the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Characteristics of Heatwaves in Africa: Morocco 2000 and South Africa 2015/16
Abstract. Heatwaves pose an ever increasing risk to African communities as exposure to heat extremes can have a drastic effect on individuals and in some cases can even result in death. This study presents new information about the characteristics of historical African heatwaves including a comprehensive synopsis of documented heatwave events from 1980 until 2020.Detailed research on heatwave case studies helps to inform the development of early warning systems and forecasting, which is an urgent priority. Here, the focus is on two reported heatwaves, Morocco 2000 and South Africa 2015/16. Both heatwaves feature in the Emergency Events Disaster database (EM-DAT) and include reported impacts, with the Morocco heat being the only hazard to be associated with an economic cost. In addition, these heatwaves reveal how the mechanisms behind them are closely influenced by synoptic systems and geography of their regions. Further, It is demonstrated there is some reporting by African Nations for heatwaves but that this needs significant improvement.
- Preprint
(618 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on nhess-2021-242', Anonymous Referee #1, 28 Sep 2021
The work here presented has the potential for being a relevant and useful study. The subject under discussion is of extreme importance considering the current climate change context, the natural and socio-economic impacts linked to heatwaves, the demographic projections for Africa and, finally, the lack of analysis conducted for this continent, where the living conditions of most of the population are poor and climate change is far from being a priority for the policymakers. I believe the authors share this perception and have in mind the relevance that studies of this kind have.
However, the manuscript calls for major revisions and needs to be improved in many aspects. First, regarding the quality of writing, in many sections the text is confused, disconnected and with grammatical and syntax errors. The reader often struggles to understand the rational and the ideas that authors pretend to share. Therefore, I strongly recommend a thorough review of the English writing throughout all the paper.
Some of the methods and metrics adopted by the authors must be clarified. For instance, the UTCI index lacks a mathematical definition and a more detailed interpretation of its values. I also have some doubts regarding some of the methodologies that were adopted by authors to achieve the objectives that were proposed. In addition, some results are not consistent with the interpretation and discussion conducted throughout the manuscript text. I found even some results “unexpected” and very strange, calling for an urgent review. As a consequence of this, most of the conclusions made by the authors find a weak support in the results presented.
The manuscript needs major reviews. I ask the authors to kindly consider the following comments and suggestions:
Line 23: What is “Morocco heat”?
Line 26-27: Please clarify this sentence… I’m not sure what authors are trying to say here.
Line 32-33: These deaths occurred mostly during two particular mega-heatwaves events in Europe. Authors should detail the particularities of these two episodes (duration, severity of the induced heat-stress levels) and explain why they were responsible for such high mortality rates and other impacts in many natural and socio-economic sectors… This would reinforce the importance of analyze in detail specific historical events such as the ones studied in this work.
Lines 36 – 45: This paragraph is very important to show the relevance of this study. It represents a substantial contribution to the understanding of natural hazards and their consequences. The authors should detail how the poor living conditions of most of the African population are crucial to explain the high levels of heat-related mortality and how the fragile economies struggle to recover from heatwaves and other climate extremes – showing once again the relevance of this type of studies, particularly for the African continent.
Line 75: Authors should provide a mathematical definition for the UTCI index. They should also provide a thorough interpretation of what this index really means. What are the variables involved and what are the implications for humans when extreme high/low values of this index are recorded?
Line 93-97: Check the syntax of this sentence.
Line 98: I’m not sure what authors mean by “notably area”; check the syntax. Also, try to use a more robust and objective reason (severity, magnitude of the events, spatial extend……) to justify the decision of choosing these two events in particular… I’m not convinced. Also, “Reported Impacts”? Which ones?
Line 118-119: Poor English writing.
Line 122: Showing the coordinates of these regions only makes sense if the authors reference a figure….
Line 125-126: I’m confused. This is not clear. Rewrite this sentence please.
Figure 1 and Figure 2:
- The caption of the figure is very poor. Authors should provide more information about the variables and the figures that are presented: What are the panels showing? Anomalies? “Temp” is surface temperature?
- The location of South Africa (Morocco) within the African continent should be highlighted.
- I don’t like the layout (horizontal) of the figure… It’s not easy for the reader to follow the figure and the text at the same time. Try to order the panels in a vertical way.
- Try to save some space by removing the colorbars and the legends of the x-axis/y-axis that are repeated throughout the several panels.
- The bottom panel with the UTCI time series is supposed to be Figure 1d, isn’t it? Also, are these area averaged UTCI values for South Africa (Morocco in Figure 2)? This is not mentioned… Please clarify.
Line 157: I’m not sure about the need to show the Z850 anomalies considering that you are already showing the 500-hpa geopotential anomaly. If your idea is to prove that these heatwaves are linked to the establishment of a quasi-stationary anticyclonic system, the 500-hpa captures these structures well. Typically, in this type of analysis performed for mid-latitude regions, the 500hpa atmospheric level is enough to find these synoptic systems. In addition, it seems that authors are not extracting any relevant information from z850 field. My suggestion would be to go to a single panel where the 500-hpa geopotential anomaly would be represented by contours and the 850-hpa temperature anomaly by colors. The temperature at this level (just above the boundary layer where the diurnal cycle is almost negligible) would give you information about potential warm air masses that are being advected to the regions of interest and that could be heated due to subsidence. This would also allow the authors to save a panel on an already heavy figure.
Line 160 – 161: Is this true for Temp? Over South Africa, the anomalies are close to zero! I have doubts about the quality of these results. If the temperature values are within the expected values, we can’t consider this as a Heatwave. For instance, Pretoria region has even negatives anomalies of Temp (once again, what is Temp??)… In table 1 authors say the following: “Heatwave Concurrent with a drought and an ENSO event. Temperatures reached 42.7 °C in Pretoria and 38.9 °C in Johannesburg on 6 th January.” Something here is not right. I know that the heatwave definition used by authors is based on the UTCI values, but, isn’t it strange to have a Heatwave event with lower temperatures than the expected values? Once again, how are the UTCI values obtained? This needs a careful and urgent review!!
Line 165: It should be Z850… Once again, the authors are not extracting any relevant information from the Z850 anomaly field.
Line 166 – 167: Not sure about what the authors mean to say here. Please clarify
Line 190: Slight cool anomalies? According to the results, there are regions reaching values between -5ºC and -10ºC.
Line 196 – 198: Not sure about this… Morocco is still under the influence of high anomalies of Temp (reaching levels very close to 5ºC). Authors need to Check the data because the results are not consistent with what is discussed in the text.
Line 198 – 200: Please rewrite this sentence.
Lines 200 – 201: Please rewrite this sentence. Not sure about what authors are trying to say here.
Lines 232 – 234: Authors should include some possible explanations for this, for instance, the topography, the influence of more local meteorological processes, the influence of the Sahara Desert, different land covers….
Lines 234 – 236: Which pressure pattern? Authors were able to link heatwaves in South Africa and Morocco with geopotential anomalies. Here, in the discussion, they should detail what are the synoptic processes that are behind this link.
Lines 236 – 239: Authors should rewrite this sentence please.
Line 240: This would be the ideal paragraph to authors mention some of the limitations of this study.
Lines 259 – 260: Authors should rewrite this sentence please.
Lines 267 – 276. Authors should rewrite all the Conclusion paragraph. There are a lot of syntax and grammatical errors. The text is confused and disconnected. I’m not sure if authors were successful in covering the ideal bullets for the Conclusion section (Why is this study relevant? What were the key findings and what can we learn from them? Is this a novel approach for Africa? How could these results be used to predict future events? Should policymakers and authorities start to look at these climatic extreme with more attention? Basically, why this work should be publish and shared.
-
AC1: 'Reply on RC1', Chloe Brimicombe, 01 Nov 2021
The work here presented has the potential for being a relevant and useful study. The subject under discussion is of extreme importance considering the current climate change context, the natural and socio-economic impacts linked to heatwaves, the demographic projections for Africa and, finally, the lack of analysis conducted for this continent, where the living conditions of most of the population are poor and climate change is far from being a priority for the policymakers. I believe the authors share this perception and have in mind the relevance that studies of this kind have.
