Strategic Planning for Sustainable Urban Park Vitality: Spatiotemporal Typologies and Land Use Implications in Hangzhou’s Gongshu District via Multi-Source Big Data

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Recommendation: Major Revision

Manuscript Number: land-3602389

Title: Strategic Planning for Sustainable Urban Park Vitality: Spatiotemporal Typologies and Land Use Implications in Hangzhou’s Gongshu District via Multi-source Big Data

 

The idea of a framework to research on urban park vitality of spatiotemporal typologies was interesting. The manuscript has done a good preliminary work. I think that the descriptions of some points were inadequate. I recommend that a major revision is warranted. I explain my concerns in more detail below. I ask that the authors specifically address each of my comments in their response.

 

1) How is the term "land use" in the title concretely and systematically represented in the research?

2) What were the research questions in this manuscript? Is it “to construct a ‘temporal behavior-spatial attribute-vitality typology’ framework, analyzing periodic patterns and spatial differentiation mechanisms across park vitality types” from the Line 124 to 126? How are the mechanisms and results of spatial differentiation explained in the manuscript?

3) What core result was expressed in Figure 4? The language of the illustration is not clear. It is suggested to revise it.

4) In the discussion, some activity types and POI types were explained, and these activity types and POI types are related to the demographic attributes of the activities. Can this content be further deepened in the results or discussions?

5) Why can this research promote the understanding of optimizing resource efficiency and public space equity, which mentioned in Abstract?

6) It is suggested to further explain the innovativeness and significance of contributions of this manuscript.

7) The figures’ resolution was not high, and it is recommended to change the figures.

Author Response

Comment 1: How is the term "land use" in the title concretely and systematically represented in the research?

 

Response 1: Thank you for your comment. The term "land use" in the title is systematically represented in the research through three interconnected dimensions: First, classifying parks based on temporal vitality to guide zoning priorities. Morning-Exercise-Dominated: highlights the need for flexible zoning (e.g., allocating space for fitness trails, transit hubs) and temporal land management (e.g., morning-focused amenities). All-Day-Balanced: supports mixed-use development (e.g., integrating parks with commercial/transit hubs via TOD strategies). Evening-Aggregation-Dominated: requires dense residential–commercial interfaces (e.g., prioritizing evening facilities like outdoor seating, lighting). Second, quantifying spatial drivers (e.g., accessibility, POIs) to optimize land use allocation. Third, proposing adaptive policies (e.g., mixed-use TOD, temporal zoning) that align park design with urban sustainability goals. We added about land use in Lines 137–139 as follows: This study embeds land use theory into park vitality analysis through three dimensions: functional zoning, spatial allocation and temporal regulation. Furthermore, we have added Section 4.1 titled Land Use Planning and Sustainability Enhancement based on vitality types in Chapter 4 (Conclusion).

 

Comment 2: What were the research questions in this manuscript? Is it “to construct a ‘temporal behavior-spatial attribute-vitality typology’ framework, analyzing periodic patterns and spatial differentiation mechanisms across park vitality types” from the Line 124 to 126? How are the mechanisms and results of spatial differentiation explained in the manuscript?

 

Response 2: Thank you for pointing this out. The Geographical Detector method (Section 2.3.3) inherently reveals causal mechanisms by quantifying driver dominance. Figure 5 (Spatial Distribution of Vitality Types) already provides critical visual evidence for spatial differentiation mechanisms.

The descriptions in Lines 126–137 outline the research methods and objectives, and we have made revisions accordingly:

The study addresses three core questions:

1. How do the temporal vitality patterns of urban parks vary across different time periods (weekdays vs. weekends) and park typologies?
2. What are the key spatial attributes and their interactions that drive the spatiotemporal differentiation of park vitality?
3. How can time-sensitive planning strategies be developed to align park design with dynamic user demands and sustainable resource allocation?

This research integrates multi-source big data to construct a "temporal behavior–spatial attribute–vitality typology" framework that transcends traditional static planning theories and finds that vitality differentiation is influenced by the interaction of factors rather than being determined by a single factor. Additionally, the Geodetector model quantifies the dynamic impacts of 14 factors, providing a replicable technical procedure.

 

Comment 3: What core result was expressed in Figure 4? The language of the illustration is not clear. It is suggested to revise it.

 

Response 3: Thank you for your comment. Figure 4 is the parallel coordinate map of the characteristic index of each activity type park. The core result of Figure 4 is provided in Lines 445-449: Morning Exercise-Dominated parks exhibit distinct daytime peak clusters characterized by high morning peak ratios; All-Day-Balanced parks (predominantly comprehensive parks) display stable visitation profiles with lower fluctuation entropy values; Evening-Aggregation-Dominated parks (largely special-category parks) experience sharp post-18:00 peaks, with high evening activity levels.

For clarity of expression, we move the sentence “Typological indices are presented in Figure 4, with spatial distributions mapped in Figure 5.” to Line 467 and add “can be observed in Figure 4” in Line 468.

Moreover, the title of Figure 4 is supplemented: “(MPR: Morning Peak Ratio, DNDC: Day-Night Difference Coefficient, EAL: Evening Activity Level, FLE: Fluctuation Entropy, PVDC: Peak-Valley Difference Coefficient. Data are standardized by Z - score, and the larger the value, the more significant the characteristics of the indicator are.)”

