiocarbon Age Offsets from a Database of AMS Radiocarbon Age Estimates
Abstract. Radiocarbon dates are an essential tool for dating non-varved lake sediments, however their interpretation is hindered by issues such as reservoir age or contamination which culminate in age estimates that can be thousands of years younger or older than the true depositional age of that sediment (we call this an age offset). Often, precise estimators of the radiocarbon age offset are not available, as in the case of Lake Baikal. Linear regression of uncalibrated radiocarbon dates has been used to estimate the age offset, with answers ranging from 0 to 1500 14C yr BP. These have been interpreted to suggest that different regions of Lake Baikal have different age offsets, although some dispute this. Other estimators have returned estimates of approximately 2000 14C yr BP. Despite this, most previous studies have not included any estimates of uncertainties for these age offsets in their proxy analysis, or have included uncertainty of, at most, ± 90 14C yr. Here, we present a complete database of published AMS radiocarbon dates from Lake Baikal sediment cores up to 2023 and, using this, review the use of linear regression on uncalibrated radiocarbon ages as a method for estimating age offsets from the sediments of Lake Baikal. We apply a standardised linear regression age offset method to all cores in our database to better quantify the age offset of Total Organic Carbon (TOC) in the lake’s sediments. We conclude that there is no statistically significant evidence from linear regression methods for a large difference in age offset in different regions of Lake Baikal. Our results return a lake-wide age offset estimate of TOC of 1.56 ± 0.75 14C kyr BP, suggesting previous studies in Lake Baikal have significantly underestimated the temporal uncertainty of radiocarbon ages. Finally, our results are a caution that linear regression-based age offset estimates in lake sediments have a large uncertainty that might only be observable with multiple datasets.
- Preprint
(1096 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 20 Sep 2025)
-
RC1: 'Different TOC flavors', Darrell Kaufman, 18 Aug 2025
reply
Lake Baikal is one of the most extensively studied lakes in the world. The compilation of published radiocarbon ages from numerous lake sediment cores represents a valuable Earth System Science dataset, and this article provides important context to support it. The dataset will serve future researchers interested in a variety of environmental changes recorded by Lake Baikal sediments. Although this audience may be relatively small, the authors could broaden the significance and novelty of their work by offering a model for reporting radiocarbon ages from sedimentary “total organic carbon” (TOC).
There are different “flavors” of TOC (also called, bulk sediment) depending on how the sediment is pretreated prior to 14C measurement. The choice of procedure can significantly influence the resulting 14C age. Documenting which pretreatment method was used for each sample is necessary not only to compare results across laboratories, as is done in this article, but also to better understand the sources of old carbon underlying the age offsets and thereby to address the authors’ motivation for compiling the dataset– for “insights into radiocarbon age offsets” as stated in the article's title.
The article would be strengthened by extending the context beyond age modeling to recognize that age offsets have been extensively studied in the broader context of watershed carbon cycling in both lake and marine depositional settings. This literature includes reports of downcore changes in age offsets, which reflect shifts in carbon cycling and directly affect the use of downcore trends to estimate long-term TOC age offsets and associated uncertainties, the focus of the data analysis in this article.
For results from different laboratories to be compared with confidence, the specific fraction of organic carbon analyzed should be the same. At first glance, the offset ages appear higher for samples measured in the Nagoya lab than in those from Woods Hole. If so, this difference may reflect pretreatment methods: the former retains all organic carbon (in-capsule acidification), while the latter removes acid-soluble organic carbon as well as organic carbon adsorbed to fine particles (acid washing through a 1 µm filter). While I am not aware of any published studies that quantify the difference in the outcome of these two procedures for 14C, I hypothesize that acid washing and filtration preferentially remove a younger fraction of 14C. I suggest that the authors use their dataset to test the possibility that the age offset depends on how the sediment is pretreated. This could contribute to the goal of using the dataset for "insights into radiocarbon age offsets."
Furthermore, the surface sediment ages reported by Fedotov et al. (2023) are conspicuously old. Although I could not easily determine the protocols used by the Golden Valley AMS lab, I wonder whether the sediments were treated with an alkaline solution to isolate the residual humin fraction, which could explain the anomalously old surface ages. I also note that the dataset lists the “dated material” for these samples as “bulk silty clay” rather than “TOC," which highlights the need for a more precise documentation of what was actually analyzed.
