Investigating West African Monsoon Features in Warm Years Using the Regional Climate Model RegCM4
Round 1
Reviewer 1 Report
The authors focus on the changes in West African monsoon (WAM) features during the warm years between 1980-2009. The analysis includes rainfall and temperature datasets and they use RegCM4 model simulations, ERA-Interim reanalysis data and GPCP observation data. The analysis is thorough and useful (in terms of mitigation and adaptation measures in West Africa), but I recommend to revise some major points. More specifically:
The authors mention that they use ERA-Interim reanalysis data to initialize RegCM4 model. Please provide further explanation for this procedure in the manuscript
In Figure 1, the authors display the simulation domain and the considered sub-domains that they use in this study. How are these sub-domains selected? Did they use an objective method for this (e.g. Principal Components Analysis)? Please refer to the manuscript
As a first step, the authors calculate the differences between RegCM4 simulation and ERA-Interim reanalysis data, as well as RegCM4 simulation and GPCP observation data. However, the resolutions of the three datasets are different (RegCM4 -> 50km, ERA-Interim -> 1.5ox1.5o, GPCP -> 2.5ox2.5o). How did they calculate the differences if the datasets have different spatial resolution? Refer to the used method in the manuscript (nearest neighbor? remapping?) I would recommend that ERA-Interim reanalysis data with 0.5ox0.5o resolution should be used
There is not sufficient information for the calculation of the anomalies (reference period etc.)
The authors mention that for the purposes of this study they choose fifteen warmest years. Additionally, in the analysis they calculate the changes of the examined indices using the average of these warmest years and the average of the other years. The authors should either change the analysis by refering to the years above and below zero Celsius degrees or use a threshold above which they characterize a year as "warm". The mean of the warm years should be compared with the mean of the cold years (threshold) or the mean of the whole period (1980-2009) in order for the results to be more robust
Table 1: I believe the units for R1mm index is Days and not mm/day as it refers to the number of days with daily cumulative precipitation greater than 1mm
Please check the formatting of the references list at the end of the manuscript. Various formatting is used, which do not comply with the journal's guides
The authors should further develop the methodology that they use in the study in Section 2.2 Data and methods in order to be clear to the reader
In line 284, it is Intergovernmental Panel on Climate Change and not International
Please check the manuscript for English language, grammar and syntax. Here I mention some of the typos and grammar mistakes in the text:
Line 64: 'are implications' -> 'are the implications'
Line 66: 'are -> 'is'
Line 72: 'were' -> 'are'
Line 84: 'scales' -> 'scale'
Line 89: 'enough large' -> 'large enough'
Line 148: 'well simulates' -> 'simulates well'
Line 219: 'starts' -> 'start'
Line 220: 'ends' -> 'end'
Lines 296, 312, 372: 'indice' -> 'index'
Lines 300, 375: 'that of' -> 'the'
Line 301: 'increases' -> increase'
Lines 307-308: the phrase 'over the Sahel and the Guinea Coast' should be moved after '(Figure 7c,d)'
Line 341: 'is' -> 'are'
Line 349: 'South' -> 'south'
Line 354: 'differences' -> 'difference'
Line 355: 'and located' -> 'and is located'
Line 375: 'decreases' -> 'decrease'
Line 376: 'increases' -> 'increase'
Line 385: 'increase of' -> 'increase in'
General grammar rule (check throughout the manuscript) -> Put commas before which and while
Author Response
Dear Editor, Dear reviewer,
Please find below the response (in bold) to the comments of the reviewers. The changes in the revised version are highlighted in blue colour. Thank you in advance for your cooperation.
Best regards
Comments and Suggestions for Authors
The authors focus on the changes in West African monsoon (WAM) features during the warm years between 1980-2009. The analysis includes rainfall and temperature datasets and they use RegCM4 model simulations, ERA-Interim reanalysis data and GPCP observation data. The analysis is thorough and useful (in terms of mitigation and adaptation measures in West Africa), but I recommend to revise some major points. More specifically:
The authors mention that they use ERA-Interim reanalysis data to initialize RegCM4 model. Please provide further explanation for this procedure in the manuscript.
First of all, we would like to thank you for your interesting comments. We took into account this suggestion in the revised manuscript.
In Figure 1, the authors display the simulation domain and the considered sub-domains that they use in this study. How are these sub-domains selected? Did they use an objective method for this (e.g. Principal Components Analysis)? Please refer to the manuscript.
