The United States Energy Consumption and Carbon Dioxide Emissions: A Comprehensive Forecast Using a Regression Model

Round 1

Reviewer 1 Report

Comments for author File: Comments.pdf

Author Response

Response to reviewer comments

Dear Reviewer,

We thank Reviewer for the valuable and helpful comments. The following are the replies to the comments and the related revision has been made in the manuscript.

Reviewer 1

In this manuscript, the authors summarized and discussed the total FFCO₂ emissions, total population, and GDP per capita using a regression mode. On this basis, they further we forecast the emissions until 2050 in consequent 5 years intervals. This work provides new insight and opinion into measures to reduce emissions through carbon capture and sequestration, and help in the development of improved GHG mitigations for the nation. The manuscript is well-organized and clearly stated. I would like to put forward some suggestions and concerns as follow:

1. In the introduction section,

(1) the author provides sufficient background information about GHG emissions,

CO₂ emissions caused by energy systems to show what create the climate change in the US. But there are no more details about the previous related researches around this topic, thus it is difficult to know what are the innovations and what is the difference between this paper and the previous ones.

Authors reply: We thank the reviewer for your valuable comments. We have added a few literatures of the previous studies in the introduction section and the comparison of results are shown in the discussions section.

(2) After the second paragraph in this part, figure 1 shows CO₂ contribution in the four major energy sectors including industry, transportation, residence and commerce while the accompanying words emphasizes the use and portion of fossil fuels. There is no correlation between the figure and the text.

Authors reply: We thank the reviewer for your valuable comments. We have changed it to an appropriate image.

2. In the data and methodology part,

(1) the author gives data collected from webpage and the reason to choose the

selected variables, but the specific process of using the SPSS tool to gain the descriptive statistics is neglected. And the meaning of the data displayed in the table should be more lucid.

Authors reply: We thank the reviewer for your valuable comments. We have added a few details to the paragraph. We have summarized the table in few lines because the mean and standard deviation are in common highly relative and the we have also spoken about the skewness and kurtosis of the entire table.

(2) At the end of this part, the author puts up forward another tool--mean absolute

percentage error (MAPE), and it is used to check the accuracy of the forecasted values. So the mathematical expression of MAPE and its usage should be put in this section ahead for methodology introduction. It is better to analyse the MAPE model results in the next part results and discussions directly.

Authors reply: We thank the reviewer for your valuable comments. We have changed the same in the manuscript.

3. In the results and discussions section, the author says LS and MI show high contribution when considered state-wise. But in the table 5, OH, IL and NY occupy more percentage of GDPPC(%) and TP(%) than LS, it might need a clearer explanation.

Authors reply: We thank the reviewer for your valuable comments. We have added a few words elaborating the contribution of variables.

4. There are some type-setting problems with the pictures.

(1) For example, the natural gas part in the third section mentions about figure 6 which showing total natural gas production and consumption in the US. But the picture is set for two paragraphs away the natural gas description.

Authors reply: We thank the reviewer for your valuable suggestion. We have moved the figures closer to the relevant text.

(2) Figure 7, table 7 and table 8 all share the same question, they need to get closer

to the words mentioning them.

Authors reply: We thank the reviewer for your valuable comments. We have moved the figures closer to the relevant text.

5. This paper lacks acknowledgement section.

Authors reply: We thank the reviewer for their valuable comments. We have added the same to the manuscript.

Reviewer 2 Report

The paper "The United States Energy Consumption And Carbon Dioxide Emissions: A Comprehensive Forecast Using A Regression Model" thoroughly forecast the carbon dioxide emissions in the energy sector. However, a few corrections need to be addressed before its acceptance and publication, as listed minor comments below:

Comments:

1.     Why has the author chosen the Regression model despite the other forecasting models? What is the significance?

2.     Figure 3 has data on the fossil fuels emitted sector-wise. It shows traces of ‘electricity’ as one of the fossil fuels in CO2 emissions. But why did the author not brief it in the manuscript?

3.     There are minor complex sentences that hinder readability. Corrections can be made to make it simpler.

4.     The measures of the forecasted EFF can be included in Table 6.

5.     Don't use abbreviations in the directly without explaining before.

Author Response

Response to reviewer comments

 Dear Reviewer,

We thank Reviewer for the valuable and helpful comments. The following are the replies to the comments and the related revision has been made in the manuscript.

