the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Aug 2024
A direct comparison of single grain and multi-grain aliquot luminescence dating of feldspars from colluvial deposits in KwaZulu-Natal, South Africa
Abstract. The erosional landscape of the Jojosi Dongas in KwaZulu-Natal, South Africa, expose accretionary slope deposits that preserve important geological and archaeological information. This landscape has been occupied by modern humans during the stone age for many thousands of years as evidenced by the presence of numerous stone artefacts interbedded within at least three phases of gully cut-and-fill deposits. A contextualisation of the artefacts and their role for human evolution in southern Africa, as well as developing an understanding of the environmental conditions that shaped this inhabited landscape, is only possible by establishing a robust chronological framework.
Here we use luminescence dating of feldspars to constrain the geochronological framework for the sequence of accretionary hillslope deposition at Jojosi at three sampling locations. Initial suspicion of poor bleaching led us to measure single grains of feldspar, which revealed low luminescence sensitivity of the individual grains and a variable proportion of grains in saturation. Summing the luminescence signal of individual grains and creating synthetic aliquots enabled us to study the effect of signal averaging on the luminescence sensitivity, signal saturation, and dose distributions. We then compare the results from individual grain measurements and synthetic aliquots to true multi-grain aliquots. To allow for a quantification of the results, we apply four different dose models to the distributions, including the Central Age Model, the Average Dose Model, BayLum, and a standardised growth curve (SGC) approach using an averaged Ln/Tn value interpolated onto the SGC. Doses calculated for the different samples range from ~80 Gy to ~800 Gy and contain up to 67 % saturated grains. We evaluate the performance of the different dose models over this large dose range, with samples close to the saturation level of feldspar luminescence.
On average we find good agreement between the results obtained using the different dose models, but observe that samples with a large number of saturated grains impact the consistency of the result. Overall, all dose models and data sets give consistent results below a saturated grain threshold of ~15 %, corresponding to a dose of ~120 Gy in this study.
Finally, we favoured BayLum for age calculations of the single grain and multi-grain aliquot data sets, representing the opportunity to refine the chronology by including stratigraphic information in the age calculations. We were able to establish a chronology for the three sampled sections within the Jojosi Dongas constraining erosional and depositional processes from ~100 ka to ~700 ka, and human occupation of the area in the early MIS 5 and late MIS 6.
- Preprint
(5541 KB) - Metadata XML
-
Supplement
(6052 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on gchron-2024-19', Anonymous Referee #1, 01 Oct 2024
Dear editor,
This manuscript presents a systematic comparison of single and multi-grain luminescence dating of feldspars to date colluvial sediments from KwaZulu region in South Africa. The authors comprehensively investigate the saturation of the IR signal on single and multi-grain aliquots. Then, they apply four different dose models to determine the equivalent dose (De) of their samples, including the central age model (CAM), the average dose model (ADM), BayLum, and the standardized growth curve (SGC) approach. The influence of the saturated aliquots on the De is also investigated. At the end they choose the BayLum for age calculation as it includes the saturated grains. The new ages constrain erosional and depositional processes from ~100 ka to ~700 ka, and human occupation in the area to MIS 5-6.
The manuscript is well written and represent an important contribution to the growing knowledge on feldspars behavior and its application in luminescence dating. I recommend receiving this manuscript for publication in Geochronology after addressing the following comments.
General comments
The feldspar samples dated in this study have very low K2O concentration of <0.5 %. Even 0 for one of the samples. Single grains from two samples were analyzed for mineralogy and oxides contents. This presents an opportunity to look into the correlation between the different minerals and signals. The authors looked into the correlation between K2O and the IR signals (in the supplementary), but not other oxides.
The manuscript deals with dating of colluvial sediments. This kind of sediments is hard to bleach. The authors conclude that there is no partial bleaching, based on the appearance of the dose distribution and the bleaching test of Buylaert et al. (2013). I’m not convinced that the samples are bleached. At high doses, the traps are filled in a slower rate, which results in a smoother dose distribution in case of partial bleaching (Fig. 3). In addition, the saturated grains/aliquots, are not represented in the dose distribution. Section 4 describes the amount of the saturated grains in the samples and their influence on the equivalent dose (De). But, the reasons for the presence of the saturated grains are not discussed. If these grains are not bleached, they should not be included in age calculation of the colluvial samples. The final conclusion of preferring BayLum is adequate. Yet, a more in-depth discussion is needed.