A number of radiocarbon ages in the dataset are cited as “personal communication.” The associated pretreatment procedures are therefore not available. Moreover, including unpublished data seems inconsistent with the authors’ statements that the dataset is derived from published sources. If this article is the first publication to report these ages, then say so; if not, cite the publication.
A full accounting of what constitutes “TOC” should include:
- Whether sediments were sieved to remove macrofossils or ground to incorporate particulate organic matter, which some scientists consider part of TOC.
- How sediments were acidified to remove inorganic carbon (decarbonated), either:
a. in-capsule, which retains acid-soluble organic carbon (fumigation or liquid), or
b. by rinsing, which removes acid-soluble organic carbon. - Whether the sediments were filtered, and at what pore size (typically 1 µm), which removes fine clay and other particles along with adsorbed organic carbon.
- Whether the sediments were treated with an alkaline solution to remove base-soluble organic carbon (humic acids), leaving behind humins, which are nearly always older than the total 14C population of the sediment.
A dataset designed to provide “insights into radiocarbon age offsets” would also benefit from including, where available, measurements of both carbon and nitrogen abundance in the analyzed material. Alongside δ13C, which is already included, these data can help estimate the proportion of terrestrial versus aquatic carbon sources. Moreover, carbon concentration (C%) is typically inversely related to offset age and can be used to estimate how age offsets vary downcore.
In my view, the text requires extensive revision to adopt a style more appropriate for a journal audience. As it stands, much of the article reads like a first draft. It could be shortened by at least one quarter, perhaps one third, by eliminating repetition, trimming unnecessary words and phrases, using more precise wording, and moving details of specific corrected errors into the dataset’s “comment” field.
Finally, the article would benefit from a concluding figure that clearly illustrates its main findings. Figure 4 provides a useful overview of all ages, but a new figure focusing on the 19 accepted core chronologies is needed. I suggest plotting ages by depth for each of the six regions in Table 5, using distinct symbols for different cores from Buguldeika Saddle and Academician Ridge.
In summary, the manuscript presents a valuable compilation of radiocarbon ages from Lake Baikal sediments. However, the manuscript requires substantial revision before it I would recommend it for publication. In particular, greater attention is needed to (1) clearly documenting what constitutes “TOC” across laboratories by specifying the pretreatment method used for each sample, (2) engaging more fully with the literature on age offsets, including those that specifically investigate temporal changes in age offsets, and (3) improving clarity and conciseness throughout the text.
Minor Comments
- Line 11:The article does not address “contamination.” I recommend avoiding confusion between age offset, which generally reflects TOC ages older than the time of deposition, and contamination, which results an age that is too young.
- Line 15 and throughout:When presenting age offset values, omit “BP.” Offsets are reported as (radiocarbon) years, not years "before present."
- Line 36:Not all radiocarbon ages require offset correction.
- Line 48 and elsewhere:“Age-offset” does not require a hyphen.
- Line 49:Replace “true age” with “timing of deposition” or similar phrasing.
- Line 56:The hardwater effect involves more than “the presence of carbonate rocks in the lake” and can significantly contribute to age offsets. It deserves more attention in the article, even if it is not a significant issue in Lake Baikal. Also, I recommend not including contamination as part of the definition of age offset, since contamination is typically spurious, post-depositional, and unrelated to watershed carbon cycling—unlike offsets, which provide meaningful insights into the carbon cycle.
- Line 75:Change “2009” to “2009a” or “2009b.”
- Line 79:Replace “Ramsey” with “Bronk Ramsey” here and in the references.
- Line 137:When reporting depth, specify the datum: (1) core interval top (often embedded in the sample ID), (2) lake bottom (below lake floor), or (3) lake water surface.
- Line 140:If the template is not publicly available, there is no need to advertise it.
