the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 05 Mar 2024
Fifty-year seasonal variability of East African droughts and floods recorded in Central Afar lake sediments (Ethiopia) and their connections with ENSO
Abstract. Understanding past and present hydro-system feedbacks to global ocean-atmospheric interactions represents one of the main challenges to preventing droughts, extreme events and related human catastrophes in the face of global warming, especially in arid and semiarid environments. In eastern Africa, the El Niño-Southern Oscillation (ENSO) was identified as one of the primary drivers of precipitation variability affecting water availability. However, the northern East African Rift System (EARS) still suffers from ENSO climate teleconnection and the underrepresentation of predictive models because of the scarcity of local-to-regional historical or palaeo-data.
In this paper, we provide a 50-year seasonal flood/drought chronicle of the Awash River catchment from the study of laminated sediment from Gemeri and Afambo lakes (Central Afar region, Ethiopia), with the aim of reconstructing the magnitude of regional hydro-climatic events. Pluri-centimetric micro-laminated lithogenic facies alternating with pluri-millimetric carbonate-enriched facies are investigated in both lakes. We couple dating methods including radiocarbon, short-lived radionuclides, palaeomagnetic field variations and varve counting on both lake deposits to build a high-resolution age model and to discuss the regional hydro-sedimentary dynamics of the Awash River over the last ~700 years, with a focus on the last fifty years.
Using a multiproxy approach, we observe that following a multi-centennial enhanced hydrological period, the two lakes experienced a gradual decrease in river load inflow since 1979 CE, attaining extreme drought and high evaporative conditions between 1991 and 1997 CE. In 2014, the construction of a dam and the improvement of agricultural hydraulic management in the lower Awash River plain impacted the erodibility of local soils and the hydro-sedimentary balance of the lake basins, as evidenced by a disproportionate sediment accumulation rate.
Comparison of our quantitative reconstruction with i) lake water surface evolution expressed in Km2, ii) the interannual Awash River flow rates expressed in mm/yr, and iii) the El Niño 3.4 model highlights the intermittent connections between ENSO SST anomalies, regional droughts and hydrological conditions in the northern EARS.
- Preprint
(2917 KB) - Metadata XML
-
Supplement
(3234 KB) - BibTeX
- EndNote
Status: open (until 30 Apr 2024)
-
RC1: 'Comment on egusphere-2024-310', Anonymous Referee #1, 02 Apr 2024
reply
Revision
Fifty-year seasonal variability of East African droughts and floods recorded in Central Afar lake sediments (Ethiopia) and their connections with ENSO
Carlo Mologni, Marie Revel, Eric Chaumillon, Emmanuel Malet, Thibault Coulombier, Pierre Sabatier, Pierre Brigode, Gwenael Hervé, Anne-Lise Develle, Laure Schenini, Medhi Messous, Gourguen Davtian, Alain Carré, Delphine Bosch, Natacha Volto,
Clément Ménard, Lamya Khalidi, Fabien Arnaud
Dear Editor and Authors of the manuscript " Fifty-year seasonal variability of East African droughts and floods recorded in Central Afar lake sediments (Ethiopia) and their connections with ENSO",
Thank you for considering me to read this article and to be part of the revision process. Below, I am sending the revisions of the manuscript.
Sincerely.
General comments
In this manuscript, Mologni et al. provide a multiproxy study of two lakes in the East African Rift System that is under-represented in hydroclimate models. The authors combine radionuclides, paleomagnetic, and petrophysic parameters for characterizing the sediment sequences of lakes Gemeri and Afambo. They also gather instrumental data based on satellite images of the variation of the lake areas, as well as flow models and documentary information in the last 50 years. Considering the main criteria of the Journal Climate of The Past, I include the main comments of my revision, hoping they are constructive to improve the manuscript with major modifications.
Principal Criteria
Scientific significance.