The considered sub-domains (western Sahel, central Sahel and the Guinea region) are those commonly used for local studies in the West African regions. They have different landscapes and surface conditions (forest over Guinea region, arid and semi-arid areas with short and tall grass savanna and trees over the western and central Sahel). These regions are characterized by different rainfall regimes when considering the annual cycle. The Guinea region is characterized by two rainy seasons, while the western and eastern Sahel are characterized by a single wet season from june to september.
Thank you.
As a first step, the authors calculate the differences between RegCM4 simulation and ERA-Interim reanalysis data, as well as RegCM4 simulation and GPCP observation data. However, the resolutions of the three datasets are different (RegCM4 -> 50km, ERA-Interim -> 1.5ox1.5o, GPCP -> 2.5ox2.5o). How did they calculate the differences if the datasets have different spatial resolution? Refer to the used method in the manuscript (nearest neighbor? remapping?) I would recommend that ERA-Interim reanalysis data with 0.5ox0.5o resolution should be used.
To calculate the differences between the datasets, we just interpolated the high resolution data onto the coarse one. For an example, the RegCM4 simulation (50km horizontal grid resolution) is interpolated onto the GPCP data (resolution of 2.5ox2.5o) using the Grid Analysis and Display System (GrADS) tools when calculating the difference between these two datasets. Please, for more details see the references manuscript of this software by following this link: http://cola.gmu.edu/grads/gadoc/reference_card.pdf.
The Era-interim reanalysis at 1.5deg resolution are used to initialize and drive the RegCM4 simulation, that’s the reason why we used this same dataset and resolution for the subsequent analysis.
There is not sufficient information for the calculation of the anomalies (reference period etc.)
We stated in the manuscript that the Kraus method (1977) is used for the calculation of the temperature and rainfall anomalies. This method is also used and exposed in details in Diba et al. (2018). The reference period for the study of the interannual anomalies is the same period considered in this paper (1980-2009).
Thank you.
1. Kraus, E.B. Subtropical droughts and cross equatorial energy transports. Monthly Weather Review. 1977, 105: 1009-18.
2. Diba, I.; Camara, M.; Sarr, A.B.; Diedhhiou, A. Potential Impacts of Land Cover Change on the Interannual Variability of Rainfall and Surface Temperature over West Africa. Atmosphere 2018, 9(10), 376; 32 pages. https://doi.org/10.3390/atmos9100376.
The authors mention that for the purposes of this study they choose fifteen warmest years. Additionally, in the analysis they calculate the changes of the examined indices using the average of these warmest years and the average of the other years. The authors should either change the analysis by refering to the years above and below zero Celsius degrees or use a threshold above which they characterize a year as "warm". The mean of the warm years should be compared with the mean of the cold years (threshold) or the mean of the whole period (1980-2009) in order for the results to be more robust.
Thank you for this important remark. In this revised version, we compared the mean of the nine (9) warmest years with the mean of the nine (9) coldest years. A Warm/Cold year refers to a year with temperatures normalized anomalies above/below 0.5 Celsius degrees over West Africa.
Table 1: I believe the units for R1mm index is Days and not mm/day as it refers to the number of days with daily cumulative precipitation greater than 1mm
Fixed. Thank you for this remark.
Please check the formatting of the references list at the end of the manuscript. Various formatting is used, which do not comply with the journal's guides
Fixed. Thank you.
The authors should further develop the methodology that they use in the study in Section 2.2 Data and methods in order to be clear to the reader.
Supplementary information are added about the methodology in the revised manuscript. Thank you.
In line 284, it is Intergovernmental Panel on Climate Change and not International
Fixed. Thank you.
Please check the manuscript for English language, grammar and syntax. Here I mention some of the typos and grammar mistakes in the text:
Line 64: 'are implications' -> 'are the implications'
Fixed. Thank you.
Line 66: 'are -> 'is'
Fixed. Thank you.
Line 72: 'were' -> 'are'
Fixed. Thank you.
Line 84: 'scales' -> 'scale'
Fixed. Thank you.
Line 89: 'enough large' -> 'large enough'
Fixed. Thank you.
Line 148: 'well simulates' -> 'simulates well'
Fixed. Thank you.
Line 219: 'starts' -> 'start'
Fixed. Thank you.
Line 220: 'ends' -> 'end'
Fixed. Thank you.
Lines 296, 312, 372: 'indice' -> 'index'
Fixed. Thank you.
Lines 300, 375: 'that of' -> 'the'
Fixed. Thank you.
Line 301: 'increases' -> increase'
Fixed. Thank you.