We thank the reviewer, our research does indeed work to address the under discussion of extreme heat and climate change.
However, the manuscript calls for major revisions and needs to be improved in many aspects. First, regarding the quality of writing, in many sections the text is confused, disconnected and with grammatical and syntax errors. The reader often struggles to understand the rational and the ideas that authors pretend to share. Therefore, I strongly recommend a thorough review of the English writing throughout all the paper.
Thank you for this valuable feedback. We will thoroughly review and revise the manuscript.
Some of the methods and metrics adopted by the authors must be clarified. For instance, the UTCI index lacks a mathematical definition and a more detailed interpretation of its values.
We will clarify the methods section. The UTCI is a robust index and has been previously used for African Nations please refer to citations in this paper: Guigma, et al. 2020
In addition, We will include the mathematical definition of the UTCI which is:
Where Ta= air temperature, Tr= mean radiant temperature, Va= wind speed Pa= water vapour pressure.
I also have some doubts regarding some of the methodologies that were adopted by authors to achieve the objectives that were proposed. In addition, some results are not consistent with the interpretation and discussion conducted throughout the manuscript text.
We have clarified our methods to ensure that the way in which they achieve the objectives is clear to the reader. We have also reviewed our results and find them to be consistent with our interpretation. We will address further comments on results below.
I found even some results “unexpected” and very strange, calling for an urgent review. As a consequence of this, most of the conclusions made by the authors find a weak support in the results presented.
We have clarified the result section to discuss why the results are not unexpected and to make clear how our conclusions are supported by the evidence.
The manuscript needs major reviews. I ask the authors to kindly consider the following comments and suggestions:
Line 23: What is “Morocco heat”?
We will change this to Morocco Heatwave.
Line 26-27: Please clarify this sentence… I’m not sure what authors are trying to say here.
We will change this to “Further it can be seen that in some African Nations heatwaves are reported but this needs improvement in order to capture impacts across the continent.”
Line 32-33: These deaths occurred mostly during two particular mega-heatwaves events in Europe. Authors should detail the particularities of these two episodes (duration, severity of the induced heat-stress levels) and explain why they were responsible for such high mortality rates and other impacts in many natural and socio-economic sectors… This would reinforce the importance of analyze in detail specific historical events such as the ones studied in this work.
We have a recent paper published showing that these mega-heatwaves were not just in Europe Brimicombe, et al., 2021. which will now make reference to. It shows that there were high heat stress levels during both the 2003 and 2010 August Heatwaves in parts of the African continent and that the impacts of heatwaves are under-reported.
Lines 36 – 45: This paragraph is very important to show the relevance of this study. It represents a substantial contribution to the understanding of natural hazards and their consequences. The authors should detail how the poor living conditions of most of the African population are crucial to explain the high levels of heat-related mortality and how the fragile economies struggle to recover from heatwaves and other climate extremes – showing once again the relevance of this type of studies, particularly for the African continent.
We agree with the reviewer entirely and will add this to the manuscript as suggest. We will also include a nod to how building material can increase the indoor temperature. However, as noted in this paragraph there is sparse recorded evidence for heatwave impacts across Africa and therefore a lack of evidence recorded of economic struggles.
Line 75: Authors should provide a mathematical definition for the UTCI index. They should also provide a thorough interpretation of what this index really means. What are the variables involved and what are the implications for humans when extreme high/low values of this index are recorded?
We have included the mathematical definition of the UTCI which is:
Where Ta= air temperature, Tr= mean radiant temperature, Va= wind speed Pa= water vapour pressure.
In addition to the results section: The UTCI is a thermal index which makes use of the meteorological parameters of 2m temperature, water vapour pressure, 10m wind speed and mean radiant temperature and a body model (Di Napoli et al 2021). It has been compared to many other thermal indices such as apparent temperature and heat index and captures well an average body response to the thermal environment (Zare et al 2018, Jendritzky et al 2012, Blazejczyk et al 2012). It has further been shown to be able to forecast heatwaves internationally (Pappenberger et al 2015) and accurately indicate extreme heat for Africa (Guigma et al 2020).
Line 93-97: Check the syntax of this sentence.
Thank you we have checked the syntax
Line 98: I’m not sure what authors mean by “notably area”; check the syntax. Also, try to use a more robust and objective reason (severity, magnitude of the events, spatial extend……) to justify the decision of choosing these two events in particular… I’m not convinced. Also, “Reported Impacts”? Which ones?
We will clarify the reason for choosing these event and thank you we will change area to spatial extent. We also have added a reference to Table 1 for reported impacts.
Line 118-119: Poor English writing.
We will simplify the sentence structure.
Academic Literature and reports show that 39 heatwaves have been reported since 1980 in Africa.
Line 122: Showing the coordinates of these regions only makes sense if the authors reference a figure….
We have asked a range of colleagues and they find the longitude and latitude references useful to understand where in Africa each region is.
Line 125-126: I’m confused. This is not clear. Rewrite this sentence please.
We will change this to: “Characteristics of heatwaves (n=26) are reported more than their impacts (n=9).”
• The caption of the figure is very poor. Authors should provide more information about the variables and the figures that are presented: What are the panels showing? Anomalies? “Temp” is surface temperature?
We will describe each figure in more detail.
• The location of South Africa (Morocco) within the African continent should be highlighted.
We agree and will add a box highlighting
• I don’t like the layout (horizontal) of the figure… It’s not easy for the reader to follow the figure and the text at the same time. Try to order the panels in a vertical way.
We agree but this was necessary because of the format of NHESS peer-review system.
• Try to save some space by removing the colorbars and the legends of the x-axis/y-axis that are repeated throughout the several panels.
Our colleagues have found it useful to have a scale on each figure.
• The bottom panel with the UTCI time series is supposed to be Figure 1d, isn’t it? Also, are these area averaged UTCI values for South Africa (Morocco in Figure 2)? This is not mentioned… Please clarify.
We will clarify this in the figure description please see as above.
Line 157: I’m not sure about the need to show the Z850 anomalies considering that you are already showing the 500-hpa geopotential anomaly.
It has previously been shown that the Saharan Heat Low is best captured by between 700 and 900hpa values, a higher height than 500hpa (Lavaysse et al 2016). We have clarified this in the text.
Line 160 – 161: Is this true for Temp? Over South Africa, the anomalies are close to zero! I have doubts about the quality of these results. If the temperature values are within the expected values, we can’t consider this as a Heatwave. For instance, Pretoria region has even negatives anomalies of Temp (once again, what is Temp??)… In table 1 authors say the following: “Heatwave Concurrent with a drought and an ENSO event. Temperatures reached 42.7 °C in Pretoria and 38.9 °C in Johannesburg on 6 th January.” Something here is not right. I know that the heatwave definition used by authors is based on the UTCI values, but, isn’t it strange to have a Heatwave event with lower temperatures than the expected values? Once again, how are the UTCI values obtained? This needs a careful and urgent review!!
The figure is showing over the part of South Africa where the heatwave is present anomalies in temperature (Temp) are up to 2C hotter than the climate. We will clarify this in the text.
The UTCI is not the same as temperature, please refer to the methods section. It is a model of a body response to the thermal environment. It assess how temperature, mean radiant temperature, water vapour pressure and wind speed effect an average body and indicates heat stress (Di Napoli, et al., 2018,2019,2021).
In the figure the parts of South Africa experiencing a heatwave are on the southern coast including over cape town. The anomalies are present at up to 2C hotter in Temperature for both Pretoria and Johannesburg.
ERA5 data as a reanalysis dataset will not give the same values as recorded observed temperature and we will also clarify this in the methods section. ERA5 however has been found to accurately identify heatwaves despite not having values that match the maximum temperature recorded (Di Napoli, et al 2018, Brimicombe et al 2021,Hersbach et al 2020).
Line 165: It should be Z850… Once again, the authors are not extracting any relevant information from the Z850 anomaly field.
