the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 14 May 2025
Short communication: Estimating radiocarbon reservoir effects in Bolivian Amazon freshwater lakes
Abstract. The Llanos de Moxos, in the Bolivian Amazon, preserves a remarkable archaeological record, featuring thousands of forest islands. These anthropogenic sites emerged as a result of activities of the earliest inhabitants of Amazonia during the Early and Middle Holocene. Excavations conducted to date on the forest islands have revealed that many assemblages contain a high number of ancient freshwater snail remains. In these shell middens, the most represented mollusc taxon, and in most cases the sole one, is Pomacea spp., a genus that inhabits inland shallow lakes and wetlands. Although human burials and faunal remains are typically recovered from these sites, their collagen is often not preserved or is of poor quality, and shell carbonates from Pomacea shells, along with carbonised plant remains, are often used for 14C measurements. However, it remains undetermined if these measurements are subject to radiocarbon reservoir effect (RRE). To determine if a freshwater RRE could affect the age estimations of Amazonian archaeological and other paleoecological deposits, we collected modern coeval Pomacea shells and tree leaves from four locations across the Llanos de Moxos area for AMS radiocarbon dating. The radiocarbon results combined with the environmental history of Llanos de Moxos during the Holocene, confirm an absence of significant RREs, and support the continued use of freshwater molluscs as viable material for radiocarbon dating in the region.
- Preprint
(530 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 03 Jul 2025)
-
RC1: 'Comment on egusphere-2025-1960', Anonymous Referee #1, 28 May 2025
reply
The manuscript describes a small set of contemporary paired Pomacea snail shells and terrestrial plants from lakes in the Llano de Moxos, Bolivia, to test whether there is a freshwater reservoir effect -- currently -- that would need to be corrected for in radiocarbon dating lake cores or archaeological materials in the region. The article is a Short Communication, which is appropriate given the limited scope of the number and contexts of the samples. It is a simple and straightforward report of some empirical work to investigate the problem, a simple contribution. As simple as the experiment is, there is actually almost no analysis of the data presented, and the reader gets the sense that the authors don't have a full grasp of the technical aspects of radiocarbon dating. This is odd enough given that several co-authors have worked with 14C in archaeology for years and one is a self-appointed expert on reservoir corrections. Even if it is a small contribution, which doesn't show a strong reservoir effect in the Llanos de Moxos, the analysis can be presented to help guide others that will (should) replicate aspects of the study in other parts of the region or world.
Things to address:
Abstract, Ln 28-29: "confirm and absence of RREs". You can't confirm an absence. Change this to say that there is no evidence of a reservoir effect in the contemporary setting. This is overstated compared to what is discussed in the body of the paper, where the authors acknowledge the small number of samples and contexts and that no samples address time depth. cf. Line 60 where characterizing reservoirs over time is given as an important goal for research.
Ln 63: "Llanos de Moxos is a noncalcareous region". In terms of bedrock geology this might be true, but are there not caliche deposits and cements in the forest island middens of mid Holocene age? Aren't the problems of collagen preservation directly attributable to high pH created by carbonates, phosphates and sulfates? All of that implies "hard water" even if it is created by excessive evaporation of groundwater. Those carbonates themselves can be a source of dead carbon. Note this applies to the discussion of evaporites in Ln 120-124 as well.
Ln. 88 - 101 Methods in general. This is a weird salad of pieces of methods boilerplates. It is incoherent. I would suggest you describe the pretreatments for shell and plant matter separately, taking each process to the production of CO2 and/or graphite, and then say the AMS used. As written the methods described are technically impossible.
Ln. 88: How much HCl, at what temperature for how long? Was msucle and the periostracum removed? Any bleaching to remove organic carbon?
Ln. 89: The shells were not combusted in an EA. Shells are hydrolyzes with e.g., phosphoric acid on a gas bench or simple blood sampling vials. Presumably the plants were combusted in the EA. Was any pretreatment done on them (e.g., ABA?)