 Reviewer 2

The paper "The United States Energy Consumption And Carbon Dioxide Emissions: A Comprehensive Forecast Using A Regression Model" thoroughly forecast the carbon dioxide emissions in the energy sector. However, a few corrections need to be addressed before its acceptance and publication, as listed minor comments below:

Comments:

1. Why has the author chosen the Regression model despite the other forecasting models? What is the significance?

Authors reply: We thank the reviewer for your valuable comments. We chose Linear regression over other models for forecasting time series like the Auto Regressive model, Moving Averages model and Autoregressive integrated moving average model because

1. They directly incorporate with the time component.
2. They are available on almost every statistical toolbox.
3. They have “few black boxes” and many resources available to help apply the technique.

2. Figure 3 has data on the fossil fuels emitted sector-wise. It shows traces of ‘electricity’ as one of the fossil fuels in CO₂ emissions. But why did the author not brief it in the manuscript?

Authors reply: We thank the reviewer for your valuable comments. We have not elaborated the details of electricity in the manuscript because electricity itself is a vast topic to be covered and the fossil fuel emissions for electricity is emerging to be replaced with renewable resources already. So we feel it is not more appropriate to choose electricity as well.

3. There are minor complex sentences that hinder readability. Corrections can be made to make it simpler.

Authors reply: We thank the reviewer for your valuable suggestions. The corrections are reflected in the manuscript.

4.The measures of the forecasted EFF can be included in Table 6.

Authors reply: We thank the reviewer for your valuable suggestions. The measures of Mmt CO₂^e is depicted in table 6.

5. Don't use abbreviations in the directly without explaining before.

Authors reply: We thank the reviewer for your valuable suggestion. We have added the state names with abbreviations prior to the explanation to make it easier to read.

Reviewer 3 Report

The article raises a relevant topic, but needs a lot of clarifications and improvements.

1) lines 31-32. "Energy consumption across the globe has more than doubled, eventually resulting in global warming". Strictly speaking, the causal relationship still has not been proven. This is only one of the hypotheses, based on a correlation. One of the other causes could be, for example, earthquakes (see [https://doi.org/10.3390/geosciences12100372](https://doi.org/10.3390/geosciences12100372))

2) lines 36 and 40 use Fahrenheit and Celsius scales. It would be better to reduce them to one scale.

3) Figure 1 must be improved. Using the bar chart with accumulation instead of pie charts might help.

4) The objective should be more focused. What is the need for such a forecast? What is new about it compared to IPCC projections and other studies?

5) Figure 2 is interesting but complex to read. State abbreviations are incomprehensible to anyone who does not live in the US. A lot of bars are very small and due to that fact useless (for instance, minerals and metals in figure 2B). It is recommended to reduce the number of components on the graph.

6) Figures 4 and 5 need adding measurement units

7) lines 332-333. "Yet some researchers have argued that it can be considered 332 a normal distribution if both values are between ±1.5." Needs references. 

8) My main concern is the methodology of econometric modelling. When modeling a time series, there are a number of mandatory tests that need to be done. For example, tests for stationarity, heteroscedasticity, autocorrelation. Without them the interpretation of regression estimates makes no sense, because all these effects can bias the results.

9) lines 399-401. "renewable energy has the potential to meet more than 80% of the country's electricity 400 demand". Your own figure 5B proves that's impossible. "Potential" can be estimated but the consumption of fossil fuels is stable. Some of the states of course can diversify their energy structure, but most of them have natural restrictions.

10) lines 422-423. "there are currently 28 large-scale carbon capture and storage projects operating or under construction worldwide". Indeed, there are investments and projects, but are there any big results? Would be nice to mention them. Also, see  DOI: 10.1038/s41558-020-00960-0

11) line 429. It would be useful to indicate the share of EVs in US.

12) Discussion section should be added. The main results should be compared with the other studies.

Author Response

Response to reviewer comments

Dear Reviewer,

We thank Reviewer for the valuable and helpful comments. The following are the replies to the comments and the related revision has been made in the manuscript.

Reviewer 3

The article raises a relevant topic, but needs a lot of clarifications and improvements.

1) lines 31-32. "Energy consumption across the globe has more than doubled, eventually resulting in global warming". Strictly speaking, the causal relationship still has not been proven. This is only one of the hypotheses, based on a correlation. One of the other causes could be, for example, earthquakes (see [https://doi.org/10.3390/geosciences12100372])

Authors reply: We thank the reviewer for your valuable comments. We have made the corrections to the introduction.

2) lines 36 and 40 use Fahrenheit and Celsius scales. It would be better to reduce them to one scale.

Authors reply: We thank the reviewer for your valuable comments. We have changed it to a single scale of Fahrenheit in the revised manuscript.

