Parks, Green Space, and Happiness: A Spatially Specific Sentiment Analysis Using Microblogs in Shanghai, China

Round 1

Reviewer 1 Report

Dear Authors,

The manuscript is well written and could address all concerns regarding the hypothesis, there are several points which requires your attention.

I would suggest a revision prior the acceptance.

- Usually in scientific articles the third person style uses.

- Figure 2 Please improve the photo quality.
- Figure 3 what was the reason to omit the most important factor "gender" from the model? do you believe the gender has not have any role on happiness?

- Line 263: happiness is an imaginary value "conceptual factor" how could you distinguish happiness among the participants. how participant could express if the happiness directly conected to the park not other daily events.

- Discussion and conclusions: how did you eliminate other factors which would directly or indirectly influence on the results such as wealth, job, stress, kids, etc.

- reference: please follow the journal structure to correct the style.

Thank You,

Regards

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript used SEM to evaluate the factors that affect the sentiment in parks based on social media data. I recommend a major revision.

(1) I cannot agree with the statement “few studies …, and few have further …” in lines 9-11 of Abstract and the similar sentence in lines 36-37 of Introduction. Many relative studies have been conducted, some of which are listed below. The authors should do a more comprehensive literature review.

Seresinhe C I, Preis T, MacKerron G, Moat H S (2019) Happiness is greater in more scenic locations. Scientific Reports, 9: 4498

Wang Z, Zhu Z, Xu M, Qureshi S (2021) Fine-grained assessment of greenspace satisfaction at regional scale using content analysis of social media and machine learning. Science of the Total Environment, 776: 145908

Ghahramani M, Galle N J, Ratti C, Pilla F (2021) Tales of a city: sentiment analysis of urban green space in Dublin. Cities, 119: 103395

Gao Y, Chen Y, Mu L, Gong S, Zhang P, Liu Y (2022) Measuring urban sentiments from social media data: a dual‑polarity metric approach. Journal of Geographical Systems, 24: 199-221

(2) In figure 1, parks are visualized in a cluttered way with both green polygons and colored triangles. I think it would be better to only use two colors of polygons to represent parks.

(3) What is the criterion to classify the frequency of park visit?

(4) This manuscript used time cost to improve traditional G2SFCA. Some work has been conducted to employ time cost in 2SFCA approaches. I recommend some references as following,

Xia N, Cheng L, Chen S, et al. (2018). Accessibility based on gravity-radiation model and Google Maps API: A case study in Australia. Journal of Transport Geography, 72, 178–190.

Tao Z, Yao Z, Kong H, Duan F, Li G. (2018). Spatial accessibility to healthcare services in Shenzhen, China: Improving the multi-modal two-step floating catchment area method by estimating travel time via online map APIs. BMC Health Services Research, 18(1), 345.

Tao Z, Cheng Y. (2019). Modelling the spatial accessibility of the elderly to healthcare services in Beijing, China. Environment & Planning B, 46(6), 1132–1147

García-Albertos P, Picornell M, Salas-Olmedo M H, Gutiérrez J. (2019). Exploring the potential of mobile phone records and online route planners for dynamic accessibility analysis. Transportation Research Part A: Policy & Practice, 125, 294–307.

Zheng Z, Xia H, Ambinakudige S, et al. (2019). Spatial accessibility to hospitals based on web mapping API: An empirical study in Kaifeng, China. Sustainability, 11(4), 1160.

Gong S, Gao Y, Zhang F, Mu L, Kang C, Liu Y (2021). Evaluating healthcare resource inequality in Beijing, China based on an improved spatial accessibility measurement. Transactions in GIS, 25(3): 1504-1521.

(5) In equations (1) and (2), accessibility is calculated through walking time. However, the walking time to a park does not change much at different times of the day, so what is the value of cost time in this model? On the contrary, driving time is affected by “traffic congestion, and other conditions” (line 298), so people will spend different amounts of time when they drive to the same park at different times of the day. How did the authors determine the driving time that are reliably comparable to walking time? Meanwhile, I am not sure whether biking time is affected by those conditions.

(6) Also in equations (1) and (2), what is the spatial unit of origin j?

(7) Some details of this work should be clarified. (a) How many Weibo posts are inside and outside parks, respectively? Are the numbers balanced and comparable? (b) How are residents and non-residents distinguished, and what is the accuracy?  (c) What is the precision of satellite images classification (lines 317-319)?

(8) What is the advantage of SEM over correlation analysis when assessing the effects of factors? Why did the authors emphasize causality rather than correlation? Why are the relationships in table 5 causalities instead of correlations?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

In the manuscript entitled “Parks, Green Space, and Happiness: a spatially specific sentiment analysis using microblogs in Shanghai, China” authors tried to link happiness and well-being with the urban park / green space. I found it an interesting work and rightly pointed to enhancing the green space over the urban area. The manuscript is well-written and has the quality to be published, but I have some general queries and it would be better if the authors address and incorporate them into the final version of the manuscript.

A)    Authors may improve Figure 4 – It is not clear what x and y axis explains (add axis title)

B)    Authors are saying total 60K users in the captions of Fig 4. Can they break out the users i.e. outside and inside? Why outside park has a smoother line in comparison to the inside park? -Is it coming from the difference in the users?

C)    The time authors used here is Jan 2021 - June 2021. It is the time when the whole world is facing travel restrictions due to COVID. The manuscript is silent about it. Please comment and add it.

D)    Authors may incorporate weather, Pollution and Urban Heat Island formation in the city and link them to work.

E)     Lone 381-382: Authors are saying 18 months of observations but in line 214 it is 181 days. Figure 4 data is also showing 6 months of observations.  Is it not contradictory or something else? Please clarify it.

Some relevant publications to improve the manuscript are:

·         Pandey, A. K., Singh, S., Berwal, S., Kumar, D., Pandey, P., Prakash, A., Lodhi, N., Maithani, S., Jain, V. K. and Kumar, K.: "Spatio - temporal variations of urban heat island over Delhi", Urban Climate, 10(P1), 119–133, 2014  https://doi.org/10.1016/j.uclim.2014.10.005

·         Pandey, A. K., Kumar, R. P., & Kumar, K. (2015, October). Satellite and ground-based seasonal variability of NO2 and SO2 over New Delhi, India. In Remote Sensing of Clouds and the Atmosphere XX (Vol. 9640, pp. 212-218). SPIE.

·         Pandey, A. K., Rawal R. S., Gairola S., Bhatt I. D., Kumar R. P., Priyadarshini N., and Kumar A.: “Protection from Anthropogenic Disturbances Contributed to the Recovery of Vegetation in the Kumaon Himalaya: A Case Study” International Journal of Geology, Earth & Environmental Sciences, 7(2), 39–50, 2017

I congratulate the authors for a good job and recommend a minor revision of this manuscript before publication.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear Authors,

Thanks for efforts which improved the manuscript quality. All concerns were considered and the correction are well applied.

thanks for clarifying the methodology and additional information added to manuscript.

Thanks.

Author Response

Dear reviewer,

We really appreciate your comments.

Thanks.

Reviewer 2 Report

The manuscript has been well revised. I have two minor comments.

(1) The authors said “The spatial unit of origin j is subdistrict” in their response. However, I did not find the word of “subdistrict” in the main text. The authors should clearly descript the spatial unit in section 2. 

(2) The response to Point 7 should also be explained in the manuscript for clarification.

Author Response

Please see the attachment.

Author Response File: Author Response.docx