 

Comment 4: In the discussion, some activity types and POI types were explained, and these activity types and POI types are related to the demographic attributes of the activities. Can this content be further deepened in the results or discussions?

 

Response 4: Thank you for suggesting the direction for extending the research. Due to the anonymized processing of signaling data, this study did not obtain direct demographic labels such as age and gender. However, through analysis and inference from existing data, we have enhanced this section by adding content before the optimization strategies for each park type, and we think we can study this topic in our future work:

Lines 624–630: On weekdays, morning-exercise-dominated parks are mostly community parks and riverside green spaces that are, relatively close to the city center and located in high-density residential areas. Field research shows that their facilities primarily include fitness trails and rest pavilions, aligning with the behavioral patterns of middle-aged and elderly groups. The interaction effect of POI density and PFD (Park Facil-ity Density) is stronger on weekdays than on weekends. To further enhance aging-friendly attributes, future improvements could involve adding community medical facilities.

Lines 646–648: The cross-time-period balance of all-day-balanced parks may result from the diversity of their served populations, including both middle-aged and elderly morning exercisers, weekend parent–child families, and young tourists.

Lines 689–693: The strong correlation between evening vitality and commercial POIs (on week-ends, SPOI q = 0.7693; on weekdays, SPOI q = 0.6281) indicates that young office workers and families are the main users. Such groups prefer composite spaces of "park + convenience store + parent-child facilities". It is recommended to add smart food pickup lockers and temporary child care areas in the facility configuration.

Lines 756–758: Future research can build a finer population-space interaction model through multi-source data fusion, such as signaling age tags + street view image analysis.

 

Comment 5: Why can this research promote the understanding of optimizing resource efficiency and public space equity, which mentioned in Abstract?

 

Response 5: Thank you for your comment. This research advances the understanding of optimizing resource efficiency and enhancing public space equity through three aspects: temporal optimization of resources, spatial equity through accessibility and facility, and typology-based planning for design. By exposing temporal–spatial mismatches in park usage and linking them to land use, accessibility, and facility distribution, the research provides a replicable blueprint to optimize efficiency: align resources with time-specific demand, and reduce waste.

We revised the Conclusion as follows (Lines 764-784): The three vitality types identified through DTW-KMeans clustering highlight spatiotemporal resource imbalances in urban park systems: morning-dominated parks experience low evening usage, while night-aggregating parks underutilize facilities during the day (as detailed in Figure 4). This temporal inefficiency underscores shortcomings in standardized planning approaches. Notably, Geodetector analysis shows that surrounding POI density (q=0.47–0.77) influences vitality 2–5 times more strongly than traditional area metrics (q=0.15–0.38), proving that spatial inequalities amplify service exclusion through temporal compounding.

The “temporal behavior–spatial attribute–vitality typology” framework advances park planning by replacing static indicators with dynamic demand-based models. Incorporating vitality typologies into land use and transportation strategies can shift urban development from expansion-focused to efficiency-driven, fostering sustainable symbiosis between cities, parks, and communities. Based on mobile signaling data, the methodology offers adaptable models for data-informed park management and siting, particularly supporting urban park planning in high-density cities and global urban regeneration efforts.

By analyzing how park vitality arises from the strategic alignment of time, space, and human behavior, this study redirects sustainability planning; integrating hourly usage data enables parks to adapt to evolving community needs while reducing resource waste. For high-density urban areas, this approach provides a blueprint to transform green spaces into resilient, inclusive assets through temporal responsiveness and multi-factor optimization.

 

Comment 6: It is suggested to further explain the innovativeness and significance of contributions of this manuscript.

 

Response 6: Thank you for your suggestion. To emphasize the innovativeness and significance of contributions of our study, we revised the Introduction in Lines 133–137 as follows: This research integrates multi-source big data to construct a "temporal behavior–spatial attribute–vitality typology" framework that transcends traditional static planning theories and finds that vitality differentiation is influenced by the interaction of factors rather than being determined by a single factor. Additionally, the Geodetector model quantifies the dynamic impacts of 14 factors, providing a replicable technical procedure.

 

Comment 7: The figures’ resolution was not high, and it is recommended to change the figures.

 

Response 7: Thank you for pointing this out. We have replaced the old images with new, high-resolution images (Figures 1, 2, 3, 5, 6, 7, and 8), with a resolution of 1000 dpi.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Based on the research framework of "temporal behavior-spatial attributes-park typology", this paper analyzed the vitality types of 59 parks in Gongshu District, Hangzhou, and explored the influencing factors of park vitality based on geographic detector analysis. The theme of this article is interesting, but there are some issues that need further revision.