Thank you we have corrected this typo
Line 166 – 167: Not sure about what the authors mean to say here. Please clarify
We will re-write this to clarify:
Considering the drop in temperature and UTCI anomalies it can be inferred that this low pressure has a cooling influence on the region.
Line 190: Slight cool anomalies? According to the results, there are regions reaching values between -5ºC and -10ºC.
Thank you this is a typo we have clarified that anomalies are between -5C and -10C
Line 196 – 198: Not sure about this… Morocco is still under the influence of high anomalies of Temp (reaching levels very close to 5ºC). Authors need to Check the data because the results are not consistent with what is discussed in the text.
We will add a box to the figure to aid the comparison of the written text and figure. The results are consistent with what is discussed in the text.
Line 198 – 200: Please rewrite this sentence.
We will restructure this sentence ‘the z850 and z500 negative geopotential anomalies to the east of Morocco indicate a low pressure system which could be in part having a cooling influence on the region.’
Lines 200 – 201: Please rewrite this sentence. Not sure about what authors are trying to say here.
We will restructure the sentence.
‘The peak of the UTCI and temperature are similar during the Morocco 2000 and South Africa 2015/16 Heatwave, with anomalies of up 2°C respectively.’
Lines 232 – 234: Authors should include some possible explanations for this, for instance, the topography, the influence of more local meteorological processes, the influence of the Sahara Desert, different land covers….
Yes we agree and have expanded this section to include information about the desert winds that influence the temperature regime of Morocco (Filahi, et al 2016) As well as discussing the position of the Saharan Heat Low which also influences the air flows for Northern Africa (Lavaysse et al 2016).
Lines 234 – 236: Which pressure pattern? Authors were able to link heatwaves in South Africa and Morocco with geopotential anomalies. Here, in the discussion, they should detail what are the synoptic processes that are behind this link.
Thank you we have clarified that this is a high pressure system.
Lines 236 – 239: Authors should rewrite this sentence please.
We will rewrite this as follows: Heatwaves drivers are closely linked to the climatology and synoptic systems in a region for both of the heatwaves examples
Line 240: This would be the ideal paragraph to authors mention some of the limitations of this study.
We agree and have clarified this paragraph’s aim:
Further work should consider more fully the meteorology of the heatwave this study has more of a focus on extreme heat impacts, such as heat stress.
Lines 259 – 260: Authors should rewrite this sentence please.
We will rewrite the following as African heatwaves have a lack of observations, under-reporting fuels a lack of evidence and a patchy research field, intrinsic factors that has a growing mandate to be addressed globally
Lines 267 – 276. Authors should rewrite all the Conclusion paragraph. There are a lot of syntax and grammatical errors. The text is confused and disconnected. I’m not sure if authors were successful in covering the ideal bullets for the Conclusion section (Why is this study relevant? What were the key findings and what can we learn from them? Is this a novel approach for Africa? How could these results be used to predict future events? Should policymakers and authorities start to look at these climatic extreme with more attention? Basically, why this work should be publish and shared.
We agree here’s how we would approach your questions in the conclusion.
Why is this study relevant?
This study is relevant because it highlights extreme heat risk for a continent where this research is sparse and where the impacts to heatwaves are ever increasing.
What were the key findings and what can we learn from them?
The key finding is that reporting although patchy for the continent is not captured in international databases, even in retrospect. And that heatwave characteristics although a rise in temperature and heat stress are diverse when taking into account the local climate and geography of a region. This means that specialist information should be provided to nations weather services to suite.
Is this a novel approach for Africa?
This will be the first peer-reviewed systematic list of heatwaves including recorded physical characteristics and impacts for the continent. This is also the first time that the Universal Thermal Climate Index has been used at a continental scale to assess heat stress during a heatwave in Africa.
How could these results be used to predict future events?
These results can be used to improve nations understanding of the characteristics and impacts of heatwaves, in the short term. In addition, the methods can be adapted and applied to climate projections to look into future risk.
Should policymakers and authorities start to look at these climatic extreme with more attention?
The main finding is that policy-makers and authorities should put in place adaptations to address extreme heat as a growing risk for their countries. And our results demonstrate heat is not just a future risk but a real risk with devastating impacts now.
Citation: https://doi.org/10.5194/nhess-2021-242-AC1
-
RC2: 'Comment on nhess-2021-242', Anonymous Referee #2, 28 Sep 2021
General Comments
This paper discusses heat waves in Africa, and provides a list of African heat wave events compiled from peer reviewed literature, as well as grey literature and media reports. Case studies of two heatwaves is provided, with discussion of air temperature, UTCI, and pressure systems.
This paper is important because the impacts of heatwaves in sub-Saharan Africa are possibly under-reported.
Before publication, the method of the literature search needs to be more thoroughly described.
There are a few sentences that are confusing. The text is well organized and about the right length. It refers to appropriate related literature.
It is not explained how the 21 academic papers and grey literature where arrived at. Typically, if a paper relies on a systematic search of the literature, then the search engine and search terms used should be give. The reader is not able to judge whether the search was systematic or totally ad-hoc, and therefore whether or not the search is exhaustive.
The only rows in the table that have human impacts have a reference to CRED. Are all of the human impacts from CRED? Do none of the reports from other sources mention human impacts? If so, this should be highlighted.
Page 14 line 236. "This is the first time heatwaves from two different African regions have been presented and compared using both their physical characteristics and reported impacts" The impacts are barely discussed in this paper. They are stated in the table, but the information comes directly from EM-DAT and no further information is added. No comparison is made between the impacts of the heatwaves.
Page 11, figure 1. During the heatwave, it does not appear that temperatures are really higher than usual in South Africa, whereas there is a high temperature anomaly over Namibia. Perhaps this is something to do with the resolution of the image, or the choice of color scale. Actual values for the temperature anomaly during the heatwave are not given in the text, and the reader is left wondering if temperatures were actually anomalously high during this period according to the reanalysis.
Is there any reporting for this time period in Namibia? It might be good to point out the discrepancy in reporting.
The figures are quite hard to view in the format given. It is difficult to compare different time steps in the same variable as they are not adjacent. The maps are tiny, so appear low resolution in the PDF version.
The word "mandate" is used a few times to mean a motivation for further study - I am not sure this is a correct use of the word.
On page 3, line 66 the phrase "between 1980 and 2020 to date" is confusing. Does the author mean 2020 up to the date at which they are writing? They should instead say when that date was specifically.
EM-DAT is first mentioned in section 2.1, but is not defined or referenced.
On page 4, line 90. "It [UTCI] has further been shown to be able to forecast heatwaves internationally (Pappenberger et al 2015)". The phrasing suggests that UTCI in isolation is able to create international forecasts. The referenced paper retrospectively shows that a NWP forecast of UTCI could have been used to forecast the 2010 Russia heatwave. The point is that UTCI forecasts can be used, not that UTCI produces forecasts.
Page 5, line 118. Hard to read sentence. "Academic Literature and reports for Africa show that 39 heatwaves are reported for somewhere in Africa almost every year since 1980." This reads as if each year 39 heatwaves are reported somewhere.
Page 8, in the table. "Was the warmest April to date at the time in the Sahel" is in the "Heatwave Impacts" column, it should be in the "Heatwave Characteristics" column.
Page 12, line 205. The phrase "heat stress" is used instead of "UTCI" making it unclear what is being measured.
Page 14 line 222, "suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa". These other studies are not suggesting that EM-DAT is not accurate, rather that it is not complete. Harrington & Otto make no mention of accuracy.
Page 14 line 222. "our study supports others (i.e. Harrington and Otto, 2020; van der Walt and Fitchett , 2021) suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa including sub-Saharan Africa only including 7 out of our 39 listed heatwaves in their records, which is less than 20%." This sentence is hard to read. Break it into two sentences.
Page 14, line 229. "However, there are some heatwave warning systems (Hafez and Almazroui 2016, Boubaker
230 2010)" Are these systems another potential source for identifying heatwaves? Were the heatwaves in your study identified by these systems?