- Line 141:The claim that the dataset “draws heavily from Millar (2014) and Stuiver and Polach (1977)” is confusing. Millar (2014) identifies six elements that should accompany each radiocarbon age, three of which concern calibration, which is not addressed in this article, and one of which is “pretreatment procedure,” which I strongly recommend including. Stuiver and Polach (1977) focus on conventions for radiocarbon age calculations, which do not appear to be central to the reporting template
- Line 147:Provide details on how the reported depths were "corrected."
- Line 149:Among the δ13C methods listed in the dataset is “AMS.” My understanding is that graphitization and ionization during AMS analysis can cause δ13C fractionation. While measured and used to calculate 14C ages, the AMS-derived δ13C may not reflect the true value. For clarity, replace “AMS” with “Approximated from AMS” or include a note of explanation.
- Line 150:Clarify terminology. To me, “carbon yield” refers to the mass of carbon measured, not the weight percent carbon in the dated material. Both are relevant: very small carbon masses can produce unreliable AMS results, while carbon abundance often scales with age offset.
- Line 165:The effect of changing sedimentation rate on age offsets deserves further discussion. Consider the differing impacts of millennial-scale sedimentation trends (potentially a major effect) versus shorter-term (century-scale or shorter) fluctuations around a mean (little or no effect), as well as the unique challenges of estimating near-surface sedimentation rates, which are strongly influenced by porosity and core handling.
- Line 168:I agree that assuming a constant age offset is a major weakness when using TOC for core chronologies. I suggest including estimates from the literature that document shifts in age offsets, and discussing how such variability could influence intercept ages.
- Line 180:Clarify “median depth of those ages.” Does this mean the median of all depths, or the midpoint of each sample?
- Line 181:Using 14C ages rather than calibrated ages is reasonable given other potentially more important uncertainties. However, the rationale that the ages first need to be corrected before they are calibrated is weak. Calibrated ages can be used to estimate age offsets prior to correction. However, intercept (offset) values derived this way would be somewhat too low, as the slope of the age–depth regression would be steeper. The best estimate is probably somewhere between intercepts derived from 14C ages (too high) and from calibrated ages assuming no offset (too low). Presumably, code could be developed to iterate on the optimal intercept value.
- Line 197:Specify that the reported ages are based on total organic carbon.
- Line 197:Clarify what is meant by “mean” radiocarbon age. Mean of how many replicates?
- Line 240:I disagree with the statement that “POM” and “FOM” represent “two different styles of sampling TOC.” TOC includes molecular and other forms of organic carbon not retained by sieving, unlike POM and FOM.
- Line 453:I disagree with the claim that “diatomaceous sediment cannot be used to infer the age offset for TOC.” Colman et al. (1996) state that Lake Baikal sediments are diatomaceous muds during interglacial intervals.
- Line 455:The key issue with using surface and trapped sediments to estimate age offsets is that they may contain bomb carbon, which could lead to underestimating the offset.
- Line 481:Water residence time is directly related to reservoir age. I believe the authors mean “age offset.” A long residence time can cause apparently old ages in plants assimilating dissolved carbon.
- Line 495:“Identical” is too strong. Consider “indistinguishable” or a similar term.
- Line 519:Be precise. For example: “an estimated 1σ uncertainty of ± 0.5 kyr.”
- Line 519:The suggestion that the primary source of uncertainty in the offset is “error in the radiocarbon dates themselves” seems unlikely to me. The average error for 14C ages <13 ka in the dataset is ± 65 years. A more plausible source of uncertainty is temporal variability in offsets, long-term trend in sedimentation rate, and challenges in recovering undisturbed sediment–water interfaces. If sources of "dating error" other than analytical precision are suspected, they should be explained.
- Punctuation errors: Lines 91, 105, 235, 383, 517.
- Typos :Lines 219, 312.
- Reference citation style: Many inconsistencies.
- Data check:Verify the reported 14C age of –13.365 ka (NUTA-3372).
Citation: https://doi.org/10.5194/essd-2025-109-RC1
Data sets
Lake Baikal sediment core AMS radiocarbon data S. R. S. Newall et al. https://doi.org/10.1594/PANGAEA.973799
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 157 | 23 | 9 | 189 | 3 | 3 |
- HTML: 157
- PDF: 23
- XML: 9
- Total: 189
- BibTeX: 3
- EndNote: 3
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1