1. This article involves the collection of a large amount of basic data. However, the article did not fully list all the basic data sources.
2. In formula (3), the symbol of visitation volume is inconsistent with formulas (1) and (2). It is recommended to express it uniformly.
3. What data did the authors analyze when calculating landscape indices related to parks (such as LSI, WP)? What is the spatial resolution of these data?
4. Formula (13) is incorrect. On the one hand, this formula is not for calculating LSI, but for calculating patch shape index. The meanings of these two indicators are completely different. On the other hand,  the patch shape index is calculated based on the perimeter-area ratio, rather than the area-perimeter ratio in formula (13). The error in this calculation formula may affect the accuracy of all subsequent analyses.
5. In L354，how to identify city centers using nighttime light data? The author needs to provide a detailed explanation.
6. In L362， what is the grid size?
7. The Discussion section of this article is almost entirely focused on the management and planning strategies of parks in Hangzhou, but there is no in-depth interpretation of the innovation of this research or reasons for the results of this article. For me, the writing style in this section is more like a project report than an academic paper. So, what new knowledge can international readers gain from this study? 
8. The Conclusion section of this article is too lengthy. At the same time, it is not appropriate to describe research limitations in this section.

Author Response

Comment 1: This article involves the collection of a large amount of basic data. However, the article did not fully list all the basic data sources.

 Response 1: Thank you for your valuable suggestion. We fully recognize the importance of clarifying data sources for research credibility. To address this issue, we have systematically supplemented the data source information in the manuscript. Specifically, we revised the title of Section 2.2 to Multi-source Data and Processing to better reflect the content, and added a detailed description of various basic data sources in the newly expanded text (Lines 182–206) as follows: The mobile signaling data were provided by China Mobile Zhejiang Company and anonymized to protect user privacy, with a spatial resolution of 50-meter grids and hourly temporal accuracy. The park vector data were sourced from the publicly available “Hangzhou Urban Green Space System Plan” issued by the Hangzhou Planning and Natural Resources Bureau.

POI (point of interest) data were collected from AMAP (https://map.gaode.com/) in 2023, and used after coordinate correction and duplicate removal. Hangzhou Gongshu District POI information was obtained via the AMAP API (https://lbs.amap.com/api/android-sdk/guide/map-data/poi/). This API enables web crawler programs to gather data, yielding attributes such as IDs, names, categories, descriptions, coordinates, addresses, longitudes, and latitudes. Data preprocessing included coordinate correction, filtering, duplicate removal, and deletion of incomplete records, resulting in 57,946 valid POI entries for Gongshu District. The average POI density is 25.6 points per hectare. Notably, the number of POIs within a 1,200-meter buffer zone around parks reaches 32,458, meeting the sample size requirements for statistical analysis (each park corresponds to an average of 550 POI records). Eight primary categories were identified as positively influencing park usage: Food and Beverage, Shopping and Consumption, Hotels and Accommodation, Science, Education, and Culture, Lifestyle Services, Healthcare, Leisure and Entertainment, and Sports and Fitness.

Water data were derived by interpreting water area boundaries using 0.5-meter-resolution remote sensing imagery. Road network data were sourced from the OpenStreetMap open-source map data platform, encompassing some minor roads and internal roads within communities. Population density was calculated using a permanent population identification model based on signaling data and calibrated with the Hangzhou Statistical Yearbook 2022.

 

Comment 2: In formula (3), the symbol of visitation volume is inconsistent with formulas (1) and (2). It is recommended to express it uniformly.

Response 2: Thank you for your comment. In Formulas (1) and (2), the symbol for visitation volume is P. However, in Formula (3), it was erroneously written as Pop—both should denote the same variable. The corrected formula is as follows:

（The formula here is not visible; please refer to the Word file for details.）

Comment 3: What data did the authors analyze when calculating landscape indices related to parks (such as LSI, WP)? What is the spatial resolution of these data?

Response 3: We apologize for the confusion. When calculating the Landscape Shape Index (LSI), ArcGIS 10.2 was used to extract the perimeter and area from park vector boundaries. For calculating Water Proportion (WP), the water area within parks was derived from 0.5-meter-resolution remote sensing imagery obtained via Google Earth. The WP is calculated as the ratio of water area to the total park area. Following the revisions to Section 2.2 above, we have added the following content on Lines 396-399: The LSI is calculated based on park vector boundaries.

Water proportion (WP) is calculated after interpreting the water area boundaries from 0.5-meter-resolution remote sensing imagery.

 

Comment 4: Formula (13) is incorrect. On the one hand, this formula is not for calculating LSI, but for calculating patch shape index. The meanings of these two indicators are completely different. On the other hand, the patch shape index is calculated based on the perimeter-area ratio, rather than the area-perimeter ratio in formula (13). The error in this calculation formula may affect the accuracy of all subsequent analyses.

Response 4: Thank you for pointing this out. The original Formula (13) was indeed conceptually confusing. We have carefully revised Formula (13), referenced literatures[1][2], recalculated the data, and updated Table 2 and Table 3, and the corresponding Figure 6, Figure 7, and Figure 8.

 

Comment 5: In L354，how to identify city centers using nighttime light data? The author needs to provide a detailed explanation.

Response 5: Thank you for your comment. Wulin Square is officially designated as Hangzhou’s city center by local authorities and has been validated through POI and night-time light data in prior research. In this paper, Wulin Square is used as the city center to calculate the parameter of "distance to the city center", so the text does not elaborate on how to identify the urban core. We have provided the reference and the rationale for using Wulin Square as the city center on Lines 410–413 as follows: According to the Territorial Spatial Master Plan of Hangzhou (2021–2035) and the research on identifying urban centers through the integration of POI (Point of Interest) and NTL (Nighttime Light) data[3], Wulin Square serves as the city center of Hangzhou.