The manuscript represents a novel hydroclimate contribution to an understudied area at Central Afar region, Ethiopia. Using classical concepts, the value of this work resides in the large amount of information about these sites provided for the last 50 years. Multiple-sourced data is obtained from two different lacustrine sediment sequencies, which is important to stress the local vs regional processes. Although the period corresponding to the last 50 years is sufficiently analyzed, the rest of the period (since 1300 CE) is only superficially discussed and I personally do not think they are conclusive, and perhaps they are only marginal to this work. I imagine they will be analyzed in detail in future works considering the extracted cores. Locally, the environmental significance of the hydroclimate variability in the region and locally is well developed, while the difference between the natural climate and anthropogenic changes are well represented.
Scientific quality.
The scientific approaches and methodologies applied are valid and adequate, they have been used and proved in several paleolimnological researches. Some methodological details are missing and should be provided (e.g. the model used for Pb-210 age models, see the specific comments, it is not clear if the lake area variations correspond to several images averaged within a year, the wet period, the dry period, see the specific comments). Although I think they can easily be included by the authors.
Presentation quality.
I consider that the presentation of the results and their structure must be improved before publication and this could take major work. The organization in the presentation of the results and figures should be revised in detailed and adjusted. As general examples, the results between both lake records are unbalanced (A), the criteria of the description are very different in both lakes. While one lake is described by facies, in the other lake facies are not mentioned, and authors use geochemical units, which are not applied for the first lake record. The figures are cited without following an order (B), they are mixed in the text. There is data in the Figures that is not discussed or included in the results (C). The figures in the supplementary material are not correctly explained (D), there are parts of the Figures that because of their size are not visible for the reader (E). The excel tables in the supplement do not present untis (F), etc. Further examples are indicated in the specific reviews. Therefore, I consider that the structure of the Results (and the supplementary data) require major modifications before publication.
In the Discussion section, there are important subjects such as the chronology, and the relationship between instrumental and paleolimnological records that are not included.
For instance, the authors include a large set of radiocarbon data, they present them in the results where they reject most of them without major analysis of the processes that might be causing errors in these data. In this case, if they decide that the radiocarbon data do not require a discussion, I do not see the point of presenting these results. Similarly, the authors do not discuss how the hiatus in lake Gemeri affected the chronological models.
In addition, the authors point in their conclusion (line 793) that the results have similarities and divergencies between instrumental data and the paleolimnological data. Nevertheless, the discussion about this subject is not provided.
Finally, I think the authors gathered a large high-resolution dataset provided by natural proxies (Ti/Sr, laminae thickness) superposed to instrumental information (lake area, SST, streamflow rate), which is sufficient to attempt a statistical or numerical comparison of both data that support the conclusions. This could add a special quantitative value to their work and would allow them to extend the data for the pre-instrumental period if longer cores were analyzed. A comparison that is restricted to descriptive observations (as it is now) of this enormous dataset, reduces the significance of the paleolimnological proxies in this work. In this sense, if there is direct hydro-climatic data enough to characterize the exact lake variability without the use of the paleolimnological proxies, then why should the latter be incorporated?
I explain the specific observations in the following comments.
MAIN MANUSCRIPT
Title: “and their connection with ENSO”. I think that this part of the title would be valuable once the authors have established a numerical connection with ENSO (see the comments below).
Abstract
Line 31-33: The laminated facies correspond only to one lake (L. Afambo), not both lakes.
Line 43: I don't think there is a quantitative reconstruction presented in this manuscript. Although there is quantitative data, the relationships are only relative (not quantitative).
Line 45: Please, indicate the meaning of the acronym.
Study site
Line 120-121: Can you enlarge the C) pannel of the figure? It is too small and the dots are difficult to discern. The Tendaho dam is not visible in the maps.
Materials and methods
Line 172-173: Please, indicate if the obtained lake area dataset is annual, monthly, seasonal, if there is missing data, etc.
Line 212-214: Were the values normalized beofre PCA? The figures in the Sup. Mat. are not clearly explained, some variables are missing. Please, provide a better description of the analysis.
Line 246-247: Please, indicate and justify the model selected for the Pb-210 chronology.
Line 264: Please, indicate how the counting was performed. Indicate also if varves were identified, why this term was no longer used in the text?
Results
Sedimentology and geochemistry results
Line 323: Which are the components of the couplets? F1/F2 or F1/F3, or both.