Page 14, line 225 to 228 repeats points already made in lines 221-224.
Page 14, line 231. "and there is some reporting in place for African Nations that is not always captured at an international scale" what is meant here by the "international scale"? Does it mean the international databases referenced above?
Page 15, line 250 - 253. The evidence provided on lines 250-252 does not directly support the hypothesis on line 253. Comparing an air temperature mortality effect with a UTCI anomaly in this way is misleading. We can see from figure 1 that the UTCI anomaly was large during the South Africa heatwave, but we cannot see that the air temperature anomaly is large. You should state what the air temperature anomaly was. This probably still leads to more deaths than the 11 reported.
Citation: https://doi.org/10.5194/nhess-2021-242-RC2 -
AC2: 'Reply on RC2', Chloe Brimicombe, 01 Nov 2021
This paper discusses heat waves in Africa, and provides a list of African heat wave events compiled from peer reviewed literature, as well as grey literature and media reports. Case studies of two heatwaves is provided, with discussion of air temperature, UTCI, and pressure systems.
Thank you for your comments.
This paper is important because the impacts of heatwaves in sub-Saharan Africa are possibly under-reported.
Yes however sub-Saharan African Heatwaves are under-reported.
Before publication, the method of the literature search needs to be more thoroughly described.
We agree and will clarify in the methods: . A systematic search was carried out using google scholar whereby chosen literature had to meet the criteria of having a focus on historic heatwaves in Africa and include either a physical characteristic of the heatwave or impact.
There are a few sentences that are confusing. The text is well organized and about the right length. It refers to appropriate related literature.
Thank you, all co-authors have read the paper again to avoid confusing sentence structure.
It is not explained how the 21 academic papers and grey literature where arrived at. Typically, if a paper relies on a systematic search of the literature, then the search engine and search terms used should be give. The reader is not able to judge whether the search was systematic or totally ad-hoc, and therefore whether or not the search is exhaustive.
As above we will clarify this.
The only rows in the table that have human impacts have a reference to CRED. Are all of the human impacts from CRED? Do none of the reports from other sources mention human impacts? If so, this should be highlighted.
They were reports that supported the evidence of impacts in em-dat and they are also referenced.
Page 14 line 236. "This is the first time heatwaves from two different African regions have been presented and compared using both their physical characteristics and reported impacts" The impacts are barely discussed in this paper. They are stated in the table, but the information comes directly from EM-DAT and no further information is added. No comparison is made between the impacts of the heatwaves.
We will add a discussion comparing the reported impacts.
Considering that our analysis shows positive anomalies in air temperature of up to 2°C and heat stress indicated by the UTCI of up to 4°C, we hypothesise that impacts of the South Africa heatwaves have been underreported. For the Morocco heatwave the reported impacts are 4 million chickens dying leading to up to 809,000 USD of damages (table 1). Interestingly, this is the only heatwave with an economic loss associated with it. This reiterates the complexities of identify loss and damage due to heat extremes, especially for Africa where research on heat-related impacts is still limited (Campbell et al 2018).
Page 11, figure 1. During the heatwave, it does not appear that temperatures are really higher than usual in South Africa, whereas there is a high temperature anomaly over Namibia. Perhaps this is something to do with the resolution of the image, or the choice of color scale. Actual values for the temperature anomaly during the heatwave are not given in the text, and the reader is left wondering if temperatures were actually anomalously high during this period according to the reanalysis.
There is a temperature anomaly of up to 2C across parts of South Africa. We will clarify also that reanalysis will not give the same values as those reported.
Is there any reporting for this time period in Namibia? It might be good to point out the discrepancy in reporting.
Namibia is a region that we would like to look into further it is not the focus of heatwave studies or any reports we found.
The figures are quite hard to view in the format given. It is difficult to compare different time steps in the same variable as they are not adjacent. The maps are tiny, so appear low resolution in the PDF version.
This was due to NHESS peer-review requirements.
The word "mandate" is used a few times to mean a motivation for further study - I am not sure this is a correct use of the word.
We have reviewed the use of the word.
On page 3, line 66 the phrase "between 1980 and 2020 to date" is confusing. Does the author mean 2020 up to the date at which they are writing? They should instead say when that date was specifically.
We will clarfiy this to be up to the end of 2020.
EM-DAT is first mentioned in section 2.1, but is not defined or referenced.
Thank you we have now referenced EM-DAT.
On page 4, line 90. "It [UTCI] has further been shown to be able to forecast heatwaves internationally (Pappenberger et al 2015)". The phrasing suggests that UTCI in isolation is able to create international forecasts. The referenced paper retrospectively shows that a NWP forecast of UTCI could have been used to forecast the 2010 Russia heatwave. The point is that UTCI forecasts can be used, not that UTCI produces forecasts.
Thank you we have changed the phrasing:
It has further been shown that the UTCI as a thermal index can be used to forecast heatwaves internationally (Pappenberger et al 2015)
Page 5, line 118. Hard to read sentence. "Academic Literature and reports for Africa show that 39 heatwaves are reported for somewhere in Africa almost every year since 1980." This reads as if each year 39 heatwaves are reported somewhere.
We have re-worded this sentence:
Academic Literature and reports show that 39 heatwaves have been reported since 1980 in Africa.
Page 8, in the table. "Was the warmest April to date at the time in the Sahel" is in the "Heatwave Impacts" column, it should be in the "Heatwave Characteristics" column.
Thank you this is a typo we have moved this to the correct column.
Page 12, line 205. The phrase "heat stress" is used instead of "UTCI" making it unclear what is being measured.
We have changed this to be UTCI.
Page 14 line 222, "suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa". These other studies are not suggesting that EM-DAT is not accurate, rather that it is not complete. Harrington & Otto make no mention of accuracy.
We have changed to in-complete instead of accuracy to clarify our meaning.
Page 14 line 222. "our study supports others (i.e. Harrington and Otto, 2020; van der Walt and Fitchett , 2021) suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa including sub-Saharan Africa only including 7 out of our 39 listed heatwaves in their records, which is less than 20%." This sentence is hard to read. Break it into two sentences.
We agree: Further, our study supports others (i.e. Harrington and Otto, 2020; van der Walt and Fitchett , 2021) suggesting that international databases such as EM-DAT (CRED 2020) are not recording all heatwaves for Africa. For Example, sub-Saharan Africa only including 7 out of our 39 listed heatwaves in their records, which is less than 20%. However, it’s important to acknowledge there are some heatwave warning systems (Hafez and Almazroui 2016, Boubaker 2010) and there is some reporting in place for African Nations that is not always captured at an international scale within reports or databases (Table 1).
Page 14, line 229. "However, there are some heatwave warning systems (Hafez and Almazroui 2016, Boubaker
230 2010)" Are these systems another potential source for identifying heatwaves? Were the heatwaves in your study identified by these systems?
Yes – Tunisia and Egypt. We have included this in the text (i.e. Tunisia and Egypt)
Page 14, line 225 to 228 repeats points already made in lines 221-224.
Thank you, we have removed this.
Page 14, line 231. "and there is some reporting in place for African Nations that is not always captured at an international scale" what is meant here by the "international scale"? Does it mean the international databases referenced above?
We mean International Databases and Reports we have now clarified this.
Page 15, line 250 - 253. The evidence provided on lines 250-252 does not directly support the hypothesis on line 253. Comparing an air temperature mortality effect with a UTCI anomaly in this way is misleading. We can see from figure 1 that the UTCI anomaly was large during the South Africa heatwave, but we cannot see that the air temperature anomaly is large. You should state what the air temperature anomaly was. This probably still leads to more deaths than the 11 reported.
We have clarified this so that it is not misleading. As above there is an air temperature anomaly for south Africa. In addition UTCI has been shown to cause a rise in mortality.
Considering that our analysis shows positive anomalies in air temperature of up to 2°C and heat stress indicated by the UTCI of up to 4°C, we hypothesise that impacts of the South Africa heatwaves have been underreported.
