Staying Cool in A Warming Climate: Temperature, Electricity and Air Conditioning in Saudi Arabia

Round 1

Reviewer 1 Report

In the beginnings of the first two paragraphs, two ‘according to’ were used. Can you make a fine adjustment for them? In the section 3, data sources and analyzing methods should be given the subtitles. Because, the essentially statistical methods such as descriptive statistics, normality test, and correlation should be given. In Figure 11, please add the df and p values to show the statistically significant level. Besides, if it is possible, please show the fitted regression models depicting the cause-effect relationship between average hourly temperatures and hourly power generation. Since some of the trends of the scatterplots did not follow the linear shapes, it is necessary to draw the right fitted lines with reasonable fitted regression models.

Author Response

We thank the reviewer for their comments.

We have made minor edits to opening text as requested at line 36. 

In section three we have organised the section into two sections as advised 'data sources' and 'Analyzing methods'. 

We have amended Fig 11 to include a line of best fit and added an appendix A1 with details of the regression results to give the further detail requested. P values were so close to zero (e.g. 0.000000000001) that it was clearer to use t values.  The reviewer's remark that some of the regions follow non-linear form was already noted in the text. We had opted to present the regression in very simple terms for two reasons: 1) the results are already highly significant with a very simple linear  regression, even though some non-linearity is a feature 2) a more complex regression analysis would lengthen the article and not add much extra insight regarding the ultimate value for beta (TEMP). However we do consider this a very good topic for future research and are working on a more involved methodologically focused paper. As we wish to keep this current paper focused on policy messages we prefer to keep the OLS as simple as possible here.

 

 

 

 

Reviewer 2 Report

Review of  climate-665617

 

“Staying cool in a warming climate: Temperature, electricity and air conditioning in Saudi Arabia”

Comments to author:

Thank you for the opportunity to review this manuscript. This paper examines trends in electricity consumption in Saudia Arabia related to rising temperatures and the use of air conditioner units. The paper overall is ambitious in scope, but they do a great job delivering the information in a concise fashion. Overall, this is a very useful contribution to the literature, because rising energy demand due to AC use is an understudied and overlooked area, but is poised to be very impactful in both areas of climate adaptation and mitigation. They use mostly descriptive quantitative methods, coupled with some simple statistical tests, to highlight the interplay of regional climate change, electricity consumption, and use of air conditioners. Furthermore, they integrate an analysis of existing, relevant policy in a qualitative manner that is useful for contextualizing their results. Last, the paper is also especially useful as it shows trends by region, highlighting the importance of considering variation at a sub-national scale.

The particular case of Saudia Arabia is especially interesting given 1) how hard its already being hit by rising temperatures, 2) its affluence, which allows for more and more residential AC unit purchases, 3) its unique policy to pay citizens to compensation for rising energy prices, 4) its prominence as an oil producing state. With an eye to these factors, the authors have done a very good job of introducing and justifying the importance of this study. Furthermore, their literature review is well written and connects their work well to previous studies.

The area that could use the most improvement is the description of the methods used – please refer to my comments below.

Thanks again – this was a pleasure to read.

 

Methods

Page 11, line 316 – before you start delving into the data, can you provide a brief overview of all the data sources you will be using, as well as what different analyses you will undertake? Right now the first paragraph reads as if the only thing you are empirically analyzing is temperature data. For example, explain that you’ll use correlation of hourly temp and power data, plus OLS regression modeling. Page 12, paragraph starting on line 354: Your correlation analysis is unclear – please specify more carefully what you correlated, what type of correlation test you used, and what your n was. Also, if you are throwing all years together, how do you account for changes by year? Page 12, line 359: It’s unclear what table A1 is? Is it in the appendix? Please specify. Page 12, lines 357-359: this is confusing, does the reader have to refer to Table 1 and Table A1 at the same time to be able to following this sentence? Page 12, lines 369-374: Please specify what it is you are going to do with the Enerdata. This is important to do before proceeding to results. Page 12 – put Table 1 in the results section, since it is a result.

Results

Why do you go from Figure 1 to Figure 8? What happened to Figures 2-7? Overall, figures are easy to comprehend and are presented well.

Discussion

I appreciated the nod to interdisciplinarity. I think the authors do a good job of partitioning their discussion by thematic grouping.

Minor points:

Page 2, lines 56-58: I think you can delete this paragraph since you describe this later in the paper (page 5) and indeed use the same phraseology. Page 5, line 159: I think you mean per capita income, not GDP. If you do mean GDP, please explain why you are comparing per capita GDP to the per capita income threshold mentioned above for household purchasing of AC units. Page 6, line 205: I suggest you use the word “contrasts” rather than “compares.” Also, please explain some possible causes for this discrepancy.

Author Response

We thank the reviewer for their positive comments and valuable suggestions to improve the paper.

As requested, we have expanded section 3: Materials and methods to more fully explain our data sources, especially regarding the energy and CO2 data as well as the temperature data. We have also added extra detail as requested regarding the analyzing methods, including detailed regression results, Pearson correlation test and sample size. 

We did not throw all the years together, but rather did the analysis for each year individually (presented in the Appendix) and took 2015 (our most recent year of data) as the example to highlight in the text. We have clarified the tables and text and how we use Enerdata to make this and other points relating to the analysis clearer as requested. We have also added a new appendix with details of the regression analysis and results.

Results: thank you for noticing this on line 520 the Figure was mislabelled figure 1 and we have updated this and other relevant figures. 

Page 2 lines 56-58: thank you for your suggestion, agree this paragraph can be deleted. 

Page 5 line 159: The two numbers 'per capita income' and 'per capita GDP' in 2015 purchasing power parity terms are the same. We have amended the text to refer only to per capita GDP and double checked with the original sources. 

Page 6, line 205: we agree that using 'contrasts' is more appropriate than 'compares' and we have explained in the text why these differences in the proportion of AC in 'electricity use' often arise across the literature.  

We have also made a small number of minor editing changes to improve the language and clarify important points. 

All changes have been made in track mode so can be easily identified by reviewers.