Line 338-339: There is a high correlation between XRF and ICP-MS from Sr and Ca. Why do you think the correlation of the Ti is lower with both techniques?
Line 348. The soft sediment deformation structures are not defined in the text.
Line 355. The cite of this Figure is not in the correct order. Please, check this for the rest of the Figures.
Line 357. I'm sorry, I don`t understand this. Are F2 and F3 well sorted only at those pointed levels (40 cm and 60 cm?), or is it a typing error and you meant micrometers instead of cm?
Line 358. Explain what F means in the Table.
Line 374. Please, justify the Log ratio use.
Line 374-380. I noticed that you used facies for the AF1802 core, but now you are using geochemical units for classifying this core. It is difficult to compare both cores, when their description and the sediment classification is different.
Line 382. I think the order of the age results should be reverted (otherwise, the table and figure order should be changed).
Line 383. I consider you should indicate and explain which model you applied for the Pb-210 derived ages, as well as the basis fot the application of this model.
Line 391. I consider you should indicate and explain which model you applied for the Pb-210 derived ages, as well as the basis for the application of this selected model.
Line 393. Please, add the core depth to the Table.
Line 397-398. Did you find evidence in your data? Is it the only explanation? Please, try to better explain these processes since there are many rejected samples, and accepted samples and rejected samples seems to be from different levels that are not directly comparable. This is an important aspect to consider for dating older sediments in future works. So, the fact that all the samples of bulk sediments are too old is a significant finding that should be highlighted. Can you see a pattern? How much older are they? You could try to explain this in the Discussion section.
Line 401-402. This is not quite clear. Although I see you calculated this rate, please provide the details (depths-ages). I think it is also not clear in the text, if the number of laminae correspond to couplets, or to a singular F1 lamina or F2 lamina.
Line 407. Figure 4. If you include in a.4 the rejected ages in the plot as a different triangle? Maybe you could see if they are related with moments of increased influxes of the river that could bring older material to the lake, as you proposed?
Line 412. Age model of Lake Gemeri. Even when I see that the authors included the hiatus in the image (Figure 5 a.4), it remains unclear how this hiatus was incorporated in the chronological models, and how it was defined in the lithology.
It is also not clearly expressed the difference of the Cs-137 control ages (1963 and 1955) from the Pb-210 CFCS derived model. In addition, since rates are indeed shown as changing throughout the core, including periods of no-deposition, please justify the application of the CFCS model.
Line 429. Figure 5. Please, incorporate the meaning of NWT and FF, RPI to the caption. The Figure a.3 shows that the Cs-137 maximum peak is not clearly discernable. The NWT letters are actually covering the curve and the sample point, but it seems that the maximum is actually the upper point (around 40 cm). Therefore, 1963 could be located between the point you marked (around 45 cm) and the point above it (40 cm). Please, include the age differences between the derived Pb-210 ages and the Cs-137 markers.
Could you also point the levels where C-14 was applied?
Line 476. The title is incomplete.
Line 484. Figure 6 and section 4.4. Please, specify the area of which lake is considered in the plot (light-blue). If both lakes are considered in this curve, you should indicate it. I think it would be more convenient to show each lake curve separately, since the description is different for both lakes, (e.g. lake Gemeri dried up). I can also notice there is a difference between the fluctuations of the sedimentological records and the lake area record. Please, explain this difference in the results and later in the discussion because the sediment response to the lake variability seems to be lagged. Otherwise, the chronological model might be the cause of this lack of synchronicity.
5- Discussion
Line 488. The chronology should be discussed in this section (it is partially discussed in the results but actually there is interpretation and ideas that deserves to be developed in a discussion. The C-14 data meaning should also be discussed. Otherwise, I do not understand why they are considered at all since the models are solely based on Pb-210 and Cs-137 and laminae counting.
Line 495. I cannot see the overlapping.
Line 507. How do the authors distinguish between total water (dissolved?) and solid load?
Line 518. In the text both facies, F2 and F3 are interpreted. In the subtitle, only F2 is mentioned. Please, add the F3 to the subtitle. Authors should also explain the interpretation for each of those facies, F2 and F3.