Citation: https://doi.org/10.5194/nhess-2021-242-AC2
-
AC2: 'Reply on RC2', Chloe Brimicombe, 01 Nov 2021
Status: closed
-
RC1: 'Comment on nhess-2021-242', Anonymous Referee #1, 28 Sep 2021
The work here presented has the potential for being a relevant and useful study. The subject under discussion is of extreme importance considering the current climate change context, the natural and socio-economic impacts linked to heatwaves, the demographic projections for Africa and, finally, the lack of analysis conducted for this continent, where the living conditions of most of the population are poor and climate change is far from being a priority for the policymakers. I believe the authors share this perception and have in mind the relevance that studies of this kind have.
However, the manuscript calls for major revisions and needs to be improved in many aspects. First, regarding the quality of writing, in many sections the text is confused, disconnected and with grammatical and syntax errors. The reader often struggles to understand the rational and the ideas that authors pretend to share. Therefore, I strongly recommend a thorough review of the English writing throughout all the paper.
Some of the methods and metrics adopted by the authors must be clarified. For instance, the UTCI index lacks a mathematical definition and a more detailed interpretation of its values. I also have some doubts regarding some of the methodologies that were adopted by authors to achieve the objectives that were proposed. In addition, some results are not consistent with the interpretation and discussion conducted throughout the manuscript text. I found even some results “unexpected” and very strange, calling for an urgent review. As a consequence of this, most of the conclusions made by the authors find a weak support in the results presented.
The manuscript needs major reviews. I ask the authors to kindly consider the following comments and suggestions:
Line 23: What is “Morocco heat”?
Line 26-27: Please clarify this sentence… I’m not sure what authors are trying to say here.
Line 32-33: These deaths occurred mostly during two particular mega-heatwaves events in Europe. Authors should detail the particularities of these two episodes (duration, severity of the induced heat-stress levels) and explain why they were responsible for such high mortality rates and other impacts in many natural and socio-economic sectors… This would reinforce the importance of analyze in detail specific historical events such as the ones studied in this work.
Lines 36 – 45: This paragraph is very important to show the relevance of this study. It represents a substantial contribution to the understanding of natural hazards and their consequences. The authors should detail how the poor living conditions of most of the African population are crucial to explain the high levels of heat-related mortality and how the fragile economies struggle to recover from heatwaves and other climate extremes – showing once again the relevance of this type of studies, particularly for the African continent.
Line 75: Authors should provide a mathematical definition for the UTCI index. They should also provide a thorough interpretation of what this index really means. What are the variables involved and what are the implications for humans when extreme high/low values of this index are recorded?
Line 93-97: Check the syntax of this sentence.
Line 98: I’m not sure what authors mean by “notably area”; check the syntax. Also, try to use a more robust and objective reason (severity, magnitude of the events, spatial extend……) to justify the decision of choosing these two events in particular… I’m not convinced. Also, “Reported Impacts”? Which ones?
Line 118-119: Poor English writing.
Line 122: Showing the coordinates of these regions only makes sense if the authors reference a figure….
Line 125-126: I’m confused. This is not clear. Rewrite this sentence please.
Figure 1 and Figure 2:
- The caption of the figure is very poor. Authors should provide more information about the variables and the figures that are presented: What are the panels showing? Anomalies? “Temp” is surface temperature?
- The location of South Africa (Morocco) within the African continent should be highlighted.
- I don’t like the layout (horizontal) of the figure… It’s not easy for the reader to follow the figure and the text at the same time. Try to order the panels in a vertical way.
- Try to save some space by removing the colorbars and the legends of the x-axis/y-axis that are repeated throughout the several panels.
- The bottom panel with the UTCI time series is supposed to be Figure 1d, isn’t it? Also, are these area averaged UTCI values for South Africa (Morocco in Figure 2)? This is not mentioned… Please clarify.
Line 157: I’m not sure about the need to show the Z850 anomalies considering that you are already showing the 500-hpa geopotential anomaly. If your idea is to prove that these heatwaves are linked to the establishment of a quasi-stationary anticyclonic system, the 500-hpa captures these structures well. Typically, in this type of analysis performed for mid-latitude regions, the 500hpa atmospheric level is enough to find these synoptic systems. In addition, it seems that authors are not extracting any relevant information from z850 field. My suggestion would be to go to a single panel where the 500-hpa geopotential anomaly would be represented by contours and the 850-hpa temperature anomaly by colors. The temperature at this level (just above the boundary layer where the diurnal cycle is almost negligible) would give you information about potential warm air masses that are being advected to the regions of interest and that could be heated due to subsidence. This would also allow the authors to save a panel on an already heavy figure.
Line 160 – 161: Is this true for Temp? Over South Africa, the anomalies are close to zero! I have doubts about the quality of these results. If the temperature values are within the expected values, we can’t consider this as a Heatwave. For instance, Pretoria region has even negatives anomalies of Temp (once again, what is Temp??)… In table 1 authors say the following: “Heatwave Concurrent with a drought and an ENSO event. Temperatures reached 42.7 °C in Pretoria and 38.9 °C in Johannesburg on 6 th January.” Something here is not right. I know that the heatwave definition used by authors is based on the UTCI values, but, isn’t it strange to have a Heatwave event with lower temperatures than the expected values? Once again, how are the UTCI values obtained? This needs a careful and urgent review!!
Line 165: It should be Z850… Once again, the authors are not extracting any relevant information from the Z850 anomaly field.
Line 166 – 167: Not sure about what the authors mean to say here. Please clarify
Line 190: Slight cool anomalies? According to the results, there are regions reaching values between -5ºC and -10ºC.
Line 196 – 198: Not sure about this… Morocco is still under the influence of high anomalies of Temp (reaching levels very close to 5ºC). Authors need to Check the data because the results are not consistent with what is discussed in the text.
Line 198 – 200: Please rewrite this sentence.
Lines 200 – 201: Please rewrite this sentence. Not sure about what authors are trying to say here.
Lines 232 – 234: Authors should include some possible explanations for this, for instance, the topography, the influence of more local meteorological processes, the influence of the Sahara Desert, different land covers….
Lines 234 – 236: Which pressure pattern? Authors were able to link heatwaves in South Africa and Morocco with geopotential anomalies. Here, in the discussion, they should detail what are the synoptic processes that are behind this link.
Lines 236 – 239: Authors should rewrite this sentence please.
Line 240: This would be the ideal paragraph to authors mention some of the limitations of this study.
Lines 259 – 260: Authors should rewrite this sentence please.
Lines 267 – 276. Authors should rewrite all the Conclusion paragraph. There are a lot of syntax and grammatical errors. The text is confused and disconnected. I’m not sure if authors were successful in covering the ideal bullets for the Conclusion section (Why is this study relevant? What were the key findings and what can we learn from them? Is this a novel approach for Africa? How could these results be used to predict future events? Should policymakers and authorities start to look at these climatic extreme with more attention? Basically, why this work should be publish and shared.
-
AC1: 'Reply on RC1', Chloe Brimicombe, 01 Nov 2021
The work here presented has the potential for being a relevant and useful study. The subject under discussion is of extreme importance considering the current climate change context, the natural and socio-economic impacts linked to heatwaves, the demographic projections for Africa and, finally, the lack of analysis conducted for this continent, where the living conditions of most of the population are poor and climate change is far from being a priority for the policymakers. I believe the authors share this perception and have in mind the relevance that studies of this kind have.
We thank the reviewer, our research does indeed work to address the under discussion of extreme heat and climate change.
However, the manuscript calls for major revisions and needs to be improved in many aspects. First, regarding the quality of writing, in many sections the text is confused, disconnected and with grammatical and syntax errors. The reader often struggles to understand the rational and the ideas that authors pretend to share. Therefore, I strongly recommend a thorough review of the English writing throughout all the paper.
Thank you for this valuable feedback. We will thoroughly review and revise the manuscript.
Some of the methods and metrics adopted by the authors must be clarified. For instance, the UTCI index lacks a mathematical definition and a more detailed interpretation of its values.