Ln 90: C/N ratio. C:N would not be calculated on the shells. You only report one C:N ratio on a plant sample, and since the C:N is not informative here (as it would be for collagen or keratin), one gets the sense that this is not actually what happened -- i.e. that someone actually carried out a "first check" on the C:N ratio. Delete or clarify.
Ln 92: The protocol outlined by Vogel 1984 .... Arguably what is meant is that the AGE system does hydrogen reduction on an iron powder catalyst. I would say that; if you read Vogel et al. 1984 you'd see that as a protocol, how to do the procedure, AGE is not the same procedure (nor are most other setups -- the point in citing that article is to refer to hydrogen reduction on an iron catalyst, per se).
Ln 94: Clarify this. Measurements were not done on both of the two AMS systems. Which samples were run on which type of AMS? And why? And where? I am pretty sure that CIRAM outsources its measurements through an "industry cooperative arrangement", so it would be good to know who runs those instruments.
Ln 95-96: "based on the comparison between the concentration measurements of 13C/12C and 14C/12C". Just cut this phrase and say whether the correction was made using a d13C value measured on the AMS, measured on an IRMS, or estimated.
Ln 96: I would prefer to see F14C given rather than pMC, and refer to Reimer et al 2004.
Ln 100: Change to: per mille or per mil; with respect to V-PDB; and once and for all: AIR is not an acronym, it is just air. The standard is Atm. N2. I.e., the d15N value of all N2 in all the air everywhere on earth is 0 per mille. "Ambient Inhalable Reservoir" is a stupid joke that Henry Schwarcz or somebody made to an archaeologist in the 1980s and the archaeologist thought they were serious. Everyone else did an emperor's new clothes after that because they didn't want to look ignorant while "doing isotopes".
Section 3 Results. Nowhere in this section is a statistical test of the 14C data made. And yet over and over it's "no statistically signifcant difference". You need a chi-squared test for these paired samples -- Ward and Wilson 1978; or just use the R_Combine command in OxCal, which generates the chi-squared stat for you. You need to report that in a table.
There are no significant differences in 14C activity between most pairs. BUT ... there is an outlier with the Laguna Azul trio, and it's not clear what the authors are trying to do to explain that. The crux is that it's actually more bomb than the other shell and leaf in the group. The authors note "we cannot fully exclude localised differences in ¹⁴C carbon sources available to the two molluscs", which is true but is calls the validity of all the sample pairs into question. Perhaps because the shell is not "older" looking, having a higher 14C activity, the implication is ignored. The presence of bomb carbon implies that the bomb spike is still being attenuated through the system at Laguna Azul (if not elsewhere). If that is the case then a dead carbon contribution diluting 14C concentration could be currently balanced or counteracted by excess bomb 14C, leaving you with apparently no reservoir effect. That needs to be considered. I get the sense that it's popular to assume that we're past the bomb peak and can use contemporary biosphere (not atmosphere) as a 100% modern baseline, but I think the vaildity of that assumption needs to be demonstrated rather assumed.
Ln 111: "The higher δ¹³C value in LA.1 when compared to LA.2 may be related to a higher consumption of potentially depleted organic matter by LA.1." Unclear what is meant by this sentence. If they are talking about the different d13C values, LA1 is higher because it's a carbonate, LA2 is lower because it's a plant. Not clear what question they are trying to respond to with this statement.
Ln 118: I would say that the results are CONSISTENT with little or no freshwater reservoir. It's not enough evidence to confirm the hypothesis. Also the same on Ln 150.