3) Figure 1 must be improved. Using the bar chart with accumulation instead of pie charts might help.

Authors reply: We thank the reviewer for your valuable suggestion. We have applied the suggestion in the manuscript.

4) The objective should be more focused. What is the need for such a forecast? What is new about it compared to IPCC projections and other studies?

Authors reply: We thank the reviewer for your valuable comments. We have added the same to the end of the introduction.

5) Figure 2 is interesting but complex to read. State abbreviations are incomprehensible to anyone who does not live in the US. A lot of bars are very small and due to that fact useless (for instance, minerals and metals in figure 2B). It is recommended to reduce the number of components on the graph.

Authors reply: We thank the reviewer for your valuable comments. We have changed it in the manuscript.

6) Figures 4 and 5 need adding measurement units

Authors reply: We thank the reviewer for your valuable suggestions. We have added the measurement units (thousand short tons) to the figures.

7) lines 332-333. "Yet some researchers have argued that it can be considered 332 a normal distribution if both values are between ±1.5." Needs references.

Authors reply: We thank the reviewer for your valuable comments. We have the following references to the main manuscript. (References: 69-71)

8) My main concern is the methodology of econometric modelling. When modeling a time series, there are a number of mandatory tests that need to be done. For example, tests for stationarity, heteroscedasticity, autocorrelation. Without them the interpretation of regression estimates makes no sense, because all these effects can bias the results.

Authors reply: We thank the reviewer for your valuable advice. We have utilized the t-test, F-test and adjusted R² but have missed the suggested tests. We assure to take it as a learning and change it in the future studies.

9) lines 399-401. "renewable energy has the potential to meet more than 80% of the country's electricity 400 demand". Your own figure 5B proves that's impossible. "Potential" can be estimated but the consumption of fossil fuels is stable. Some of the states of course can diversify their energy structure, but most of them have natural restrictions.

Authors reply: We thank the reviewer for your valuable comments. We agree to the aforesaid point to meet the level is complicated, but the lines are by National Renewable Energy Laboratory. It is said that countries like Germany have the potential to use such strategy due to the increasing renewable energy capacity.

As in reference No. 64. Brown, M. A.; Patel, D.; Frew, B. A.; Wiser, R. Renewable Energy Has the Potential to Meet More Than 80% of US Electricity Demand. IEEE J. Photovoltaics, 2018, 8, 793-798.

10) lines 422-423. "there are currently 28 large-scale carbon capture and storage projects operating or under construction worldwide". Indeed, there are investments and projects, but are there any big results? Would be nice to mention them. Also, see DOI: 10.1038/s41558-020-00960-0

Authors reply: We thank the reviewer for your valuable comments. To show an example of the big investments, we have added a few words on the achievements of the “Petra Nova CC project” by the United States. (References: 77,78)

11) line 429. It would be useful to indicate the share of EVs in US.

Authors reply: We thank the reviewer for your valuable comments. We have added a few words on the share and future growth of electric vehicles in the US.

12) Discussion section should be added. The main results should be compared with the other studies.

Authors reply: We thank the reviewer for your valuable comments. We have added comparisons in the discussions and future direction section.

Round 2

Reviewer 3 Report

I still think that the sentence "Energy consumption across the globe has more than doubled, eventually resulting in natural calamities like the earth-quakes, blizzards, heat and cold waves, glacier breaks and the current decade has seen continued warming of the Earth's climate system." might be corrected. It imposes that the energy consumption is the reason of natural calamities and global warming. In fact, we do know about the correlation between these facts. But the correlation is not the causal relationship. That's the difference. The increase in energy consumption might be the reason of global warming. Or might be one of the many reasons. Or might be not a reason at all. For example, there is some evidence that earthquakes, not human activity, may be the main cause of warming in Antarctica.

Author Response

I still think that the sentence "Energy consumption across the globe has more than doubled, eventually resulting in natural calamities like the earth-quakes, blizzards, heat and cold waves, glacier breaks and the current decade has seen continued warming of the Earth's climate system." might be corrected. It imposes that the energy consumption is the reason of natural calamities and global warming. In fact, we do know about the correlation between these facts. But the correlation is not the causal relationship. That's the difference. The increase in energy consumption might be the reason of global warming. Or might be one of the many reasons. Or might be not a reason at all. For example, there is some evidence that earthquakes, not human activity, may be the main cause of warming in Antarctica.

Author’s response: We thank the reviewer for the valuable comments. We understand the concept behind the reviewer’s suggestion, so we thought it might be better if the sentence is removed from the manuscript. After removing, it still does not hinder the concept of the writing.