We will clarify the methods section. The UTCI is a robust index and has been previously used for African Nations please refer to citations in this paper: Guigma, et al. 2020
In addition, We will include the mathematical definition of the UTCI which is:
Where Ta= air temperature, Tr= mean radiant temperature, Va= wind speed Pa= water vapour pressure.
I also have some doubts regarding some of the methodologies that were adopted by authors to achieve the objectives that were proposed. In addition, some results are not consistent with the interpretation and discussion conducted throughout the manuscript text.
We have clarified our methods to ensure that the way in which they achieve the objectives is clear to the reader. We have also reviewed our results and find them to be consistent with our interpretation. We will address further comments on results below.
I found even some results “unexpected” and very strange, calling for an urgent review. As a consequence of this, most of the conclusions made by the authors find a weak support in the results presented.
We have clarified the result section to discuss why the results are not unexpected and to make clear how our conclusions are supported by the evidence.
The manuscript needs major reviews. I ask the authors to kindly consider the following comments and suggestions:
Line 23: What is “Morocco heat”?
We will change this to Morocco Heatwave.
Line 26-27: Please clarify this sentence… I’m not sure what authors are trying to say here.
We will change this to “Further it can be seen that in some African Nations heatwaves are reported but this needs improvement in order to capture impacts across the continent.”
Line 32-33: These deaths occurred mostly during two particular mega-heatwaves events in Europe. Authors should detail the particularities of these two episodes (duration, severity of the induced heat-stress levels) and explain why they were responsible for such high mortality rates and other impacts in many natural and socio-economic sectors… This would reinforce the importance of analyze in detail specific historical events such as the ones studied in this work.
We have a recent paper published showing that these mega-heatwaves were not just in Europe Brimicombe, et al., 2021. which will now make reference to. It shows that there were high heat stress levels during both the 2003 and 2010 August Heatwaves in parts of the African continent and that the impacts of heatwaves are under-reported.
Lines 36 – 45: This paragraph is very important to show the relevance of this study. It represents a substantial contribution to the understanding of natural hazards and their consequences. The authors should detail how the poor living conditions of most of the African population are crucial to explain the high levels of heat-related mortality and how the fragile economies struggle to recover from heatwaves and other climate extremes – showing once again the relevance of this type of studies, particularly for the African continent.
We agree with the reviewer entirely and will add this to the manuscript as suggest. We will also include a nod to how building material can increase the indoor temperature. However, as noted in this paragraph there is sparse recorded evidence for heatwave impacts across Africa and therefore a lack of evidence recorded of economic struggles.
Line 75: Authors should provide a mathematical definition for the UTCI index. They should also provide a thorough interpretation of what this index really means. What are the variables involved and what are the implications for humans when extreme high/low values of this index are recorded?
We have included the mathematical definition of the UTCI which is:
Where Ta= air temperature, Tr= mean radiant temperature, Va= wind speed Pa= water vapour pressure.
In addition to the results section: The UTCI is a thermal index which makes use of the meteorological parameters of 2m temperature, water vapour pressure, 10m wind speed and mean radiant temperature and a body model (Di Napoli et al 2021). It has been compared to many other thermal indices such as apparent temperature and heat index and captures well an average body response to the thermal environment (Zare et al 2018, Jendritzky et al 2012, Blazejczyk et al 2012). It has further been shown to be able to forecast heatwaves internationally (Pappenberger et al 2015) and accurately indicate extreme heat for Africa (Guigma et al 2020).
Line 93-97: Check the syntax of this sentence.
Thank you we have checked the syntax
Line 98: I’m not sure what authors mean by “notably area”; check the syntax. Also, try to use a more robust and objective reason (severity, magnitude of the events, spatial extend……) to justify the decision of choosing these two events in particular… I’m not convinced. Also, “Reported Impacts”? Which ones?
We will clarify the reason for choosing these event and thank you we will change area to spatial extent. We also have added a reference to Table 1 for reported impacts.
Line 118-119: Poor English writing.
We will simplify the sentence structure.
Academic Literature and reports show that 39 heatwaves have been reported since 1980 in Africa.
Line 122: Showing the coordinates of these regions only makes sense if the authors reference a figure….
We have asked a range of colleagues and they find the longitude and latitude references useful to understand where in Africa each region is.
Line 125-126: I’m confused. This is not clear. Rewrite this sentence please.
We will change this to: “Characteristics of heatwaves (n=26) are reported more than their impacts (n=9).”
• The caption of the figure is very poor. Authors should provide more information about the variables and the figures that are presented: What are the panels showing? Anomalies? “Temp” is surface temperature?
We will describe each figure in more detail.
• The location of South Africa (Morocco) within the African continent should be highlighted.
We agree and will add a box highlighting
• I don’t like the layout (horizontal) of the figure… It’s not easy for the reader to follow the figure and the text at the same time. Try to order the panels in a vertical way.
We agree but this was necessary because of the format of NHESS peer-review system.
• Try to save some space by removing the colorbars and the legends of the x-axis/y-axis that are repeated throughout the several panels.
Our colleagues have found it useful to have a scale on each figure.
• The bottom panel with the UTCI time series is supposed to be Figure 1d, isn’t it? Also, are these area averaged UTCI values for South Africa (Morocco in Figure 2)? This is not mentioned… Please clarify.
We will clarify this in the figure description please see as above.
Line 157: I’m not sure about the need to show the Z850 anomalies considering that you are already showing the 500-hpa geopotential anomaly.
It has previously been shown that the Saharan Heat Low is best captured by between 700 and 900hpa values, a higher height than 500hpa (Lavaysse et al 2016). We have clarified this in the text.
Line 160 – 161: Is this true for Temp? Over South Africa, the anomalies are close to zero! I have doubts about the quality of these results. If the temperature values are within the expected values, we can’t consider this as a Heatwave. For instance, Pretoria region has even negatives anomalies of Temp (once again, what is Temp??)… In table 1 authors say the following: “Heatwave Concurrent with a drought and an ENSO event. Temperatures reached 42.7 °C in Pretoria and 38.9 °C in Johannesburg on 6 th January.” Something here is not right. I know that the heatwave definition used by authors is based on the UTCI values, but, isn’t it strange to have a Heatwave event with lower temperatures than the expected values? Once again, how are the UTCI values obtained? This needs a careful and urgent review!!
The figure is showing over the part of South Africa where the heatwave is present anomalies in temperature (Temp) are up to 2C hotter than the climate. We will clarify this in the text.
The UTCI is not the same as temperature, please refer to the methods section. It is a model of a body response to the thermal environment. It assess how temperature, mean radiant temperature, water vapour pressure and wind speed effect an average body and indicates heat stress (Di Napoli, et al., 2018,2019,2021).
In the figure the parts of South Africa experiencing a heatwave are on the southern coast including over cape town. The anomalies are present at up to 2C hotter in Temperature for both Pretoria and Johannesburg.
ERA5 data as a reanalysis dataset will not give the same values as recorded observed temperature and we will also clarify this in the methods section. ERA5 however has been found to accurately identify heatwaves despite not having values that match the maximum temperature recorded (Di Napoli, et al 2018, Brimicombe et al 2021,Hersbach et al 2020).
Line 165: It should be Z850… Once again, the authors are not extracting any relevant information from the Z850 anomaly field.
Thank you we have corrected this typo
Line 166 – 167: Not sure about what the authors mean to say here. Please clarify
We will re-write this to clarify:
Considering the drop in temperature and UTCI anomalies it can be inferred that this low pressure has a cooling influence on the region.
Line 190: Slight cool anomalies? According to the results, there are regions reaching values between -5ºC and -10ºC.
Thank you this is a typo we have clarified that anomalies are between -5C and -10C
Line 196 – 198: Not sure about this… Morocco is still under the influence of high anomalies of Temp (reaching levels very close to 5ºC). Authors need to Check the data because the results are not consistent with what is discussed in the text.
We will add a box to the figure to aid the comparison of the written text and figure. The results are consistent with what is discussed in the text.
Line 198 – 200: Please rewrite this sentence.