Finally, with respect to suitability of these shells for dating, what about diagenesis? There needs to be some discussion of carbonate exchange and alteration in the depositional context. Again, there are cemented middle Holocene middens there, and they got bicarbonate ions from somewhere, and they exchange in the sediments. So in addition to the paragraphs explaining why climate would not affect reservoir conditions through the Holocene, one does need to consider whether the snail shells are good candidates given more mundane factors affecting their isotope ratios like diagenesis.Citation: https://doi.org/10.5194/egusphere-2025-1960-RC1 -
RC2: 'Comment on egusphere-2025-1960', Anonymous Referee #2, 10 Jun 2025
reply
The article aims to determine the reservoir age, or lack thereof, for Pomacea shells that commonly found in and often used for radiocarbon dating of archaeological sites in the Bolivian Amazon. Modern shells were radiocarbon dated and compared to leaves of terrestrial plants taken from the nearby locations at the same time. The radiocarbon content of the corresponding samples is generally in agreement so that the article concludes that there is no radiocarbon reservoir effect in the shells. The article further discusses possible effects that could have changed the reservoir age in the region throughout time but concludes that there are no large changes to be expected.
General comments:
The article gives a good summary of the situation around mollusc dating in the study area. Since the focus is on Pomcea sp., some more details about them should be given, especially things like lifetime or growth pattern/direction of the shells that are relevant for sampling for the radiocarbon measurements.
The sampling approach of shells and terrestrial plants is reasonable, but more information on the sampling and selection criteria should be given in the article. In the end, only 9 radiocarbon dates are used for a large study area and considering there was a sample that did not agree with the findings of a negligible reservoir effect, there is chance that dates of molluscs and plants only accidentally match.
The article considers changes of the reservoir effect over time well.
Specific comments:
Line 76-80: The description mentions the size and feeding habits of the snails. The relevant part missing here is the lifetime as this is critical to assessing a reservoir age. It could also be interesting to describe to growth pattern of the shells to possibly identify the youngest parts of it for the measurement.
Line 81: It is mentioned that the specimens were collected alive, but was there any criteria live assumed age or size for the sampling, or were they just picked at random?
Line 84: Tree leaves are mentioned, but no further details are given. As the trees can store carbon for quite some time as wood, more details are needed here. The leaves will probably have significantly shorter live time than the tree itself, but they are ideally annual, or at least of a similarly age than the shells, but none of that is mentioned.
Line 88A: Were the whole shells used? If not, how were they subsampled and did that happen before or after the HCl treatment?
Line 88B: HCl treatment is mentioned for shell samples. As no reference is given, the amount of acid used (probably per sample mass) should be given for completion.
Line 88C: There is no pretreatment for the leaf samples mentioned. As there are fresh samples, pretreatment is probably not needed, but that should be mentioned in the article.
Line 90: A C/N ratio check is mentioned here. However, the value from this is not given. Also due to the material used containing barely any nitrogen (only one of the samples seem to have enough to allow a N-15 analysis according to table 2), probably just the carbon content would be a more relevant quantity to as a quality check.
Line 98: Were the samples also combusted for IRMS measurement? Or were carbonates dissolved in acid?
Line 105: The difference between 2023 and 2024 measurements is attributed to a general atmospheric decline. However, the difference of 1 pMC or more in a single year is much larger than what is usually observed in the atmosphere.
Line 110A: As a localized effect is considered here, maybe more information on the sampling/selection criteria for the samples should be included in the article to discuss this.
Line 110B: As the sample LA1 is higher than the terrestrial samples, it has to be assumed that the sample is either old (mentioned above), or that there is a reservoir of “old” bomb carbon somewhere in the system that might only be released under certain conditions. More information should be provided to discuss this outlier.
Line 118: Most of the shell data seem to fit the terrestrial dates. However, the elevated C-14 contents of sample LA.1 and maybe CHU.100, indicate that there could be a reservoir effect. Using these values could lead to an upper limit of the reservoir effect that could still be small enough to be useful for archaeological dates.
Citation: https://doi.org/10.5194/egusphere-2025-1960-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 133 | 42 | 10 | 185 | 5 | 8 |
- HTML: 133
- PDF: 42
- XML: 10
- Total: 185
- BibTeX: 5
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1