the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 17 Feb 2025
Diverse Causes of Extreme Rainfall in November 2023 over Equatorial Africa
Abstract. Understanding the atmospheric factors that lead to extreme rainfall events is essential to improve climate forecasting. This study aims to diagnose the physical processes underlying the extreme rainfall event of November 2023 in Equatorial Africa (EA), using the ERA5 reanalysis dataset. Composite, spatio-temporal and correlation analyses are used to shed light on the relationship between the November 2023 extreme precipitation events and the various associated factors. The analysis reveals that these extreme rainfall were mainly controlled by several factors that occurred during this period in the Pacific, Atlantic and Indian oceans. These factors include strong Sea-Surface-Temperature (SST) anomalies in Niño-3.4, North Tropical Atlantic, Equatorial Atlantic and Indian Ocean Dipole (IOD) oceanic regions, changes in zonal winds, the Walker circulation, the anomalous moisture flux and its divergence, the easterly jets and the activity of the Madden-Julian Oscillation (MJO). This convergence of moisture flows entered the EA region through its western and eastern boundaries, coming from the equatorial Atlantic and Indian oceans respectively. The juxtaposition of these factors has led to strong and positive rainfall anomalies in EA, with the highest values over the East African region, mainly over southern Ethiopia, Somalia, Kenya and Tanzania, which received more than 430 mm of rainfall during this month. Our findings suggest that many dynamic atmospheric effects need to be taken into account jointly to anticipate this type of extreme event. The results of the present study contribute to the improvement of sub-seasonal to seasonal rainfall forecasts by the region's national meteorological services, to enable us to increase the resilience of the region's citizens to these extreme weather conditions.
- Preprint
(3591 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 15 Apr 2025)
-
RC1: 'Comment on egusphere-2025-76', Anonymous Referee #1, 24 Feb 2025
reply
SOME GENERAL COMMENTS
Rainfall over Africa should not be evaluated using a reanalysis product. There are plenty of good observational and satellite products. It is well known that reanalysis products do not adequately represent rainfall in this region, unless the rainfall presented in the reanalysis is merely from an observational or satellite source. A recent article by Lavers et al. shows that ERA5 cannot get maximum precipitation right. Since the authors do use reanalysis rainfall, they need to find out more about the rainfall product in ERA5 and discuss this in the manuscript. They should also find an article (I think one exists) discussing how well ERA5 performs over Africa.
There is no monsoon over East Africa. There are some studies that suggest there is, but only monsoonal wind shift is in the two dry seasons. The only monsoonal area is West Africa. This should be removed from line 82 and also the caption of Fig. 11a, and wherever else it appears.The article omits what this reviewer considers to be pivotal articles on variability of the short rains, Hastenrath et al. 2010 and 2011. These clearly explain the importance of the low-level wind anomalies seen in Fig. 6.
The recent paper by Herrnegger et al. (2024) discussing the flooding in 2023 should also be added when discussing the rainfall anomalies.
A LOT OF SMALL ISSUES
Line 115 - Nicholson (2015) demonstrated that the IOD, ENSO, and zonal winds all play a role; did not state that increased rainfall is due to the presence of the IOD. This whole discussion is confusing. All three of those factors play a role. They major occur jointly, but each alone can also produce increased rainfall.
Line 170 the word "more" should be replaced by "additional"
Line 189 Perhaps I missed it, but I don't think CB cell has been defined.
Fig. 1a and d -this calculation must be off. November cannot possibly supply more than 30% of annual rainfall over Kenya and southern Somalia.
2.1 There is a units problem. SSTs should not be in K. It should be kg, not Kg. Surface pressure should be hPa, not Pa.
Line 242 This is very misleading. Rainfall was just below normal in 1992. Even 1983 I would not call a drought year.
Fig. 2. The authors need to look further at how ERA5 obtains SST data. Surely ERSST is incorporated into it, which would explain the similarities in a and b. It might just use ERSST. Again, as with rainfall, the authors should have used a bona fide SST data set.
Fig. 3 needs more in the caption. For example, what are the boxes? Indicate that the correlation is with November rainfall. This is clarified later on, but this should be put in the caption at this point.
The order in which things are discussed in the text is wrong. The sequence is 3c, 4, 3b, then 3a.
Line 290 The statement too strong. Although the anomalies are not significant, the dipole is clearly seen in the correlation patterns.
Fig. 4 The caption is wrong: b and c are not "as in a" because "a" is precipitation. Also, is DMI the DMI index. The caption implies it is SST averages over the DMI region, but that make no sense. DMI is calculated from two regions which generally have opposite SST anomalies.