 

Comment 6: In L362， what is the grid size?

Response 6: Thank you for your comment, and we apologize for the confusion. The grid file approach was proposed in the initial draft of our experimental plan, but the actual implementation utilized the ArcGIS Network Analysis functionality. Corrections have been made on Lines 421–422 to reflect this: Using the network analysis tool in ArcGIS, the range of walking accessibility for each park was calculated.

 

Comment 7: The Discussion section of this article is almost entirely focused on the management and planning strategies of parks in Hangzhou, but there is no in-depth interpretation of the innovation of this research or reasons for the results of this article. For me, the writing style in this section is more like a project report than an academic paper. So, what new knowledge can international readers gain from this study?

Response 7: Thank you for your suggestion. We agree with this comment. The existing content does indeed focus on planning applications, and its innovative contributions require strengthening by mentioning its international relevance. Therefore, we have added Section 4.2 (Lines 739–751) to the Discussion section, as follows: 4.2. Implications and Innovations

The innovative value of this study goes beyond the Hangzhou case, bringing three keys advancements to the international urban planning field.

Theoretically, it reveals for the first time that park vitality emerges from the dynamic interaction of time-based needs, spatial features, and social groups—a relationship that should be central to planning. Methodologically, integrating 50-meter-resolution signaling data with the DTW algorithm creates a practical framework for improving the accuracy of micro-scale vitality assessments in cities. In practice, time-adaptive approaches (such as designing flexible spaces for evening activity) can help developing countries make the most of limited green spaces while aiding developed nations in upgrading public areas for aging communities.

These findings enrich our understanding of Sustainable Cities and Communities (SDG 11) and drive a shift in public space research—moving from a focus on supply-driven design to one that actively responds to people’s behaviors and needs.

 

Comment 8: The Conclusion section of this article is too lengthy. At the same time, it is not appropriate to describe research limitations in this section.

Response 8: Thank you for your suggestion. We agree with this comment. The Conclusion has been streamlined to focus on key findings, with research limitations relocated to a dedicated section (4.3) for better clarity. Therefore, we have revised the Conclusion section as follows (Lines 764 to 784): The three vitality types identified through DTW-KMeans clustering highlight spatiotemporal resource imbalances in urban park systems: morning-dominated parks experience low evening usage, while night-aggregating parks underutilize facilities during the day (as detailed in Figure 4). This temporal inefficiency underscores shortcomings in standardized planning approaches. Notably, Geodetector analysis shows that surrounding POI density (q=0.47–0.77) influences vitality 2–5 times more strongly than traditional area metrics (q=0.15–0.38), proving that spatial inequalities amplify service exclusion through temporal compounding.

The “temporal behavior–spatial attribute–vitality typology” framework advances park planning by replacing static indicators with dynamic demand-based models. Incorporating vitality typologies into land use and transportation strategies can shift urban development from expansion-focused to efficiency-driven, fostering sustainable symbiosis between cities, parks, and communities. Based on mobile signaling data, the methodology offers adaptable models for data-informed park management and siting, particularly supporting urban park planning in high-density cities and global urban regeneration efforts.

By analyzing how park vitality arises from the strategic alignment of time, space, and human behavior, this study redirects sustainability planning; integrating hourly usage data enables parks to adapt to evolving community needs while reducing resource waste. For high-density urban areas, this approach provides a blueprint to transform green spaces into resilient, inclusive assets through temporal responsiveness and multi-factor optimization.

 

Furthermore, we moved the research limitations to Section 4.3 as follows (Lines 752 to 762):

4.3. Limitations and Future Work

This study acknowledges four primary limitations: first, excluding park visits with a dwell time of <30 minutes may underestimate visitor scale. Second, population com-position analysis focuses solely on residential/working demographics, lacking subdivided visitor characteristics (gender, age, income). Future research can build a finer population-space interaction model through multi-source data fusion, such as signal-ing age tags + street view image analysis. Third, the factor selection relies on a literature review and qualitative methods, potentially missing unknown determinants due to knowledge gaps. Fourth, the data collection included a midweek light rain Qingming Festival day that was treated as a weekday, limiting insights into holiday visitation patterns; future studies will incorporate May Day/National Day holiday data to investigate seasonal variations.

 

[1] Jiang, W.; Wang, Y.; Zhang, M. Exploring the Industrial Heat Island Effects and Key Influencing Factors in the Guangzhou–Foshan Metropolitan Area. Sustainability 2025, 17, 3363, doi:10.3390/su17083363.

[2] Sun, R.; Chen, L. How Can Urban Water Bodies Be Designed for Climate Adaptation? Landsc. Urban Plan. 2012, 105, 27–33, doi:10.1016/j.landurbplan.2011.11.018.

[3] Lou, G.; Chen, Q.; He, K.; Zhou, Y.; Shi, Z. Using Nighttime Light Data and POI Big Data to Detect the Urban Centers of Hangzhou. Remote Sens. 2019, 11, 1821, doi:10.3390/rs11151821.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

My suggestions are as follows.

 

In my opinion, the authors should elaborate on whether the quality and quantity of POIs used for statistical analysis are sufficient to support their quantitative study.

