Evaluation of Selected Empirical Models for Asphalt Pavement Temperature Prediction in a Tropical Climate: The Case of Ghana

Round 1

Reviewer 1 Report

This study evaluated 15 five empirical asphalt pavement temperature prediction models developed by Lukanen et al. [1], 16 Park et al. [2], Diefenderfer et al. [3], Taamneh [4] all in the United States and Asefzadeh et al. [5] in 17 Canada to ascertain their prediction accuracy in tropical climate, using Ghana as a case study. This study is interesting. The following two aspects should be noted.

1) Can references in the abstract be replaced by model names? Try not to include references in the abstract.

2) In Figs. 7-12, the value of R2 is small. Please add the related illustration. 

Minor editing of English language required.

Author Response

1) References in the abstract have been replaced with the surname of the first author of the model developers, where necessary.

2) Thanks for your suggestion, but it is not clear whether it is the font size that is small or whether the reviewer is alluding to something different. Either way, the R² value has been made more visible and an explanatory sentence has been provided on the low R² value.

3) The entire manuscript has been thoroughly reviewed to address such minor language issues.

Reviewer 2 Report

In general, it is not common practice to include citations in the abstract of a research paper. The abstract is meant to provide a concise summary of the study's objectives, methodology, key findings, and conclusions. Citations are typically reserved for the main body of the paper, where you can elaborate on the background and related work.

The abstract needs to be more descriptive and informative. It should briefly mention the research objective, the methodology used, and the main findings. The current abstract is too brief and doesn't give readers enough information about the study.

The references in the abstract part should be removed. 

The introduction should provide more context and background information on the importance of asphalt pavement temperature prediction in tropical climates, especially in Ghana. It should also highlight the limitations of existing models developed for temperate climates and the need for evaluating their applicability in tropical climates.

The methodology section should be more detailed, providing information on data collection methods, sample size, and statistical analysis procedures used for model evaluation.

The discussion section should delve deeper into the implications of the results and highlight the practical implications of using the Park et al. [2] model in the tropical climate of Ghana. It should also discuss potential areas for improvement in the Park et al. [2] model or the possibility of developing a local model.

Check the consistency in presenting equations, tables, and figures. Ensure that they are all numbered properly and are referred to correctly in the text.

The entire article doesn't look like a review article / analysis part has not been performed.

On each case of discussion, the authors don't specify about the number of samples.

Review papers are valuable contributions to scientific literature as they summarize and critically evaluate existing research in a specific field, providing insights, identifying gaps, and suggesting future research directions. Provide your justification?

Review the language and grammar throughout the article for clarity and coherence.

Author Response

1) References in the abstract have been replaced with the surname of the first author of the model developers, where necessary.

2) Thanks for your suggestion, but it is not clear whether it is the font size that is small or whether the reviewer is alluding to something different. Either way, the R² value has been made more visible and an explanatory sentence has been provided on the low R² value.

3) The entire manuscript has been thoroughly reviewed to address such minor language issues.

Author Response File: Author Response.pdf

Reviewer 3 Report

This study examines the applicability of five pavement temperature forecast models in Ghana, which has significant practical implications. However, you must consider these issues.

(1) Are citations in the abstract required? The references for these five models don't need to be cited again in the conclusion.

(2) Figure 1 should be replaced by a more precise image.

(3) It is advised that the introduction be strengthened to emphasize the climatic features of Ghana.

(4) If the results are applicable to areas with a climate comparable to Ghana, like Togo and Cote d'Ivoire, it is advisable to include a proposal in the conclusion.

Author Response

1) The citations of the models in the abstract and conclusions have been removed.

2) Figure 1 has been replaced with a clearer image and it shows the climatic zones of Ghana, and the locations (cities) where data were collected for the study. We consider this to be informative enough to orient readers. However, we be glad to consider any specific recommendation the reviewer may have. 

3) A summary of Ghana’s climatic characteristics has been provided in the introduction section.

4) The following sentence has been added to the conclusions:

The findings of this study may be generally applicable to countries with similar climatic characteristics as the Forest and Savannah climatic zones of Ghana, particularly nearby West-African countries such as Togo, Benin, Nigeria, Cote D’Ivoire, Liberia, Sierra Leone, and Burkina Faso

Reviewer 4 Report

This paper analyzes the applicability of five asphalt pavement temperature prediction models in Ghana. The results show that the Park et al. model is well applicable in Ghana. Please note the following specific comments.