Line 533-534. “Indeed, Ca and Sr are related to intra lake precipitation of CaCO3 with Sr and Ca substitution” This sentence is not clear.
Line 540. I agree that Ca and Sr indicate drought stages. How is the intensity related to the element variation? Is this relationship relative, semiquantitative, does it occur from a certain lake level? Please, explain this relationship in the manuscript.
Line 562-563. Please, include the cites of the Figures.
Line 571. Sometimes the authors cite the Tables and Figures of the supplementary material with numbers S1, S2, etc. In this case, they used the letter E, that is also not in the right order (it follows the section J of the Supplementary material). Please, apply only one citation style.
Line 582. There is a contradiction between this paragraph and the following section where changes in the last 700 years are multicentennial.
Lines 593-595. Please, cite the Figures.
Line 599. Although the authors say that they explain the trends for the time-range 1300-1964 CE, they extend the temporal range in the following paragraph to 1979 CE. In addition, notice that there is a mix between AD and CE along the manuscript.
Line 605. Please, specify higher supplies of what...
Line 605. In this interpretation, the following periods are not explained or mentioned. Please, correct this.
Line 606. I'm sorry, I could not understand what time is "this time", and the connection of this interpretation with the data.
Line 611. Figure 8. I consider that sedimentation should be compared using the same criteria in this Figure. If the authors consider homopycnical sedimentation, they refer to an interpretation of the depositional processes. Instead, when they say laminated sedimentation, they refer only to the structure, without any interpretation. So, please, use the same criteria for this image.
At the b) panel, there is an example of the drought events (e.g. 1991-1997). Therefore, an example of the wet periods should also be mentioned for the pannel a).
In addition, the Pedogenic event is not well explained in the manuscript, there is a drought event but pedogenesis is not explained.
Line 631-632. What are the bases for this assumption? Please, explain them.
Line 639. I do not think high temperatures were explained before.
Line 589. Magnitudes of the Awash River wet/drought seasons, and their connection with the impacts of ENSO events. Considering you have a continuous and high-resolution record of the last 50 years, in one or both lakes, along with the images of lake area variability, and that you try to establish a relationship with ENSO, (El Niño 3.4), why are they not statistically compared? Regressions, a percentage of El Niño o La Niña episodes explained by the paleolimnological data, a correlation between data could be options.
Line 775-776. Please, discuss the similarities and divergencies of instrumental and the obtained paleolimnological records.
Conclusion.
Line 792-293. This should be further discussed in the manuscript because if the correlation of the indirect proxies and the direct proxies is not straightforward in some periods, then the indirect proxies are not entirely sensitive to changes in precipitation in the catchment, or other variables should be involved. Please, include this discussion considering the possible causes of the discrepancies.
SUPLEMENTARY MATERIAL
Figure S1. Please, include the coordinates of the image. The names of the coring sites are superposed and not visible in the image.
Figure S2. The letters are too small.
Figure S4. Please, explain the PCA, Dim1, Dim 2, the percentages, the variables. I cannot see Si in the PCA. Then, I see that S is also analyzed and not mentioned in the text, as well as Zr and Fe. Maybe you could mark the End-member 1 within the plot.
Figure S5. Please, check this paragraph, it is not clear. Neither are the colors. What is black? I think the legend is hidden. Explain this Figure.
Figure S6. Idem. Please, explain better this Figure, by indicating what is Dim 1, Dim 2, what is area, what is Sr.Ka, S.Ka, Mg.Ka. Maybe you could add a paragraph about the results of the three PCA plots.
- Paleomagnetic study on Lake Gemeri GEM18-03/04 core. I think the pedogenesis events and their evidences should be described in the main manuscript.
- Loss on ignition. Please, add the methodology and cite this in the manuscript. The decimals are separated by commas, you should change this. The weight, residual water, total depth (I assume it is the composite core depth) lack of units.
- Clay X-Ray Diffraction. The clay composition is not explained in the text, it is not used. If this methodology is applied, you should include the results of these analysis.
Figure S14. The image at the right of the excel sheet is not readable.