 

I think the problem with research is that it makes few comparisons. In connection with the results, the authors should discuss the extent to which these findings can be generalised to a broader Chinese context—for example, whether they are valid for other cities.

 

The manuscript could include further discussion of park-use issues in Chinese cities and any minor conflicts that may arise.

 

A map showing the case study city’s location within China should be included to orient international readers.

 

Most figures are of too low a resolution to read; please provide higher-resolution versions.

 

Comments on the Quality of English Language

The English is generally clear and understandable. Some sentences could be tightened and vocabulary refined for a more professional style. 

Author Response

Comment 1: In my opinion, the authors should elaborate on whether the quality and quantity of POIs used for statistical analysis are sufficient to support their quantitative study.

 Response 1: Thank you for your suggestion. The POI data were obtained through the AMAP API, containing detailed fields such as names, coordinates, and classifications. After manual deduping and coordinate correction (with an error margin of <50 meters), the accuracy rate exceeds 95%, ensuring reliable sources, high accuracy, reasonable classification, and good alignment with park needs. A total of 57,946 valid POI records were collected in Gongshu District, with an average POI density of 25.6 per hectare. Among them, the number of POIs within the 1,200-meter buffer zone around parks reached 32,458. Both the quantity and density of POI data meet the sample size requirements of this study. Therefore, we supplemented the content in Section 2.2 to elaborate in detail on the POI data from data quality and data quantity:

Lines 187–200:

POI (point of interest) data were collected from AMAP (https://map.gaode.com/) in 2023, and used after coordinate correction and duplicate removal. Hangzhou Gongshu District POI information was obtained via the AMAP API (https://lbs.amap.com/api/android-sdk/guide/map-data/poi/). This API enables web crawler programs to gather data, yielding attributes such as IDs, names, categories, descriptions, coordinates, addresses, longitudes, and latitudes. Data preprocessing included coordinate correction, filtering, duplicate removal, and deletion of incomplete records, resulting in 57,946 valid POI entries for Gongshu District. The average POI density is 25.6 points per hectare. Notably, the number of POIs within a 1,200-meter buffer zone around parks reaches 32,458, meeting the sample size requirements for statistical analysis (each park corresponds to an average of 550 POI records). Eight primary categories were identified as positively influencing park usage: Food and Beverage, Shopping and Consumption, Hotels and Accommodation, Science, Education, and Culture, Lifestyle Services, Healthcare, Leisure and Entertainment, and Sports and Fitness.

 

Comment 2: I think the problem with research is that it makes few comparisons. In connection with the results, the authors should discuss the extent to which these findings can be generalized to a broader Chinese context—for example, whether they are valid for other cities.

Response 2: Thank you for your comment and suggestion. The generalizability of the research findings can be discussed from three perspectives: urban commonalities, regional particularities, and promotion recommendations. As a core city in the Yangtze River Delta, Hangzhou’s urbanization stage is similar to that of most first- and second-tier cities in China, particularly in terms of high-density built environments, accelerating aging populations, and the separation of work and residence, making it a representative of these aspects. However, Hangzhou’s "Park City" construction policy may enhance the spatiotemporal equilibrium of vitality, while cities lacking policy guidance may exhibit more pronounced temporal differences. Therefore, we have revised Section 4.2 as follows:

Lines 718–732: As a core city in the Yangtze River Delta, Hangzhou's urbanization stage is similar to that of most first- and second-tier cities in China, particularly in terms of high-density built environments, accelerating aging population, and the separation of work and residence, making it representative in these aspects. The vitality characteristics and influencing factors identified in this study have also been validated in other cities such as Shanghai [7], Guangzhou [35], and Fuzhou [30]. Regarding regional specificity, Hangzhou's "Park City" construction policy may enhance the spatiotemporal equilibrium of vitality, while cities lacking policy guidance may exhibit more pronounced temporal differences. The spatiotemporal coupling mechanisms discovered in this study hold reference value for rapidly urbanizing areas in China. For example, high-density cities such as Guangzhou and Wuhan can draw lessons from the community linkage strategies of 'evening gathering-type' parks, while leisure-oriented cities such as Chengdu and Xi'an can optimize the commercial supporting facilities of comprehensive parks. However, less developed regions should pay attention to the fundamental role of transportation accessibility and avoid blindly copying the eastern model.

 

Comment 3: The manuscript could include further discussion of park-use issues in Chinese cities and any minor conflicts that may arise.

Response 3: Thank you for your comment and suggestion. This study deepens the understanding of conflicts in supply–demand mismatch, generational needs, and management policies. Land scarcity in first-tier cities has led to a proliferation of "mini-parks" and a shortage of large-scale parks. While older adults prefer quiet morning exercise spaces, younger populations have diversified needs, causing functional competition in community parks (e.g., complaints about square dance noise in a Hangzhou community park increased by 15% in 2023 compared to 2022). Additionally, some cities implement a "one-size-fits-all park closing time" policy (e.g., uniform closure at 21:00), which suppresses evening vitality potential and contrasts with this study’s recommendation to "extend opening hours for suburban parks." Therefore, we have provided the following information in the Section:

Liness 733–738: Chinese urban parks face a conflict between "standardized management and diversified needs," as fee-based parks generally close earlier than the peak of residents' evening activities, resulting in loss of facility utilization. In the future, it is necessary to promote "one district, one policy" management—for example, allowing early opening of morning exercise areas in aging communities and piloting night-time opening in zones with concentrated young populations.