Lines 307-308: the phrase 'over the Sahel and the Guinea Coast' should be moved after '(Figure 7c,d)'
Fixed. Thank you.
Line 341: 'is' -> 'are'
Fixed. Thank you.
Line 349: 'South' -> 'south'
Fixed. Thank you.
Line 354: 'differences' -> 'difference'
Fixed. Thank you.
Line 355: 'and located' -> 'and is located'
Fixed. Thank you.
Line 375: 'decreases' -> 'decrease'
Fixed. Thank you.
Line 376: 'increases' -> 'increase'
Fixed. Thank you.
Line 385: 'increase of' -> 'increase in'
Fixed. Thank you.
General grammar rule (check throughout the manuscript) -> Put commas before which and while
Fixed. Thank you.
THANK YOU FOR YOUR VALUABLE COMMENTS.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Review of "Investigating West African Monsoon Features in Warm Years using RegCM4 climate model" by Ibrahima et al.
This manuscript presented the West African Monsoon (WAM) features during warm years from 1980 to 2009 using ERA-interim reanalysis as well as RegCM4 model simulations, with the focus on spatial patterns, interannual variability and annual cycle of surface temperature and rainfall, and some climate indices recommended by WMO. The authors then discussed the changes in the monsoon flow, the magnitude of the African Easterly Jet and the Tropical Easterly Jet, from low-level to the upper troposphere.
Overall, I find this study very informative and useful, especially in the context of what those ‘2oC’ or ‘1.5oC’ global temperature reduction goals can really mean for the regional climate conditions as in West Africa. However, I suggest some major points be addressed before the paper is considered for publication, as follows.
Major comments:
1. The motivation of this research is not well presented, or may be quite misleading. After reading the title and introduction, I have a feeling that the authors proposed to study WAM features using the regional climate because it has a relatively high resolution so it can resolve fine-scale dynamics of the atmosphere. I would expect to see insightful discussions on these physical processes that current GCMs are missing. But it seems that in the rest of the manuscript the authors are simply doing a model-data comparison, without an emphasis on how the regional climate model simulation can help us to improve our understanding of the WAM dynamics during warm years. If it cannot be achieved, why not just show the results from reanalysis? Does any conclusion change when the model simulation results are included? Therefore, I think it to be a priority that the justification of the using of RCM4 simulations needs to be clarified.
2. There are some puzzling features in terms of the model-data comparison. For example, I find that the differences between reanalysis and RCM4 results are relatively small in Figs. 2,4,5, but in Figs. 6-8 the differences are quite substantial. How to explain the increased model biases in Figs. 6-8? Also, the model performance seems ok in terms of mean climate, what about (interannual) climate variability?
Minor comments:
1. Line 185-186. There is a total of 30-year (1980-2009) period of this study. For these 30 years, the authors chose 15 ‘warmest’ years. Please clarify why not use a smaller subset of the 15 years, like 5 years. Because the choice may easily include years that are not warm enough, thus damping the main features of the results.
2. Fig. 3. Is there a negative correlation between rainfall and temperature anomalies time series? If so, what’s the implication of the correlation?
3. Line216. ‘Second peak’ may not be a precise description because there is only one peak in Fig. 4e, the June rainfall cannot be considered as a peak, because July rainfall is larger than it.
4. Conclusions. The implications of this study can be better expressed. For example, are these results in general supporting the conclusions associated with WAM features in the recent IPCC 1.5oC special report?
5. Line 354. ‘differences’ should be ‘difference’.
Author Response
Dear Editor, Dear reviewer,
Please find below (in bold) the response to the comments of the reviewers. The changes in the revised version are highlighted in blue colour. Thank you in advance for your cooperation.
Best regards
Comments and Suggestions for Authors
Review of "Investigating West African Monsoon Features in Warm Years using RegCM4 climate model" by Ibrahima et al.
This manuscript presented the West African Monsoon (WAM) features during warm years from 1980 to 2009 using ERA-interim reanalysis as well as RegCM4 model simulations, with the focus on spatial patterns, interannual variability and annual cycle of surface temperature and rainfall, and some climate indices recommended by WMO. The authors then discussed the changes in the monsoon flow, the magnitude of the African Easterly Jet and the Tropical Easterly Jet, from low-level to the upper troposphere.
Overall, I find this study very informative and useful, especially in the context of what those ‘2oC’ or ‘1.5oC’ global temperature reduction goals can really mean for the regional climate conditions as in West Africa. However, I suggest some major points be addressed before the paper is considered for publication, as follows.