In the discussion of Fig. 6, the Hastenrath papers really need to be included.Citation: https://doi.org/10.5194/egusphere-2025-76-RC1 -
RC2: 'Comment on egusphere-2025-76', Anonymous Referee #2, 21 Mar 2025
reply
The authors investigate atmospheric and ocean drivers of the Novermber 2023 extreme high rainfall month in equatorial Africa. They study a number of drivers previously studied in the literature in relation to the season, and place the 2023 event in the context of the climatology. The authors highlight in their introduction a number of direct and dramatic impacts arising from the extreme weather, and therefore justify the focus on this particular period.
The paper is well-written and provides a useful description of factors influencing an important extreme month of weather. The most persistent issue I had with this paper is the poor description of figure annotations. That is easily sorted, but I place the comment under “major” so the issue is clear. There are a few minor comments I have that either require small corrections or re-consideration by the authors. Once these are addressed, I can see the paper being publishable.
Major comments
Figs1, 3, 5,68, 9, 10 and 12 – all include boxes or lines annotating the figure, but which are not mentioned in the caption. Please explain what they are in the captions.
Minor comments
Sec2.1 - says you use ERA5 SST and ERSST for sea surface temperatures. Be clear throughout which you are using. I suggest you add to the caption of every relevant figure, unless there is a simple blanket statement you can make in the methods.
L189 – what is “CB cell”? It is not defined anywhere. Given it’s referred to a number of times, it might be worth annotating on an early plot.
L195 – Can you explicitly state if this is mean to be a scalar (as it is in the equation you present)? Often I would think V might be vector of u and v winds, with Q also being a vector with u and v components (or even z component too). Are you using “wind speed” magnitude for V? I.e. you are losing the direction information?
L201 – add the term “climatology” or similar to be clear about this. Also in the caption
L203 - “long-term mean (LTM) NOVEMBER rainfall” - “novermber” is needed.
L280-282 – You highlight the contradiction in the DMI term between the 2019 and 2023 response. However, I can’t see that you explain why they are different. If you cannot explain it, can you be explicit at this point in the text and say so. If you do explain it, can you give a brief mention to what the discerning factor was at this point in the text.
L334-334 - “These LLWs...” sentence. I’m not sure this sentence is entirely precise or correct. I’m not sure what you’re saying is cooling (subsiding air warms as it follows the dry adiabat). Is this just about land ocean temperature contrasts, and consequent changes in circulation. It sounds like your suggesting the circulation is causing the colder temperature over the ocean. Can you reconsider this sentence, and ensure your confident in it? Ideally, provide a reference which justifies the statement.
L375 - “...upper troposphere...” - Your plot only goes to 400hPa. I’d say 400hPa is borderline upper troposphere so I’m not sure it’s a fairly persuasive statement. I think you’d either need to show the anomalies make it to 300-200hPa, or just say they do stretch deep into the troposphere up to 400hPa.
Fig 8 caption – “Omega” has units hPa s-1. Vertical velocity, w, has units m s-1. Please be clear about which is being plotted here.
Fig8 cpation – what does the 10E-3 ms-1 refer to? Is it the value of the arrow on the figure panels? If so it must be the units of the horizontal winds too, not just vertical velocity? This needs clarity
L477 - “no vectors have entered...” - I’m not sure what you mean here. Whether arrows actually cross the box edges is an arbitrary choice of plotting position for the arrows. There clear are some meridional components to the winds along those 2 boundaries. I suggest you need to be precise about what you mean and quantify it. Do you mean no net meridional component along the boundary relative to the size of the net zonal component at the east and west boundaries?
L517 – I'm not sure the red contours would be called “southwest”. They look like “central” or “west central” to me, since the very southwest (10-20E, 6-10S) is white contours.
Technical comments
L149 – kgkg-1 should have small letter “k”
L345 - “JAS, The” needs to be a lower-case “t”
L462 – “tropospheric” to “troposphere”?
L506 - “confirmes” typo
L557 - “favorasing” typo
Citation: https://doi.org/10.5194/egusphere-2025-76-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 84 | 25 | 5 | 114 | 4 | 2 |
- HTML: 84
- PDF: 25
- XML: 5
- Total: 114
- BibTeX: 4
- EndNote: 2
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 50 | 35 |
| China | 2 | 18 | 12 |
| Cameroon | 3 | 16 | 11 |
| Germany | 4 | 8 | 5 |
| France | 5 | 5 | 3 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 50