 

Comment 4: A map showing the case study city’s location within China should be included to orient international readers.

Response 4: Thank you for pointing this out. We agree that adding a locator map will significantly improve the geographical context for international readers. We have reproduced Figure 1.

(Figure: please refer to the Word file for details.)

 

Comment 5: Most figures are of too low a resolution to read; please provide higher-resolution versions.

Response 5: Thank you for pointing this out. We have replaced the old images with new, high-resolution images (Figures1, 2, 3, 5, 6, 7, and 8), with a resolution of 1000 dpi.

 

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

 

This study constructs and applies a "temporal behavior-spacial attrbutes-park typology" framework. It applies this partially new approach to 59 parks in the Chinese Hangzhou Gongshu district.

 

The manuscript attracts mixed comments from a science point of view:

strong and positive aspects include:

• The significant amount of research on SD relevant criteria.
• The attempt to combine these criteria in complex indices, reflecting the multi-disciplinar character of the problem.

Weaker points include:

. the difficult interpretation of the resulting indices.

. The application potential and repeatability of the resulting methods.

. The heterogeneity of the resulting data and their sustainability implication.

The "Introduction" offers a clear and convincing motivation of the study, based on previous, published works.

The extended "Materials and Methods"-section provides an idea about the high number of heterogeneous data that is the basis for the proposed method.

Figure 3 is difficult to read due to the small characters used.

The "Results"-section introduces the "Park Vitality Typologies" as a concept. 

Figures 6-9 should be explained in the text. The abcises can hardly be read.

The "Discussion" explains the theoretical applicability of the method. The logical links with the "Results" are unclear. The text entails only selected elements of a classical dission.

The "Conclusions"-section puts emphasis on the strengths and limitations of the developed method.

 

Author Response

Comment 1: the difficult interpretation of the resulting indices.

Response 1: Thank you for pointing this out. To enhance clarity, we have (1) added plain-language descriptions for each index definition in 2.3.1 part as follows:

Lines 264–265:

The higher the ratio, the more concentrated the morning exercise crowd.

Line 271-274:

A ratio greater than 1 indicates busier daytime activity (e.g., office workers’ lunch breaks), while a ratio less than 1 signifies more active nighttime engagement (e.g., residents taking walks after work). The calculation formula is as follows:

Lines 278–280:

The higher the value, the greater the potential of the park's night-time economy (e.g., 0.3 indicates that 30% of visitors are active at night). The calculation formula is as follows:

Lines283–287:

The fluctuation entropy value is used to evaluate the regularity of human flow fluctuations. A higher entropy value indicates more random visitor distribution, such as being evenly dispersed throughout the day; a lower entropy value signifies stronger periodicity, such as being crowded only during morning and evening peaks.

Lines 293–296:

The peak–valley difference coefficient is used to characterize the disparity between peak and trough human flow volumes, similar to a "tidal phenomenon." The larger the ratio, the more crowded the peak periods and the emptier the trough periods become.

 

 (2) explicitly stated the practical interpretation of each index range in 3.1 part as follows:

Lines 448–464:

The temporal characteristic indicators and their thresholds are as follows: The morning peak ratio reflects the intensity of morning exercise demand, with values <0.3 indicating low demand; 0.3–0.5 signifying typical community parks serving both morning exercise and daily leisure; and >0.5 denoting professional morning exercise venues that may require additional morning-oriented facilities. The day-night difference coefficient measures activity balance: <0.8 indicates night-dominated parks, 0.8–1.2 signifies balanced day-night activity, and >1.2 reflects predominantly daytime use. For nighttime vitality, <0.25 indicates low nighttime activity, 0.25–0.35 represents basic nighttime demand, and >0.35 signals high potential for night-time economy development. The fluctuation entropy value evaluates flow regularity: <2.5 indicates strong periodicity (concentrated in specific periods), 2.5–3.2 denotes moderate fluctuation, and >3.2 suggests random all-day distribution (potentially indicating high facility utilization). The peak-valley difference coefficient characterizes crowding disparity: <5 implies stable flow with low facility pressure, 5–10 indicates significant tidal effects, and >10 reflects extreme peak-valley gaps requiring crowd-diversion measures. These thresholds, derived from clustering results of 59 parks and validated through field observations, provide a rapid diagnostic tool for planning decisions to optimize park management and resource allocation.

 

Comment 2: The application potential and repeatability of the resulting methods.