1. Page 6, Figure 1 - The illustration is slightly blurred and it is recommended that it be replaced. In addition, the figure shows that the coastal climate zone is also an important part of Ghana. Should this area also be studied?

2. Page 6, lines 204-207 - "The roads were within 3 km of a Ghana Meteorological Agency (GMet) synoptic weather station...". Why are these three characteristics used as selection criteria?Page 7, lines 214-216 - A diagram of the installation of a digital thermometer in the drilled hole at the mid-depth of the pavement should be provided.

3. Only one sample is tested in both areas. Can the results be considered representative?

4. Figure 3-12, What is the meaning of the directional arrow in the illustrations?

5. This paper has analyzed the prediction results of the five temperature prediction models. It is recommended to supplement the analysis of the reasons for the significant differences in the five models.

6. In the conclusion of this paper, there seems to be no highlight. The author only briefly describes the prediction results of these five commonly used empirical models, making it difficult to reflect the innovation of this paper.

Moderate editing of English language required.

Author Response

1) Figure 1 has been replaced with a clearer image. Thanks for your note regarding the coastal climate. Unfortunately, resources were not available to include the coastal belt in the current study. Notwithstanding, climate conditions are less severe along the coast, and one may reason that models applicable to the forest zone might as well be applicable in the coastal zone.

2) The selection of the study roads was not based on these criteria. It is our attempt to provide a general description of the nature of the study roads. We could have selected any other asphalt roads. Regardless, the flat terrain ensured safety of the data enumerators.

2b) Thanks for this useful suggestion, but unfortunately, apart from the AC layer thickness, information on the granular layers was not readily available to allow a sketch of the pavement cross-section showing the thermometer installation depths. However, we endeavored to improve the introduction of Figure 2, which contains a photo of the in-situ pavement temperature measurement set up.

3) In the opinion of the authors, the single samples are considered to be representative of the study zones.

4) The directional arrows distinguish between the line of equality and the regression lines on respective Figures.

5) The section on summary of model evaluation results has been strengthened to include possible factors accounting for the low prediction accuracy of the Diefenderfer [11], Taamneh [12] and Asefzadeh [13] models.

6) The following sentence has been added to the conclusions:

The findings of this study may be generally applicable to countries with similar climatic characteristics as the Forest and Savannah climatic zones of Ghana, particularly nearby West-African countries such as Togo, Benin, Nigeria, Cote D’Ivoire, Liberia, Sierra Leone, and Burkina Faso

7) The entire manuscript has been thoroughly reviewed to address such minor language issues.

Round 2

Reviewer 2 Report

1. The introduction and methods sections are relatively clear, but the presentation of results and discussion lacks structure.

2. Why are subheadings required in Section 2.1 to have references?

3. Please review the figure captions for Figure 2 and 3. They look similar.

4. The abstract should provide a concise summary of the key findings and implications of the study.

5. Dataset information about the instances and features is not clear.

6. Please provide justification for the data availability statement: "The data is available upon request." Since this work has already been reported, why are the authors not disclosing a publicly accessible link?

7. Address the limitations of the study. Were there any constraints or assumptions made during the research that could have affected the results?

8. The article lacks a clear concluding section that summarizes the main findings and their significance. Consider adding a concluding section to tie together the key points and implications of the study.

9. Proofread the article for grammar and style issues. Some sentences are long and complex, which can make them difficult to follow. Simplify sentences, when possible, for better readability.

10. In the discussion section, provide a more detailed interpretation of the results. For example, explain why certain models performed better or worse in specific climatic zones. Discuss the practical implications of these findings for pavement engineering in Ghana.

11. Why references have been made on figure captions. Is it the reported results?

The article requires major revision.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 3

Reviewer 2 Report

1. The article briefly describes the data collection process, including the instruments used and the locations chosen for data collection. However, it lacks comprehensive details regarding the sampling strategy, measurement frequency, and data quality control. Providing more information on these aspects would enhance the reproducibility and reliability of the study.

2. While the article discusses the criteria used for model evaluation (statistical significance, error statistics, LOE, and R-squared), it doesn't provide a clear rationale for these criteria's selection. A more detailed explanation of why these specific criteria were chosen and how they relate to the study's objectives would be beneficial.

3. The article presents various error statistics and scatter plots to evaluate model performance. However, it lacks a comprehensive interpretation of these results. A detailed discussion of what the error statistics reveal about each model's accuracy and how well they align with the study's objectives is needed.

4. Please suggest directions in the results and discussion part / conclusion for future research or areas where improvements in pavement temperature prediction models are needed.

English proofreading is required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf