Review Reports - Spatial Fairness and Changes in Transport Infrastructure in the Qinghai-Tibet Plateau Area from 1976 to 2016

Round 1

Reviewer 1 Report

Pretty nice paper but not entirely clear to me:

1. What is their goal? They state that they studied the distributive effects and the fairness of transport infrastructure…and so forth. Okay, they study ‘something’, but why? What are they aiming for? Support decision-making? Finding new scientific ways of research? See also my point 6

2. Additionally what is actually their scientific contribution in their view?

3. Highly unclear is how they operationalize ‘fairness’ and ‘distributive’ effects. They present modelling results of five indicators but I still do not know if transport infrastructure in this region has made the region fairer from an accessibility perspective. In lines 124 -126 they give some sort of definition, I think, but it seems only to indicate that they measure ‘accessibility changes’. Closest to some form of fairness measurement are their indicators 4 and 5 but I do not understand these entirely (e.g. lines 26 and 31 in the summary are highly unclear. This could be a language problem but I just do not understand these sentences. They seem also grammar wise incorrect). Lines 424 to 442 are a bit clearer to me but also puzzling (again, this could also be a language problem). I would recommend to write the summary and conclusions less ‘technical’. I would like to see that they try to interpret their results for non-Chinese readers in less technical language. Language which explains far clearer if the region has become fairer accessibility wise (or not). And why?

4. What is anchor city?

5. They speak sometimes of ‘accessibility dropped’ (e.g. line 200). I think, in contrast, they mean that accessibility increased though.

6. Related to point 1 and 3, after reading their final section you think as a reader ‘so what?’. What are the implications for policy-making and/or for science?

Author Response

Dear Reviewer:

Thanks for your comments concerning our manuscript entitled “Spatial fairness and changes in transport infrastructure in the Qinghai-Tibet Plateau area from 1976-2016” (manuscript ID: sustainability-389552). These comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Revised portion are marked in red in the paper. The main corrections in the paper and the responds to your comments are as flowing:

Point 1: What is their goal? They state that they studied the distributive effects and the fairness of transport infrastructure…and so forth. Okay, they study ‘something’, but why? What are they aiming for? Support decision-making? Finding new scientific ways of research? See also my point 6

Response 1: Thanks for your kind advice. We reorganized the literature and rewrote the introduction. Fairness is an important cornerstone of a legal and democratic society. In the past, the Chinese government made policies and planned transportation routes, it paid attention to the superiority of efficiency over fairness, but in recent years, it has gradually shifted to paying equal attention to efficiency and fairness, or even giving priority to fairness. The European and American scholars have earlier noticed the fairness of social resources, which is exactly what China's current regional planning lacks. China is currently implementing a "rural road" project in rural areas to replace dirt roads with cement pavements. In pastoral areas, however, such projects are much more complex because the herdsmen' residence is not fixed and the area is relatively sparsely populated.

Point 2: Additionally, what is actually their scientific contribution in their view?

Response 2: Thanks for your comments. We believe that fairness is not only a philosophical problem, but also a practical issue worth discussing to solve some social problems. In related studies of urban bus rapid transit (BRT) planning, scholars have discussed public participation in planning to enable ordinary people to benefit from the development of transportation. In other researches on fairness, scholars have explored the distribution of public facilities among different groups to solve the unfair distribution of social welfare. It can be said that fairness research plays a positive role in social development.

Point 3: Highly unclear is how they operationalize ‘fairness’ and ‘distributive’ effects. They present modelling results of five indicators but I still do not know if transport infrastructure in this region has made the region fairer from an accessibility perspective. In lines 124 -126 they give some sort of definition, I think, but it seems only to indicate that they measure ‘accessibility changes’. Closest to some form of fairness measurement are their indicators 4 and 5 but I do not understand these entirely (e.g. lines 26 and 31 in the summary are highly unclear. This could be a language problem but I just do not understand these sentences. They seem also grammar wise incorrect). Lines 424 to 442 are a bit clearer to me but also puzzling (again, this could also be a language problem). I would recommend to write the summary and conclusions less ‘technical’. I would like to see that they try to interpret their results for non-Chinese readers in less technical language. Language which explains far clearer if the region has become fairer accessibility wise (or not). And why?

Response 3: Thanks for your valuable comments.

All ambiguous statements have been revised (e.g. lines 26 and 31, lines 183 and lines 424 to 442).

By referring to the literature, we found that there were two broad ways to conceive of fairness: social fairness and spatial fairness. In this study, we focus on the spatial fairness, which concerns the distributive effect of transport infrastructure in space, while ignoring social factors (such as city size, residents’ traffic needs in different places).

Considering the environment characteristics of the Qinghai-Tibet Plateau area and the way of life of people [Tibetans, almost herdsmen, are seasonal rotational grazing, who graze at higher altitudes in the summer and at lower altitudes (usually along roads or around towns) in the winter], we believe that compared with social fairness, spatial fairness can better solve the current and future transportation planning and construction in a period of time. Due to the particularity of the plateau, the problem of backward transportation is the first thing that policy makers and planners should solve. Transport infrastructure provides the possibility for most Tibetans to access to the city. For example, in terms of economic benefits, the Qinghai-Tibet railway has little economic return, but it encourages Hoh Xil's children to go to school in Golmud (Golmud has much better education resources and much better living environment than Hoh Xil). I saw lots of pupils go to school in Golmud by train in Golmud railway station (these pupils live on campus during their schooling, and the government provides good living conditions for them).

Point 4: What is anchor city?

Response 4: Thanks for your comments. We are very sorry for our incorrect writing “anchor city (cities)”. “Anchor city (cities)” is derived from inaccurate expression in the literature, “Chen Y.; Jin F.; Lu Y.; et al. Development history and accessibility evolution of land transportation network in Beijing-Tianjin-Hebei region. Acta Geographica Sinica, 2017, 72”. “Anchor city (cities)” has been replaced with “major city (cities)” in this study.

Point 5: They speak sometimes of ‘accessibility dropped’ (e.g. line 200). I think, in contrast, they mean that accessibility increased though.

Response 5: Thanks for your kind advice. Actually, it should be “accessibility value dropped”, which means “accessibility time to the nearest major city decreased”. All statements have been revised.

Point 6: Related to point 1 and 3, after reading their final section you think as a reader ‘so what?’. What are the implications for policy-making and/or for science?

Response 6: Thanks for your kind advices, which are of great importance. We believe that the current research on traffic fairness at the urban scale is of great significance to discuss the sustainable development of cities. Similarly, the discussion traffic fairness on a larger scale is conducive to a comprehensive understanding of the impact of a country or region's transport development, and also has important guiding significance for balancing the unbalanced development between regions.

The Qinghai-Tibet Plateau area is ecologically fragile and not conducive to large-scale human economic activities (such as industrialization), but it is also the region in greatest need of development. Transportation provides people with the most basic travel possibilities and the opportunity to improve their living standards. As of 2016, there is only one Qinghai-Tibet Railway in the region. In the future, the region will also plan to build major transport infrastructure such as Sichuan-Tibet Railway and Xinjiang-Tibet Railway. The social impact of these transport infrastructures is worth further study.

Special thanks to you for your valuable comments!

Reviewer 2 Report

This paper uses a highly dubious technique to draw conclusions about distributional implications and fairness of transport accessibility on the Qinghai-Tibet Plateau. My major concern with the paper is that it purports to draw these conclusions using a ranking tool that, on my reading, takes no account of the size of various locations, or opportunities available at those locations, simply considering their travel time from various anchor locations. Thus being (say) 50 kilometres from a large centre has the same effect as being 50 kilometres from a much smaller location, even though the accessibility, in more usual usage of the term, would differ considerably between these two circumstances. Without taking the distribution of population (/opportunity) size into account, the paper is essentially calculating the relative accessibility, in a time sense, of a piece of dirt or location, compared to others, which has little relevance to anything of societal significance.

These weaknesses partly follow from the absence of any serious attempt in the paper to explain the authors' perspectives on distributive effects and fairness, apart from a definition of fairness as justice from another source. If they had thought more deeply about distributional effects and fairness, they might have grappled with the problem that their approach is about space not people, whereas both are needed to draw meaningful conclusions in this kind of analysis.

In addition to this flaw, which I believe is fatal, there are a number of other weaknesses.For example, the paper mainly talks about road travel times but also includes rail in its analysis, without saying how the two modes are treated in determining travel times. Is some weighting approach used, based on modal shares, or is the shortest travel time used, irrespective of mode? Or is something else done?

In discussing its measures of accessibility, the paper uses both time and distance in its elaboration, whereas the analysis is all about time. This poor use of language confuses the reader.

Author Response

Dear Reviewer:

Point 1: This paper uses a highly dubious technique to draw conclusions about distributional implications and fairness of transport accessibility on the Qinghai-Tibet Plateau. My major concern with the paper is that it purports to draw these conclusions using a ranking tool that, on my reading, takes no account of the size of various locations, or opportunities available at those locations, simply considering their travel time from various anchor locations. Thus being (say) 50 kilometres from a large centre has the same effect as being 50 kilometres from a much smaller location, even though the accessibility, in more usual usage of the term, would differ considerably between these two circumstances. Without taking the distribution of population (/opportunity) size into account, the paper is essentially calculating the relative accessibility, in a time sense, of a piece of dirt or location, compared to others, which has little relevance to anything of societal significance.

Response 1: Thanks for your kind advices. First, for Tibetans (almost herdsmen) living on the Qinghai-Tibetan plateau area, their lifestyle is seasonal rotational grazing, who graze at higher altitudes in the summer and at lower altitudes (usually along roads or around towns) in the winter. For them, the round-trip time is more of a concern than the size of the city. Second, we concern about horizontal fairness, that is, the regional distribution and spatial effects of transport infrastructure. Transport infrastructure provides the possibility for most Tibetans to access to the city. Last, in the survey, we found that the vast majority of herdsmen reached the farthest place, is the nearest city. Unless there is a special event (such as a critical illness, pilgrimage), only a few people will go to further cities. Therefore, we believe that for people in this region, the time of accessibility to the major city is more important than the size of the city. However, it must be admitted that the size of a city, especially the quality and quantity of service facilities provided by different cities, especially medical services, will also have a great impact on people's travel choices. This needs further study.

Point 2: These weaknesses partly follow from the absence of any serious attempt in the paper to explain the authors' perspectives on distributive effects and fairness, apart from a definition of fairness as justice from another source. If they had thought more deeply about distributional effects and fairness, they might have grappled with the problem that their approach is about space not people, whereas both are needed to draw meaningful conclusions in this kind of analysis.

Response 2: Thanks for your valuable comments. Concepts of distributive effects and fairness still do not have a single, widely accepted definition. In this study, we adopt the viewpoints from Forth et al. and Delbosc et al., and define the spatial fairness of traffic as that transport infrastructure provides benefits evenly for all groups (Page 2).

Point 3: In addition to this flaw, which I believe is fatal, there are a number of other weaknesses. For example, the paper mainly talks about road travel times but also includes rail in its analysis, without saying how the two modes are treated in determining travel times. Is some weighting approach used, based on modal shares, or is the shortest travel time used, irrespective of mode? Or is something else done?

Response 3: Thanks for your comments. In data processing, first, the ordinary road network was completely intersected, and then the highway entrances and exits and train stations were connected with the ordinary road network, so as to obtain the shortest accessibility time between nodes. In addition, considering the actual situation in the region and the focus of our study, we did not adopt weighting approach. In our field survey in 2016, we found that few herdsmen in Longbao town of Yushu had been to the provincial city - Xining, and most of them had been to the farthest city, the nearest city, Yushu city. Therefore, we only focused the shortest travel time.

Point 4: In discussing its measures of accessibility, the paper uses both time and distance in its elaboration, whereas the analysis is all about time. This poor use of language confuses the reader.

Response 4: Thanks for your comments. Actually, we used one measure of accessibility, the time accessibility model. What happens was that when we looked up literatures, we didn't pay enough attention to the differences between different languages. Some Chinese scholars regard time as a kind of distance and time as a kind of cost, hence the expression of “time distance”. And we had tried our best to revise the language in the study.

Special thanks to you for your valuable comments!

Reviewer 3 Report

This is a very interesting research. However, it is strongly recommended to revise the manuscript by addressing the following comments before considering it for publication.

· In the abstract ‘CV value’ should be explained/defined.

· In the introduction the following terms should de defined: spatial fairness, traffic fairness, fairness of transport infrastructure, and anchor city (since these terms are extensively used in the paper).

· In the introduction I do not think the last paragraph is needed unless it is required by the journal.

· There should be separate sections for study site, data source/data collection, and research methods.

· I suggest to use the term, ‘This study’ instead of This paper’.

· The used data should be listed in tabular format and explained further.

· There should be a figure that shows all major administrative boundaries along with the anchor cities and transport networks. It would be better to have the temporal variation of transport network development shown in figures.

· The quality of all figures should be improved, especially the image resolution. It is extremely difficult to read the text and comprehend the figure properly due to poor quality.

· Figure 1 should only use one map in larger scale or use the other as an index map.

· At line 183, the sentence is incomplete.

· In addition to the above comments the authors should provide the detail explanation about the techniques used to convert the results obtained from network analysis to 2D spatial distribution in the study area.

Author Response

Dear Reviewer:

Point 1: In the abstract ‘CV value’ should be explained/defined.

Response 1: Thanks for your comments. The abbreviation of the coefficient of variation value is CV value. The statement has been revised.

Point 2: In the introduction the following terms should be defined: spatial fairness, traffic fairness, fairness of transport infrastructure, and anchor city (since these terms are extensively used in the paper).

Response 2: Thanks for your kind advice, which is of great importance. The concept of fairness still does not have a single, widely accepted definition. We choose spatial fairness o instead of social fairness, because the former focuses on the possibility of travel and accessibility to major cities provided by transport infrastructure, regardless of the different abilities of each person. And we adopt the viewpoints from Forth et al. and Delbosc et al., and define the spatial fairness of traffic as that transport infrastructure provides benefits evenly for all groups.

“Anchor city” has been replaced by “major city”. “Anchor city (cities)” is derived from inaccurate expression in the literature, “Chen Y.; Jin F.; Lu Y.; et al. Development history and accessibility evolution of land transportation network in Beijing-Tianjin-Hebei region. Acta Geographica Sinica, 2017, 72”.

Point 3: In the introduction I do not think the last paragraph is needed unless it is required by the journal.

Response 3: Thanks for your kind advice. The last paragraph in the introduction has been deleted from this study.

Point 4: There should be separate sections for study site, data source/data collection, and research methods.

Response 4: Thanks for your valuable comments. The structure of the article has been adjusted as follows: introduction, methodology, study area and data, results and analysis, discussion, and conclusion.

Point 5: I suggest to use the term, ‘This study’ instead of This paper’.

Response 5: Thanks for your kind advice. All statements have been corrected.

Point 6: The used data should be listed in tabular format and explained further.

Response 6: Thanks for your kind advice. “Table 1” has been added in page 5.

Point 7: There should be a figure that shows all major administrative boundaries along with the anchor cities and transport networks. It would be better to have the temporal variation of transport network development shown in figures.

Response 7: Thanks for your kind advice. We have redrawn Figure 1, and put the temporal variation of transport network development in Figure 3.

Point 8: The quality of all figures should be improved, especially the image resolution. It is extremely difficult to read the text and comprehend the figure properly due to poor quality.

Response 8: Thanks for your kind advice. All drawings in the document have been re-inserted. All maps have a resolution of 1000 dpi (dots per inch).

Point 9: Figure 1 should only use one map in larger scale or use the other as an index map.

Response 9: Thanks for your kind advice. We have redrawn a colour map (Figure 1).

Point 10: At line 183, the sentence is incomplete.

Response 10: Thanks for your kind advice. We have been revised the sentence.

Point 11: In addition to the above comments the authors should provide the detail explanation about the techniques used to convert the results obtained from network analysis to 2D spatial distribution in the study area.

Response 11: Thank you for the comments. Please refer to the first paragraph of section 4.1. (Page 6), “…we used spatial interpolation and the inverse distance weighting (IDW) model in ArcGIS software to obtain the spatial pattern and characteristics…”, and section 4.2. (Page 7), “Using ArcGIS zoning statistical functions…”. All steps are implemented in the interpolation model of ArcGIS software.

Special thanks to you for your valuable comments!

Round 2

Reviewer 2 Report

This paper is an improvement on the original version but needs more to be done.

First, the English needs to be improved. The paper should be edited by someone whose first language is English.

Second, there are some particular matters that need to be improved.

- The wording 'traffic fairness' is most unusual, more common wording being transport equity or fairness. I suggest this change in wording. I have written and reviewed numerous articles in the transport equity space and have never heard of 'traffic fairness'.

- The discussion between lines 112-117 seems to have the C's the wrong way around. If travel time is the indicator and it is very small, then a low value of C will result. Yet the paper asserts that if C>1, accessibility is better than the regional average. Doesn't C>1 mean that travel time is longer than the regional average, which means accessibility is worse. I think the language is confusing travel time and accessibility.

- Table 2. What verification has been undertaken on these very important speed assumptions?

- Section 4.3.2. This section talks repeatedly about a place called Ali, which I could not find on the map. Presumably there are other places also cited that are not located on the map. This makes it hard for the reader to know what the authors are talking about.

- line 355 says "As can be seen from Figure 7,...'. These things cannot be seen from the figure because the places are not marked on the maps.

- The short contextual discussion between lines 378-381 is interesting and would benefit from being placed early in the paper, when the region is described.

- line 411 talks about the overall accessibility level of the plateau decreasing. Again, as with a previous comment above, I think you mean travel time is decreasing, which means that accessibility is increasing.

Author Response

Dear Reviewer:

Thanks for your comments concerning our manuscript entitled “Spatial fairness and changes in transport infrastructure in the Qinghai-Tibet Plateau area from 1976 to 2016” (manuscript ID: sustainability-389552). These comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our researches. We have studied comments carefully and have made correction which we hope meet with approval. Revised portion are marked in red in the paper. The main corrections in the paper and the responds to your comments are as flowing:

Response to Reviewer 2 Comments

Point 1: First, the English needs to be improved. The paper should be edited by someone whose first language is English.

Response 1: Thanks for your kind advices. We have followed your suggestion and sent the manuscript to MDPI for English editing service. Meanwhile, we will try our best to standardize English writing in future studies.

Point 2: Second, there are some particular matters that need to be improved. The wording 'traffic fairness' is most unusual, more common wording being transport equity or fairness. I suggest this change in wording. I have written and reviewed numerous articles in the transport equity space and have never heard of 'traffic fairness'.

Response 2: Thanks for your valuable comments. This is a very important comment. We have replaced ‘traffic fairness’ with ‘transport fairness’. There are significant differences in the meanings of equity, fairness and justice in Chinese. But ‘traffic’ and ‘transport’ mean almost the same thing in Chinese. These suggest that we did not grasp the concepts more deeply. In future studies, we will do everything we can to avoid such mistakes again.

Point 3: The discussion between lines 112-117 seems to have the C's the wrong way around. If travel time is the indicator and it is very small, then a low value of C will result. Yet the paper asserts that if C>1, accessibility is better than the regional average. Doesn't C>1 mean that travel time is longer than the regional average, which means accessibility is worse. I think the language is confusing travel time and accessibility.

Response 3: Thanks for your comments. The relevant text has been modified, “ > 1, indicates that the accessibility level of the node is below the regional average level; if < 1, the accessibility level of the node is better than the regional average level”. We did not fully understand the logical difference between Chinese writing and English writing, which would be our focus on improvement in the future.

Point 4: Table 2. What verification has been undertaken on these very important speed assumptions?

Response 4: Thanks for your kind advice. The Technical Standard of People's Republic of China Highway Engineering (JTGB01–2003) stipulates the design speed of various existing roads, and Chen Y., et al studies roads in different periods in the Beijing-Tianjin-Hebei Region (Chen Y.; Jin F.; Lu Y.; et al. Development history and accessibility evolution of land transportation network in Beijing-Tianjin-Hebei region. Acta Geographica Sinica, 2017).