Response 2: Thank you for your comment. The core methodology (index calculation + DTW-K-means) is domain-agnostic and can be applied to any spatiotemporal activity dataset. Furthermore, we have added content on the extent in Discussion to which the findings of this study can be generalized to broader Chinese contexts (whether they are equally applicable to other cities) and implications for international readers as follows:

Lines 717–751:

4.2. Implications and Innovations

As a core city in the Yangtze River Delta, Hangzhou's urbanization stage is similar to that of most first- and second-tier cities in China, particularly in terms of high-density built environments, accelerating aging population, and the separation of work and residence, making it representative in these aspects. The vitality characteristics and influencing factors identified in this study have also been validated in other cities such as Shanghai [7], Guangzhou [35], and Fuzhou [30]. Regarding regional specificity, Hangzhou's "Park City" construction policy may enhance the spatiotemporal equilibrium of vitality, while cities lacking policy guidance may exhibit more pronounced temporal differences. The spatiotemporal coupling mechanisms discovered in this study hold reference value for rapidly urbanizing areas in China. For example, high-density cities such as Guangzhou and Wuhan can draw lessons from the community linkage strategies of 'evening gathering-type' parks, while leisure-oriented cities such as Chengdu and Xi'an can optimize the commercial supporting facilities of comprehensive parks. However, less developed regions should pay attention to the fundamental role of transportation accessibility and avoid blindly copying the eastern model.

Chinese urban parks face a conflict between "standardized management and di-versified needs," as fee-based parks generally close earlier than the peak of residents' evening activities, resulting in loss of facility utilization. In the future, it is necessary to promote "one district, one policy" management—for example, allowing early opening of morning exercise areas in aging communities and piloting night-time opening in zones with concentrated young populations.

The innovative value of this study goes beyond the Hangzhou case, bringing three keys advancements to the international urban planning field.

Theoretically, it reveals for the first time that park vitality emerges from the dynamic interaction of time-based needs, spatial features, and social groups—a relation-ship that should be central to planning. Methodologically, integrating 50-meter-resolution signaling data with the DTW algorithm creates a practical framework for improving the accuracy of micro-scale vitality assessments in cities. In practice, time-adaptive approaches (such as designing flexible spaces for evening activity) can help developing countries make the most of limited green spaces while aiding developed nations in upgrading public areas for aging communities.

These findings enrich our understanding of Sustainable Cities and Communities (SDG 11) and drive a shift in public space research—moving from a focus on sup-ply-driven design to one that actively responds to people’s behaviors and needs.

 

Comment 3: The heterogeneity of the resulting data and their sustainability implication

Response 3: Thank you for your comment. We agree with this comment. Therefore, we have enhanced the discussion of heterogeneity and strengthened sustainability linkages as follows:

Lines 465–466: The three vitality patterns reveal the heterogeneity of crowd behavior, which give rise to differentiated resource pressures.

Line 598: 4.1. Land Use Planning and Sustainability Enhancement based on Vitality Types

 

Comment 4: Figure 3 is difficult to read due to the small characters used.

 Response 4: Thank you for pointing this out. We have redrawn Figure 3 as follows:

（Figure: please refer to the Word file for details）

 

Comment 5: The "Results"-section introduces the "Park Vitality Typologies" as a concept.

Response 5: Thank you for your comment. To more clearly elaborate on the theoretical foundation and classification logic of the vitality typology, we have added the following content to the first paragraph of Section 3.1, "Urban Park Vitality Typologies" as follows:

Lines 436–438: Based on time geography theory [45], the park vitality typology reveals the spatiotemporal consistency of vitality patterns through dynamic time-series similarity clustering.

 

Comment 6: Figures 6-9 should be explained in the text. The abscises can hardly be read.

Response 6: Thank you for pointing this out. We agree with this comment. Therefore, we have redrawn Figures 6-8 and added an explanation on Lines 506–515 (about Figure 6): Interaction detection, as shown in Figure 6, revealed the following synergistic effects among factors: the interactions of SPOIs with park type (q=0.7485), area (q=0.7567), and D-30 (q=0.8221) exerted the strongest influence on morning peaks, driven by consumption–leisure linkages in parks such as Banshan National Forest Park. For evening activity, interactions of SPOIs with facility density (q=0.763) and W-15 (q=0.8056) dominated, reflecting night market/retail synergies. The interactions of D-30 with park area (q=0.806) and SPOIs (q=0.8156) intensified pulse visits in large parks, exacerbating flow fluctuations. These findings highlight the critical role of comprehensive parks and surrounding amenities in mediating weekend visitation patterns, with implications for park planning and resource allocation.

（Figure: please refer to the Word file for details）

Given that the analytical logic for weekday and weekend results is consistent, and to avoid redundancy in length, the Geodetector results for weekends are described separately for factor detector and interaction detector outcomes, while the weekday Geodetector results present an integrated analysis of both types of results (see Lines 512–531).

（Figure: please refer to the Word file for details）

Moreover, we would like to clarify that Section 3.3 ("Influencing Factors of Vitality Types in Different Time Periods") is dedicated to explaining the findings visualized in Figure 8. To improve clarity, we have made the following revisions in the manuscript: we have added explicit references to Figure 8 within Section 3.3 to strengthen the connection between the text and the visualization.