We will restructure this sentence ‘the z850 and z500 negative geopotential anomalies to the east of Morocco indicate a low pressure system which could be in part having a cooling influence on the region.’
Lines 200 – 201: Please rewrite this sentence. Not sure about what authors are trying to say here.
We will restructure the sentence.
‘The peak of the UTCI and temperature are similar during the Morocco 2000 and South Africa 2015/16 Heatwave, with anomalies of up 2°C respectively.’
Lines 232 – 234: Authors should include some possible explanations for this, for instance, the topography, the influence of more local meteorological processes, the influence of the Sahara Desert, different land covers….
Yes we agree and have expanded this section to include information about the desert winds that influence the temperature regime of Morocco (Filahi, et al 2016) As well as discussing the position of the Saharan Heat Low which also influences the air flows for Northern Africa (Lavaysse et al 2016).
Lines 234 – 236: Which pressure pattern? Authors were able to link heatwaves in South Africa and Morocco with geopotential anomalies. Here, in the discussion, they should detail what are the synoptic processes that are behind this link.
Thank you we have clarified that this is a high pressure system.
Lines 236 – 239: Authors should rewrite this sentence please.
We will rewrite this as follows: Heatwaves drivers are closely linked to the climatology and synoptic systems in a region for both of the heatwaves examples
Line 240: This would be the ideal paragraph to authors mention some of the limitations of this study.
We agree and have clarified this paragraph’s aim:
Further work should consider more fully the meteorology of the heatwave this study has more of a focus on extreme heat impacts, such as heat stress.
Lines 259 – 260: Authors should rewrite this sentence please.
We will rewrite the following as African heatwaves have a lack of observations, under-reporting fuels a lack of evidence and a patchy research field, intrinsic factors that has a growing mandate to be addressed globally
Lines 267 – 276. Authors should rewrite all the Conclusion paragraph. There are a lot of syntax and grammatical errors. The text is confused and disconnected. I’m not sure if authors were successful in covering the ideal bullets for the Conclusion section (Why is this study relevant? What were the key findings and what can we learn from them? Is this a novel approach for Africa? How could these results be used to predict future events? Should policymakers and authorities start to look at these climatic extreme with more attention? Basically, why this work should be publish and shared.
We agree here’s how we would approach your questions in the conclusion.
Why is this study relevant?
This study is relevant because it highlights extreme heat risk for a continent where this research is sparse and where the impacts to heatwaves are ever increasing.
What were the key findings and what can we learn from them?
The key finding is that reporting although patchy for the continent is not captured in international databases, even in retrospect. And that heatwave characteristics although a rise in temperature and heat stress are diverse when taking into account the local climate and geography of a region. This means that specialist information should be provided to nations weather services to suite.
Is this a novel approach for Africa?
This will be the first peer-reviewed systematic list of heatwaves including recorded physical characteristics and impacts for the continent. This is also the first time that the Universal Thermal Climate Index has been used at a continental scale to assess heat stress during a heatwave in Africa.
How could these results be used to predict future events?
These results can be used to improve nations understanding of the characteristics and impacts of heatwaves, in the short term. In addition, the methods can be adapted and applied to climate projections to look into future risk.
Should policymakers and authorities start to look at these climatic extreme with more attention?
The main finding is that policy-makers and authorities should put in place adaptations to address extreme heat as a growing risk for their countries. And our results demonstrate heat is not just a future risk but a real risk with devastating impacts now.
Citation: https://doi.org/10.5194/nhess-2021-242-AC1
-
RC2: 'Comment on nhess-2021-242', Anonymous Referee #2, 28 Sep 2021
General Comments
This paper discusses heat waves in Africa, and provides a list of African heat wave events compiled from peer reviewed literature, as well as grey literature and media reports. Case studies of two heatwaves is provided, with discussion of air temperature, UTCI, and pressure systems.
This paper is important because the impacts of heatwaves in sub-Saharan Africa are possibly under-reported.
Before publication, the method of the literature search needs to be more thoroughly described.
There are a few sentences that are confusing. The text is well organized and about the right length. It refers to appropriate related literature.
It is not explained how the 21 academic papers and grey literature where arrived at. Typically, if a paper relies on a systematic search of the literature, then the search engine and search terms used should be give. The reader is not able to judge whether the search was systematic or totally ad-hoc, and therefore whether or not the search is exhaustive.
The only rows in the table that have human impacts have a reference to CRED. Are all of the human impacts from CRED? Do none of the reports from other sources mention human impacts? If so, this should be highlighted.
Page 14 line 236. "This is the first time heatwaves from two different African regions have been presented and compared using both their physical characteristics and reported impacts" The impacts are barely discussed in this paper. They are stated in the table, but the information comes directly from EM-DAT and no further information is added. No comparison is made between the impacts of the heatwaves.
Page 11, figure 1. During the heatwave, it does not appear that temperatures are really higher than usual in South Africa, whereas there is a high temperature anomaly over Namibia. Perhaps this is something to do with the resolution of the image, or the choice of color scale. Actual values for the temperature anomaly during the heatwave are not given in the text, and the reader is left wondering if temperatures were actually anomalously high during this period according to the reanalysis.
Is there any reporting for this time period in Namibia? It might be good to point out the discrepancy in reporting.
The figures are quite hard to view in the format given. It is difficult to compare different time steps in the same variable as they are not adjacent. The maps are tiny, so appear low resolution in the PDF version.
The word "mandate" is used a few times to mean a motivation for further study - I am not sure this is a correct use of the word.
On page 3, line 66 the phrase "between 1980 and 2020 to date" is confusing. Does the author mean 2020 up to the date at which they are writing? They should instead say when that date was specifically.
EM-DAT is first mentioned in section 2.1, but is not defined or referenced.
On page 4, line 90. "It [UTCI] has further been shown to be able to forecast heatwaves internationally (Pappenberger et al 2015)". The phrasing suggests that UTCI in isolation is able to create international forecasts. The referenced paper retrospectively shows that a NWP forecast of UTCI could have been used to forecast the 2010 Russia heatwave. The point is that UTCI forecasts can be used, not that UTCI produces forecasts.
Page 5, line 118. Hard to read sentence. "Academic Literature and reports for Africa show that 39 heatwaves are reported for somewhere in Africa almost every year since 1980." This reads as if each year 39 heatwaves are reported somewhere.
Page 8, in the table. "Was the warmest April to date at the time in the Sahel" is in the "Heatwave Impacts" column, it should be in the "Heatwave Characteristics" column.
Page 12, line 205. The phrase "heat stress" is used instead of "UTCI" making it unclear what is being measured.
Page 14 line 222, "suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa". These other studies are not suggesting that EM-DAT is not accurate, rather that it is not complete. Harrington & Otto make no mention of accuracy.
Page 14 line 222. "our study supports others (i.e. Harrington and Otto, 2020; van der Walt and Fitchett , 2021) suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa including sub-Saharan Africa only including 7 out of our 39 listed heatwaves in their records, which is less than 20%." This sentence is hard to read. Break it into two sentences.
Page 14, line 229. "However, there are some heatwave warning systems (Hafez and Almazroui 2016, Boubaker
230 2010)" Are these systems another potential source for identifying heatwaves? Were the heatwaves in your study identified by these systems?
Page 14, line 225 to 228 repeats points already made in lines 221-224.
Page 14, line 231. "and there is some reporting in place for African Nations that is not always captured at an international scale" what is meant here by the "international scale"? Does it mean the international databases referenced above?
Page 15, line 250 - 253. The evidence provided on lines 250-252 does not directly support the hypothesis on line 253. Comparing an air temperature mortality effect with a UTCI anomaly in this way is misleading. We can see from figure 1 that the UTCI anomaly was large during the South Africa heatwave, but we cannot see that the air temperature anomaly is large. You should state what the air temperature anomaly was. This probably still leads to more deaths than the 11 reported.