Figure S20. Please, include the units of the axes. Are they counts or clr? Please, include the coefficients that are mentioned in the manuscript, although they are not represented in these plots.
Figure S22. Authors mix AD and CE in this work, please, select one unit.
Figure S24 S25. I could not find this plot cited in the manuscript. Please, either explain these Figures or remove them.
- Stokes Law calculation of Lake Afambo sediments. This section (E) is preceded by section (J). Please, check the entire Supplementary section and the cites in the main manuscript.
-
RC2: 'Comment on egusphere-2024-310', Anonymous Referee #2, 05 Apr 2024
reply
General Comments:
This manuscript provides a very thorough analysis of sediment cores from two lakes in eastern Africa, a region that is highly susceptible to the adverse impacts of a changing climate as well as a region with sparse paleo data coverage. By combining multiple age models, XRF data, geochemical and microscopic analyses with satellite data and seismic imagery, they have produced interesting and important results.
There are a few things that should be done to strengthen this manuscript: performing the same analyses on all of the core sections so that we can directly compare the cores (e.g. Figures 4 & 5) and a reorganization to reduce the length of the manuscript and improve readability.
Given the importance of understanding this region and its sensitivities to regional and global climate, these revisions are well worth the time invested. I encourage you to revisit this manuscript. I look forward to seeing a revised version soon.
Specific Comments:
Line 27 - what is meant by “still suffers from ENSO climate teleconnection” – maybe something like “is impacted by ENSO climate teleconnections” is better?
Line 35 - ~700 years? The paper seems like it would be better to focus on the last ~50 years when you have the most data.
Line 43 – the units of these products is probably not necessary here.
Results – there are differences in the way that the cores were treated/analyzed. Is it possible to do all of the same analyses on all of the sections of cores?
There are a lot of great things in this paper but it is quite long. Are there some methods or results details that could be streamlined? Do these same processes apply to the modern times as well as the last ~700 years?
Figure 8 is very helpful for understanding a complex area – would it make sense to present it in the introduction as you are introducing the region?
Section 5.4 is really interesting and well-done. As you work to shorten and consolidate the text, this is an important piece to keep. These ideas, to me, are the big takeaways from this article!
The climate story back to 1300 CE seems a bit of an afterthought. Is it worth keeping in this manuscript?
Technical Corrections:
Abstract line 40 – I question whether hydraulic is the right word here.
Line 64 – “source of moisture fluxes from Atlantic or Indian Oceans”- this phrase seems out of place or missing a word or two.
Paragraph starting with line 89 – this should be a “wrap up” the introduction paragraph and much of it is. There are some new ideas presented here (human impacts, for example) that should either be removed or moved to another paragraph in the introduction and explored more fully.
Figure 1B – what is DEM?
Line 241 – please specify the short-lived radionuclides. I assume 210Pb, 226 Ra and 137Cs but it would be helpful to know which ones you are using.
Figure 1 uses A, B, C and Figure 2 uses a, b, c – either is fine but need to be consistent
Table 1 – Column 1 should be Depth
Line 324 – There is a “0” instead of “O” in SiO2 (also line 367)
Table 2 – there are a lot of rejected ages. Could this be improved? If not, please provide further discussion of why these ages were rejected or eliminate this part of the age model.
Figure 4 – Since you do not have Cs data prior to 1982, it is hard to know what the peak in Cs was.
Line 414 – 210Pb – the ex needs to be subscript
Line 534 and 545 – subscripts for CaCO3 and 210Pbex
Line 619 – chronicle?
Line 626 – D90 or Q90? Both are used in the text and figures. Which do you want to use?
Line 691 – the monsoon season should be referenced a bit more in the introduction. This is an important aspect of the climate system in this region.
Line 693 – are anomalous low and reduced rainy seasons different things or the same thing said twice?
Line 779 – define IOD
Citation: https://doi.org/10.5194/egusphere-2024-310-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 213 | 68 | 12 | 293 | 23 | 8 | 5 |
- HTML: 213
- PDF: 68
- XML: 12
- Total: 293
- Supplement: 23
- BibTeX: 8
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1