The geological and archeological context of the dated samples is a bit thin. Please elaborate on the significance of the new ages to the understanding of human presence in the area throughout the Pleistocene. How the ages correspond to the previous works on the local geomorphology.
Specific comments
L32-34: Please rephrase this sentence. Abstract should not contain “suspicions”.
L63: Please refer to the location map at the first mention of the study area.
L64: Do not site a paper which was not yet accepted by a journal.
L63-66: Please add a reference to this sentence.
L68-69: What is the chronological framework of the colluvial slope processes according to previous works?
L69: Li and Wintle (1991) investigates the luminescence sensitivity of samples from KwaZulu region. But these samples are not dated in the paper. This reference is not relevant here.
L73: Please add a reference at the end of the sentence.
L113: Please add a rough assessment of the distance from the source rock.
L114-129: There is a mixture of past, present, and future tenses. Please be consistent.
L127: Please add Roberts and Duller, 2004 for the SGC references.
L127-129: Please change to past tense.
L135: See comment to L64.
L213: You state the signal and background integrals for the single grain measurements, but not for the multi grain. Please add these details.
L240: Here and in the supplementary it is written that 46 grain were analyzed; in L245 and L247 it is written that 45 grain were measured; in the caption of figure S1a it is written 44 grains. There only 45 grains in tables S1a,b. Please be consistent.
L243-244: Please add a reference to the K2O contents range (0 wt% to 16.9 wt%).
L284-285: Please add an example for each type of model.
L350: Please add IRSL50 results to the supplementary data (Equivalent to table 4 in the manuscript).
L351: According to Table 4, the OD of the SG De distributions for post-IR IRSL225 ranges from 26 % (sample JOJO-TRPL-2) to 74 % (sample JOJO-5-5).
L367: According to table 3, the lowest percentage of saturated grains for IRSL50 is 0.7 % (sample JOJO-85U).
L368-370: Can you add a plot of number of saturated grains vs. SG CAM dose to the supplementary?
L380-381: This sentence is not clear.
L379-382: You start two successive sentences with “However”. Please rephrase.
L382-385: Although IR50 fading is very probable, it dose not excludes partial bleaching of the post-IR IRSL225 signal, especially if taking into consideration the sedimentary environment.
L398-400: Do you have an explanation to this observation?
L410-411: For SG (350-351) you present the OD values range without errors, while for the MG and the SynAl you present the OD values range with errors. Please be consistent.
L411: According to Table 4, the OD of the SynAl De distributions for post-IR IRSL225 ranges from 14 % (sample JOJO-85U) to 40 % (sample JOJO-5-5). Is the confusion in the locations of MG and SynAl in table 4?
L522: Consider to move Fig.S7 to the manuscript, as it has important visualization of the dose response curve characteristics between the different methods.
L424-425: Please change to past tense.
L436-438: Please move this information to the table’s caption.
L571-574: Please refer to Fig. 5.
L604: Generally, in active fluvial systems, sediments older than MIS5 are not common. Please add examples from other places.
Figure 1: If the “Luminescence samples not included in this study” do not contribute any relevant information to the manuscript, please remove them from the figure.
Figure 5: Please add the sub figures to the caption (a, b, e, f).
Figure S4: Are the signals and the dose response curves belong to the same aliquots in each sub figure? If yes, state so. If not, please show corresponding curves.
Figure S6: Sub figures e, f are not described in the caption.
Table S1: What is “n” in the table? Is it the number of measurement points per grain? In the caption it is written that “two to six point measured on each grain”, but “n” is ranging from 1 to 7. In the caption of Figure S1 it is written “Two to three points were measured per grain”.
Figure S9: Please add a legend to the figure.
Figures S11-S12: Please delete the “<>” in the X-axis title.
Technical corrections
L70: Teeme et al. (2008) and Lyons et al. (2013) are missing from the references list.