Citation: https://doi.org/10.5194/nhess-2021-242-RC2 -
AC2: 'Reply on RC2', Chloe Brimicombe, 01 Nov 2021
This paper discusses heat waves in Africa, and provides a list of African heat wave events compiled from peer reviewed literature, as well as grey literature and media reports. Case studies of two heatwaves is provided, with discussion of air temperature, UTCI, and pressure systems.
Thank you for your comments.
This paper is important because the impacts of heatwaves in sub-Saharan Africa are possibly under-reported.
Yes however sub-Saharan African Heatwaves are under-reported.
Before publication, the method of the literature search needs to be more thoroughly described.
We agree and will clarify in the methods: . A systematic search was carried out using google scholar whereby chosen literature had to meet the criteria of having a focus on historic heatwaves in Africa and include either a physical characteristic of the heatwave or impact.
There are a few sentences that are confusing. The text is well organized and about the right length. It refers to appropriate related literature.
Thank you, all co-authors have read the paper again to avoid confusing sentence structure.
It is not explained how the 21 academic papers and grey literature where arrived at. Typically, if a paper relies on a systematic search of the literature, then the search engine and search terms used should be give. The reader is not able to judge whether the search was systematic or totally ad-hoc, and therefore whether or not the search is exhaustive.
As above we will clarify this.
The only rows in the table that have human impacts have a reference to CRED. Are all of the human impacts from CRED? Do none of the reports from other sources mention human impacts? If so, this should be highlighted.
They were reports that supported the evidence of impacts in em-dat and they are also referenced.
Page 14 line 236. "This is the first time heatwaves from two different African regions have been presented and compared using both their physical characteristics and reported impacts" The impacts are barely discussed in this paper. They are stated in the table, but the information comes directly from EM-DAT and no further information is added. No comparison is made between the impacts of the heatwaves.
We will add a discussion comparing the reported impacts.
Considering that our analysis shows positive anomalies in air temperature of up to 2°C and heat stress indicated by the UTCI of up to 4°C, we hypothesise that impacts of the South Africa heatwaves have been underreported. For the Morocco heatwave the reported impacts are 4 million chickens dying leading to up to 809,000 USD of damages (table 1). Interestingly, this is the only heatwave with an economic loss associated with it. This reiterates the complexities of identify loss and damage due to heat extremes, especially for Africa where research on heat-related impacts is still limited (Campbell et al 2018).
Page 11, figure 1. During the heatwave, it does not appear that temperatures are really higher than usual in South Africa, whereas there is a high temperature anomaly over Namibia. Perhaps this is something to do with the resolution of the image, or the choice of color scale. Actual values for the temperature anomaly during the heatwave are not given in the text, and the reader is left wondering if temperatures were actually anomalously high during this period according to the reanalysis.
There is a temperature anomaly of up to 2C across parts of South Africa. We will clarify also that reanalysis will not give the same values as those reported.
Is there any reporting for this time period in Namibia? It might be good to point out the discrepancy in reporting.
Namibia is a region that we would like to look into further it is not the focus of heatwave studies or any reports we found.
The figures are quite hard to view in the format given. It is difficult to compare different time steps in the same variable as they are not adjacent. The maps are tiny, so appear low resolution in the PDF version.
This was due to NHESS peer-review requirements.
The word "mandate" is used a few times to mean a motivation for further study - I am not sure this is a correct use of the word.
We have reviewed the use of the word.
On page 3, line 66 the phrase "between 1980 and 2020 to date" is confusing. Does the author mean 2020 up to the date at which they are writing? They should instead say when that date was specifically.
We will clarfiy this to be up to the end of 2020.
EM-DAT is first mentioned in section 2.1, but is not defined or referenced.
Thank you we have now referenced EM-DAT.
On page 4, line 90. "It [UTCI] has further been shown to be able to forecast heatwaves internationally (Pappenberger et al 2015)". The phrasing suggests that UTCI in isolation is able to create international forecasts. The referenced paper retrospectively shows that a NWP forecast of UTCI could have been used to forecast the 2010 Russia heatwave. The point is that UTCI forecasts can be used, not that UTCI produces forecasts.
Thank you we have changed the phrasing:
It has further been shown that the UTCI as a thermal index can be used to forecast heatwaves internationally (Pappenberger et al 2015)
Page 5, line 118. Hard to read sentence. "Academic Literature and reports for Africa show that 39 heatwaves are reported for somewhere in Africa almost every year since 1980." This reads as if each year 39 heatwaves are reported somewhere.
We have re-worded this sentence:
Academic Literature and reports show that 39 heatwaves have been reported since 1980 in Africa.
Page 8, in the table. "Was the warmest April to date at the time in the Sahel" is in the "Heatwave Impacts" column, it should be in the "Heatwave Characteristics" column.
Thank you this is a typo we have moved this to the correct column.
Page 12, line 205. The phrase "heat stress" is used instead of "UTCI" making it unclear what is being measured.
We have changed this to be UTCI.
Page 14 line 222, "suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa". These other studies are not suggesting that EM-DAT is not accurate, rather that it is not complete. Harrington & Otto make no mention of accuracy.
We have changed to in-complete instead of accuracy to clarify our meaning.
Page 14 line 222. "our study supports others (i.e. Harrington and Otto, 2020; van der Walt and Fitchett , 2021) suggesting that international databases such as EM-DAT (CRED 2020) are not accurately recording heatwaves for Africa including sub-Saharan Africa only including 7 out of our 39 listed heatwaves in their records, which is less than 20%." This sentence is hard to read. Break it into two sentences.
We agree: Further, our study supports others (i.e. Harrington and Otto, 2020; van der Walt and Fitchett , 2021) suggesting that international databases such as EM-DAT (CRED 2020) are not recording all heatwaves for Africa. For Example, sub-Saharan Africa only including 7 out of our 39 listed heatwaves in their records, which is less than 20%. However, it’s important to acknowledge there are some heatwave warning systems (Hafez and Almazroui 2016, Boubaker 2010) and there is some reporting in place for African Nations that is not always captured at an international scale within reports or databases (Table 1).
Page 14, line 229. "However, there are some heatwave warning systems (Hafez and Almazroui 2016, Boubaker
230 2010)" Are these systems another potential source for identifying heatwaves? Were the heatwaves in your study identified by these systems?
Yes – Tunisia and Egypt. We have included this in the text (i.e. Tunisia and Egypt)
Page 14, line 225 to 228 repeats points already made in lines 221-224.
Thank you, we have removed this.
Page 14, line 231. "and there is some reporting in place for African Nations that is not always captured at an international scale" what is meant here by the "international scale"? Does it mean the international databases referenced above?
We mean International Databases and Reports we have now clarified this.
Page 15, line 250 - 253. The evidence provided on lines 250-252 does not directly support the hypothesis on line 253. Comparing an air temperature mortality effect with a UTCI anomaly in this way is misleading. We can see from figure 1 that the UTCI anomaly was large during the South Africa heatwave, but we cannot see that the air temperature anomaly is large. You should state what the air temperature anomaly was. This probably still leads to more deaths than the 11 reported.
We have clarified this so that it is not misleading. As above there is an air temperature anomaly for south Africa. In addition UTCI has been shown to cause a rise in mortality.
Considering that our analysis shows positive anomalies in air temperature of up to 2°C and heat stress indicated by the UTCI of up to 4°C, we hypothesise that impacts of the South Africa heatwaves have been underreported.
Citation: https://doi.org/10.5194/nhess-2021-242-AC2
-
AC2: 'Reply on RC2', Chloe Brimicombe, 01 Nov 2021
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
1,379 | 407 | 115 | 1,901 | 56 | 113 |
- HTML: 1,379
- PDF: 407
- XML: 115
- Total: 1,901
- BibTeX: 56
- EndNote: 113
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1
Cited
2 citations as recorded by crossref.
- Introducing Project-Based Climate Education in Moroccan Universities via a New Air Quality Monitoring Network in Rabat M. Lbadaoui-Darvas et al. 10.1051/e3sconf/202341805003
- Assessing the impact of climate change on future extreme temperature events in major South African cities A. Mengistu et al. 10.1007/s00704-023-04712-w