Lines 551–569: By analogizing the results of the factor detector across different time periods (weekends/weekdays), we generated Figure 8. As illustrated in Figure 8, the q-values for three characteristic indices during both weekends and weekdays highlight the dominance of surrounding POIs, and the degree of influence does not differ significantly across different time periods. As shown in Figure 8 (a), during the morning peak period, the vitality of urban parks is significantly influenced by park type, park area, and 30-minute driving accessibility on weekends, while on weekdays, it is more strongly affected by 15-minute walking accessibility and surrounding work population density. As shown in Figure 8 (b), the evening activity index is significantly influenced by park facility density and 15-minute walking accessibility. Specifically, when comparing different time periods, they are more strongly affected by 15-minute walking accessibility on weekends, while on weekdays, they are more influenced by 15-minute driving accessibility and surrounding residential population density. As shown in Figure 8 (c), the day/night difference index is primarily influenced by distance from the city center, surrounding residential population density, park type, and 30-minute driving accessibility. When comparing different time periods, it is notably more affected by surrounding residential population density and 30-minute driving accessibility on weekends, while on weekdays, park type has a stronger influence. Notably, the distance from the city center exhibits a strong impact during both weekends and weekdays.

 

Comment 7: The "Discussion" explains the theoretical applicability of the method. The logical links with the "Results" are unclear. The text entails only selected elements of a classical dission.

Response 7: Thank you for pointing this out. We agree with this comment. Therefore, we have added a transitional sentence at the beginning of the Discussion section to strengthen its connection with the Results section, as shown in Lines 590–592: Based on the spatiotemporal differentiation results of the three vitality types and the Geodetector results, this study further explores their theoretical implications and planning insights.

Furthermore, we have designated the optimization strategies section as Section 4.1, "Land Use Planning and Sustainability Enhancement based on Vitality Types", in the Discussion section.

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

Compared with the previous version, the quality of the paper has been significantly improved. I have no other comments.

Author Response

Dear Reviewer,

We sincerely appreciate your positive feedback and kind acknowledgment of the improvements made to the manuscript. It is encouraging to hear that the revisions have significantly enhanced the paper’s quality, and we are grateful for your valuable insights throughout this process, which have been instrumental in strengthening our work.

 

Sincerely,

Ge Lou

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The quality of the manuscript has improved significantly. Could you please elaborate further on the following details (e.g., the AMAP platform and web crawler applications) for the sake of global readers?

"POI (point of interest) data were collected from AMAP (https://map.gaode.com/).... Hangzhou Gongshu District POI information was obtained via the AMAP API (https://lbs.amap.com/api/android-sdk/guide/map-data/poi/). This API enables web crawler programs to gather data, providing attributes such as IDs, names, categories, descriptions, coordinates, addresses, longitudes, and latitudes."

Although the dpi resolution of the figures has improved, several figures still appear somewhat small within the text, and I believe their size needs to be increased further.

Author Response

Comment 1: The quality of the manuscript has improved significantly. Could you please elaborate further on the following details (e.g., the AMAP platform and web crawler applications) for the sake of global readers?

"POI (point of interest) data were collected from AMAP (https://map.gaode.com/).... Hangzhou Gongshu District POI information was obtained via the AMAP API (https://lbs.amap.com/api/android-sdk/guide/map-data/poi/). This API enables web crawler programs to gather data, providing attributes such as IDs, names, categories, descriptions, coordinates, addresses, longitudes, and latitudes."

 

Response 1: We appreciate your suggestion to enhance the clarity for international readers.

We have added an explanation of the AMAP platform, API usage and data compliance in the manuscript:

 

1) The explanation of the AMAP platform

AMAP is one of China’s leading digital mapping platforms, operated by AutoNavi (a subsidiary of Alibaba Group). It provides comprehensive geographic information services, including point-of-interest (POI) data, routing, and real-time traffic updates. For global readers, we have added an explanation of its significance: "a leading Chinese geographic information platform comparable to Google Maps globally " (Lines 194-195)

 

2) API usage

We clarified that the data were obtained through AMAP’s official Open API (not traditional web crawling), which is designed for developers to access authorized geographic data. This approach complies with the platform’s terms of service and ensures data legitimacy. Specifically: The API endpoint used is the "POI Search API", which supports querying POI data by keywords, categories, or geographic boundaries. The process was implemented using Python scripting. We have revised our manuscript as follows:

Lines 196-199: Hangzhou Gongshu District POI information was obtained via the AMAP’s Web Service API (https://lbs.amap.com/api/webservice/guide/api-advanced/newpoisearch), which enables programmatic data retrieval.

Lines 200-202: Structured data retrieval was performed by defining geographic boundaries (Gongshu District, Hangzhou) and POI categories using Python scripting. Retrieved attributes include...

 

3) Data compliance

We added a note on data compliance: "Data were collected in accordance with the platform’s developer guidelines, and no personal privacy information was involved." (Lines 199-200)

 

The parts modified in the manuscript this round are highlighted in green.

 

Comment 2: Although the dpi resolution of the figures has improved, several figures still appear somewhat small within the text, and I believe their size needs to be increased further.

 

Response 2: Thank you for your careful review of the figures. We fully understand the importance of readability and have attempted to enlarge the figures as suggested. We have resized Figure 4 to the maximum size that meets the formatting requirements. Other figures in the manuscript have already been adjusted to the maximum extent that can be placed according to the requirements of the format template.

Furthermore, to address your concern without violating the journal’s requirements, we have added attachments to enhance clarity. For readers seeking higher detail, enlarged versions of Figures 6, 7 and 8 are provided as Supplementary Materials, accessible via the journal’s online platform.

Author Response File: Author Response.docx