L73: Please choose “over” or “for”.
L88: Huntley (2006) and Kars et al. (2008) are missing from the references list.
L95: Riedesel et al. (2018) is missing from the references list.
L114: Please change “possibility” to “possibly”.
L123: Please change “following” to “follow”.
L127: You cite Li et al. (2017) in the manuscript, but have Li et al. (2018) in the reference list.
L158: Botter-Jensen et al. (2010) is missing from the references list.
L246: it is written “max=16.3 wt%”, but in table S1b the highest K2O content is 16.23.
L311: Please change to “There is a possibility”.
L314: Li et al. (2017) is missing from the references list.
L316: Please change “an” to “a”, and “A” to “The”.
L514: Thomsen et al. (2005) is missing from the references list.
L567: Please change Fig. “7a” to “S7a”.
Citation: https://doi.org/10.5194/gchron-2024-19-RC1 -
CC1: 'Comment on gchron-2024-19', Barbara Mauz, 14 Oct 2024
This is an empirical study that uses sand-sized K-rich feldspar as dosimeter and compares data obtained from single grain and synthetic aliquots with data obtained from multi-grain aliquots using the pIRIR225 measurement protocol. Then the data are quantified using four different dose distribution models. This sounds like an approach with (too?) many parameters and, often, such papers are hard to read. Not this paper, I feel, because It is well-structured and methods employed are well described. I was particularly interested to learn how this study evaluates samples close to saturation level but this turned out to be a bit difficult, hence my first question (see below). While reading in order to find out how saturation was defined, a couple of other questions popped up, which are summarised below.
Saturation dose and D0 – section 2 describes a lot of details (some multiple times, e.g. filter) but not how these two values were determined. I gather from the text that a grain/aliquot exhibit saturation when Ln/Tn > Imax where Imax is the “maximum asymptote of the dose response curve” (line 312). The plots suggest that betamax = 800 Gy, but the D0 plot (Fig. S7a) shows D0 >500 Gy for a significant number of grains/aliquots. This suggests that maximum beta dose does not correspond to Imax. I am therefore wondering if the difference between D0-single grain and D0-multi grain aliquot is caused by the difference between Imax and betamax?
BayLum – it includes aliquot data exhibiting Ln/Tn > Imax and fits the single exponential function to the Lx/Tx data. How was De determined if interpolation is not an option? A Gauss distribution of individual doses around De was assumed also for those samples with a high De (e.g. >150 Gy)?
Grain size of the dosimeter – you selected 200-250 micron grain size - did you assume that this coarse fraction is better bleached than, say 90-150 micron grains, because grains did roll downslope individually and were not part of the bed load? The photos in Fig. 1 suggest transport of sediment by gravity and/or by water, each mode generating a different degree of bleaching across grain size fractions.
The CLL calibrates the beta source every 6 month, presumably using the same calibration quartz sample each time, i.e. quartz grains that have received the same gamma dose (e.g. 5 Gy). How do you fit a regression line through datapoints that have approximately the same value? Is the intercept always zero?
Introduction - you say that a number of studies exists already on colluvial deposits in S-Africa – I am wondering which open questions did these studies leave behind and how did these papers guide your study?
Gamma spectrometry - this may sound like my personal hobbyhorse, but since Murray et al. (2015; interlab comparison study) we know that we have an issue with gamma spectrometry. As a consequence, I think we should report details (measurement time and geometry, calibration, peak selection) in publications.
The term “palaeodose” means past radiation dose. I suggest to follow Huntley (2001; Ancient TL): “in our work it is not the actual past radiation dose that is determined, but the beta or gamma dose that results in the same luminescence intensity during thermal or optical excitation“ where “the same” is the fundamental assumption.
Citation: https://doi.org/10.5194/gchron-2024-19-CC1 -
RC2: 'Comment on gchron-2024-19', Anonymous Referee #2, 29 Oct 2024
General comments
In this study, feldspar single grain and multi grain aliquots were used to date colluvial deposits from KwaZulu-Natal in South Africa using a pIRIR225 protocol. A systematic comparison of the De values was conducted using four different age models. The effect of saturated grains on the calculated De values was also investigated for both single grain and multi grain aliquots. Ages derived using BayLum were considered for the final interpretation.