Major comments:
1. The motivation of this research is not well presented, or may be quite misleading. After reading the title and introduction, I have a feeling that the authors proposed to study WAM features using the regional climate because it has a relatively high resolution so it can resolve fine-scale dynamics of the atmosphere. I would expect to see insightful discussions on these physical processes that current GCMs are missing. But it seems that in the rest of the manuscript the authors are simply doing a model-data comparison, without an emphasis on how the regional climate model simulation can help us to improve our understanding of the WAM dynamics during warm years. If it cannot be achieved, why not just show the results from reanalysis? Does any conclusion change when the model simulation results are included? Therefore, I think it to be a priority that the justification of the using of RCM4 simulations needs to be clarified.
First of all, we would like to thank you for your interesting comments. This study focuses on a regional climate model (RegCM4) analysis in the aim to help us to improve our understanding of the West African monsoon dynamics during warm years. Therefore in this revised version, to be consistent with this goal, reanalysis are used only for the validation of the model in the beginning and then, we focus only the analysis done with the RegCM4 model.
Thank you.
2. There are some puzzling features in terms of the model-data comparison. For example, I find that the differences between reanalysis and RCM4 results are relatively small in Figs. 2,4,5, but in Figs. 6-8 the differences are quite substantial. How to explain the increased model biases in Figs. 6-8? Also, the model performance seems ok in terms of mean climate, what about (interannual) climate variability?
In the revised version, a new approach is used. We focus on the RegCM4 simulation analysis by comparing the nine warmest years to the nine coldest years.
However, we would like to state that in the previous version fig 2, 4; 5 refer to the mean climate analysis (smooth effect due to the averaging) while fig 6-8 are about events analysis. This may be a possible explanation for the differences.
In our previous paper (Diba et al. 2018) we analyzed the ability of the RegCM4 model to simulate the interannual variability of rainfall over West Africa. The results show the good performance of the RegCM4 model to simulate the interannual variability of rainfall. However they are some discrepancies.
Thank you
Diba, I.; Camara, M.; Sarr, A.B.; Diedhhiou, A. Potential Impacts of Land Cover Change on the Interannual Variability of Rainfall and Surface Temperature over West Africa. Atmosphere 2018, 9(10), 376; 32 pages. https://doi.org/10.3390/atmos9100376.
Minor comments:
1. Line 185-186. There is a total of 30-year (1980-2009) period of this study. For these 30 years, the authors chose 15 ‘warmest’ years. Please clarify why not use a smaller subset of the 15 years, like 5 years. Because the choice may easily include years that are not warm enough, thus damping the main features of the results.
We chose a new approach in the revised version by considering only the nine warmest/coldest years. A Warm/Cold year refers to a year with a temperature anomaly above/below 0.5°C over West Africa as diagnosed in Figure 3. Thank you.
2. Fig. 3. Is there a negative correlation between rainfall and temperature anomalies time series? If so, what’s the implication of the correlation?
Yes, there is a strong negative correlation (r= – 0.60) between rainfall and temperature anomalies time series (Fig.3) suggesting that the precipitation and temperature are in opposite phase. On average, warm (cold) years are associated with dry (wet) conditions in term of rainfall.
3. Line216. ‘Second peak’ may not be a precise description because there is only one peak in Fig. 4e, the June rainfall cannot be considered as a peak, because July rainfall is larger than it.
We took into account this remark. Thank you.
4. Conclusions. The implications of this study can be better expressed. For example, are these results in general supporting the conclusions associated with WAM features in the recent IPCC 1.5oC special report?
This study shows an increase (a decrease) of thermal extremes (rainfall extremes) indices during the warmest years over the Sahel and the Guinea region. These results are generally consistent with the recent IPCC special report on the regional impacts of global warming at 1.5°C and 2°C over West Africa. Therefore, this study can be considered as a support for the West African policymakers to implement adaptation and mitigation measures in this climate change context.
5. Line 354. ‘differences’ should be ‘difference’.
Fixed. Thank you.
THANK YOU FOR YOUR VALUABLE COMMENTS.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Significant efforts have been made from the authors to improve the manuscript. I would like to thank the authors for that.
The manuscript can be accepted in present form.
I have only a last comment to make (future advice) regarding the use of ERA-Interim data with 1.5ox1.5o in the analysis. It is not obligatory to use in the analysis the same dataset that is used for the ICBC to drive the model (pre-processing of RegCM).
Reviewer 2 Report
I would like to thank the authors for their efforts to improve the manuscript.
I think the revised manuscript can be accepted in present form.