Meanwhile, we communicated with several professors (Professor Li, Professor Cao and Professor Sha) living in the area, all of whom are professors of Qinghai Normal University. Qinghai Normal University is located in Xining, provincial capital of Qinghai. Lanzhou and Xining have about 220 kilometres of railway, which took four to five hours in the 1980s. Around the year 2000, it took about 15 hours to drive from Xining to Golmud (about 770 kilometres). These are important references for speed assumptions.

Point 5: Section 4.3.2. This section talks repeatedly about a place called Ali, which I could not find on the map. Presumably there are other places also cited that are not located on the map. This makes it hard for the reader to know what the authors are talking about.

Response 5: Thanks for your comments. We noticed this problem and marked the place names in the text as much as possible on the map.

Point 6: line 355 says "As can be seen from Figure 7,...'. These things cannot be seen from the figure because the places are not marked on the maps.

Response 6: Thanks for your kind advice. We have marked the place names in the text as much as possible on the map, so that people can find them and have a better understanding of the Qinghai-Tibet Plateau area.

Point 7: The short contextual discussion between lines 378-381 is interesting and would benefit from being placed early in the paper, when the region is described.

Response 7: Thanks for your comments. The lifestyle of people in the Qinghai-Tibet Plateau area, especially Tibetans, is different from that of people in both urban areas and rural areas. Tibetans, almost herdsmen, exhibit seasonal rotational grazing. In the summer of 2015, I did a geological survey in the mountains around Qinghai Lake, where I met many herdsmen. They live in tents and live on wheaten food (such as zanba, roasted highland barley flour), meat (mainly mutton and yak beef) and tea with milk. In winter, they live in the brick houses along roads or around towns.

Point 8: line 411 talks about the overall accessibility level of the plateau decreasing. Again, as with a previous comment above, I think you mean travel time is decreasing, which means that accessibility is increasing.

Response 8: Thanks for your valuable comments. All ambiguous statements have been revised.

Special thanks to you for your valuable comments!

Sincerely,

Gao Xingchuan

Li Tao

Cao Xiaoshu

Author Response File: Author Response.pdf

Reviewer 3 Report

The revised manuscript has been improved substantially. However, the authors still need to revise the manuscript further by addressing the following comments:

· The ‘Methodology’ section should come after the ‘Study Area and Data’ section.

· More explanation is needed about the techniques used to convert the results obtained from network analysis to 2D spatial distribution in the study area. There should be detail discussions about the used interpolation technique (IDW). Why this interpolation technique was used? This is the simplest interpolation technique but not the most accurate one.

· The colors of Figure 1 need to have explanation.

· In Figure 6, why there are two color bars in the legend? It is confusing.

Author Response

Dear Reviewer:

Response to Reviewer 3 Comments

Point 1: The ‘Methodology’ section should come after the ‘Study Area and Data’ section.

Response 1: Thanks for your comments. The ‘Methodology’ section and the ‘Study Area and Data’ section have been adjusted.

Point 2: More explanation is needed about the techniques used to convert the results obtained from network analysis to 2D spatial distribution in the study area. There should be detail discussions about the used interpolation technique (IDW). Why this interpolation technique was used? This is the simplest interpolation technique but not the most accurate one.

Response 2: Thanks for your kind advice, which is of great importance. The Available interpolation tools commonly used in ArcGIS include: IDW (the inverse distance weighting model), Kriging, Natural Neighbor and Spline. Because IDW is a weighted distance average, the average cannot be greater than the highest or less than the lowest input. As accessibility is considered to be closely related to the distance, we used the IDW model for the interpolation. Meanwhile, we compared the results of several models and found that results from IDW are more consistent with the actual situation.

Point 3: The colors of Figure 1 need to have explanation.

Response 3: Thanks for your kind advice. We have redone the legend of figure 1.

Point 4: In Figure 6, why there are two color bars in the legend? It is confusing.

Response 4: Thanks for your valuable comments. We have redone the legend of figure 6 and made the necessary explanation.

Special thanks to you for your valuable comments!

Sincerely,

Gao Xingchuan

Li Tao

Cao Xiaoshu

Author Response File: Author Response.pdf