This manuscript is well written and is deserved a publication in Geochronology after addressing the comments.
For the type of environment that this study was conducted in, i.e. colluvium, partial bleaching can be an issue particularly for high temperature IRSL signals as also described in the text. I think that there are indications of partial bleaching present in some of the KDE plots displayed in Fig. 3. The MAM age model can be investigated in such a case and compared with other age models. If saturated grains are present in a sample, how one may know if these are not the partially bleached grains? Also what are the consequences of including them in the final age calculation (if they are actually not well bleached)?
Specific comments
Numbers refers to line numbering used in the text.
65: Please add a map or photo.
138, 139: Does this mean some areas were avoided and no samples were collected for dating (e.g. boundaries)
Fig. 1b can be better presented, to give a better view of the area.
188: the error is mentioned to be the standard error, is this the same for the IR50 signal?
223: Same criteria were mentioned in lines 198-199 for multi grains. So what is considered additional?
244: Please add a reference.
318: the term “extrapolation” should be avoided here as the whole process is interpolation onto the dose response curve.
338-340: Is there a reason for this difference in the accepted grain for IR50 and pIRIR?
Fig. 3d: There is only one accepted synthetic aliquot
Table 3 caption: Add something like ‘as indicated in brackets’ after “The relative number of accepted grains”
368, 369: This sentence is not clear.
380: It would be good to add the fitted equation.
381, 382: Please rephrase. This sentence is not clear.
409: Please briefly describe how you calculate this.
427: Please avoid the term “non-fading”
454: Is this D0 or 2D0?
Fig. 5: There is no description for a, b, e, and f. Please also add some more explanation to this figure caption.
645, 646: provide a value instead of using the term “on average”
647: “give consistent results” please add within what uncertainty.
Technical comments
40: Please check Ln/Tn throughout the text, sometimes it is referred to as LnTn.
68: add a comma after references
70: use a comma after Clarke et al. (2003)
74, 75: It feels like this sentence is not complete.
87-89: This sentence should be cut from here, so that “Despite these advantages,…” can be written exactly after the sentence ending in Line 87.
112: Jojosi donga: please check it throughout the text, sometimes it is written with the capital: Donga
122: c.f.: please check it throughout the text (cf., c.f.)
123: Delete “to” before following
146: use a comma after “From this fraction”
165: use a comma after “measurements”
171: use a comma after “tests”
Fig. 2 caption: omit “than” after “>”
213: omit “around”
241: add ‘the’ before “luminescent grains”
241: divided into
246: Omit “2.1 ± 0.4 wt%”
276: add ‘dose’ before “remaining”
445: Add ‘are’ after “CAM doses”
454: less than…
492: add ‘and’ before “the dose model”
540: Ln/Tn instead
567: Fig. S7a
586: Add ‘multi-grain’ before “aliquot”
605: Replace “by” with ‘of’: of Clarke et al.
606: the IRSL50 ages in Clarke et al.
615: Add ‘the’ before “Middle Pleistocene”
623 and 626: Jojosi 1 or Jojosi-1, please check throughout the text
624: use a comma after “For Jojosi 5”
639: use a comma after “the SGC”
642: CAM and ADM equivalent dose distributions results
645: use ‘while’ instead of “with these two methods”
Citation: https://doi.org/10.5194/gchron-2024-19-RC2
Data sets
A direct comparison of single grain and multi-grain aliquot measurements of feldspars from colluvial deposits in KwaZulu-Natal, South Africa Svenja Riedesel, Guillaume Guérin, Kristina J. Thomsen, Mariana Sontag-González, Matthias Blessing, Greg A. Botha, Max Hellers, Gunther Möller, Andreas Peffeköver, Christian Sommer, Anja Zander, and Manuel Will https://zenodo.org/doi/10.5281/zenodo.12759292
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 292 | 76 | 61 | 429 | 24 | 5 | 5 |
- HTML: 292
- PDF: 76
- XML: 61
- Total: 429
- Supplement: 24
- BibTeX: 5
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1