A method for assessing inequality in residential green space exposure

Abstract

The present invention belongs to the field of urban ecology and urban planning technology, specifically a method for assessing inequality in urban residents' exposure to green space. The method comprises: obtaining residential community data for the assessment area, POI data for urban green spaces and parks, and data on major urban parks; calculating accessibility based on the number of POIs for urban green spaces and parks within a specific buffer range, and normalizing the data; calculating accessibility based on the shortest path Di from residential areas to all major urban parks, and performing inverse data normalization; averaging the accessibility and accessibility data; plotting a Lorenz curve; and calculating the Gini coefficient, which is the Green Space Exposure Inequality Index (GEII). The method is service-oriented, user-friendly, and extensible, and it dispenses with complex evaluation indicators, making it operational. It is expected to be included in urban planning guidelines, pushing research on green space exposure inequality into practical urban planning. The method is of great value in alleviating inequality in urban residents' exposure to green space and scientifically optimizing landscape planning.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban ecology and urban planning, and in particular relates to a method for evaluating the inequality of green space exposure among urban residents. Background Art

[0002] Urban green spaces significantly impact the physical and mental health of residents by providing various ecosystem services. The provision and maintenance of urban natural environments is an integral part of healthy cities. Therefore, it is necessary to continuously create and improve the quantity and quality of urban natural environments through urban interventions (such as urban planning and design) to mitigate the inequality in residents' green space exposure caused by differences in urban pattern. Currently, a growing number of studies are dedicated to improving systems for assessing residents' green space exposure. However, the development of multidimensional and comprehensive evaluation indicators has made these systems increasingly complex, and there is currently little conceptual framework for assessing "inequality in green space exposure." To mitigate this inequality in residents' green space exposure through urban planning, it is necessary to develop a simple, operational, and people-oriented evaluation indicator that serves practical planning needs to serve as a theoretical basis for urban planning guidelines. Therefore, the present invention provides a simple, yet effective, and universal method for assessing inequality in residents' green space exposure. This method has three major advantages: serving practical planning as its purpose, mapping human potential green space exposure as its core, and scientific expansion and improvement as its development direction. It abandons the complex assessment process of multiple indicators, can overcome the shortcomings of existing assessment systems, and better promote the study of exposure inequality from theoretical research to practical urban planning guidance. It has important theoretical and practical significance for guiding scientific and humanistic urban planning and reducing differences in population green space exposure. The Green Space Exposure Inequality Index (GEII) includes an availability quantity index, an accessibility distance index, and a Gini coefficient. The Gini coefficient is one of the commonly used indicators used internationally to measure the income gap between residents in a country or region. The Lorenz curve and the Gini coefficient in the field of economics are widely used to assess the inequality of urban residents' green space exposure. In the present invention, the Lorenz curve is the cumulative distribution function of residents' green space exposure, and the Gini coefficient is the ratio of the area between the absolute fairness line and the Lorenz curve (actual exposure distribution curve) to the total area below the absolute fairness line, representing the overall degree of inequality in green space exposure. The maximum Gini coefficient is "1" and the minimum is "0". The closer the Gini coefficient is to 0, the more equal the distribution is. Summary of the Invention

[0003] The purpose of this invention is to provide a simple but effective and universal method for assessing the inequality of exposure to green space for residents, so as to overcome the shortcomings of the existing assessment system in serving practical planning and better promote the study of exposure inequality from theoretical research to practical urban planning guidance.

[0004] The method for assessing inequality in residents' green space exposure provided by the present invention comprises the following steps:

[0005] (1) Obtaining POI data of residential communities, and screening and cleaning them;

[0006] Specifically, we can obtain POI (point of interest) data of residential communities in the study area from the AutoNavi open platform; filter and clean the data entered multiple times in a community;

[0007] (2) Obtain POI data of urban green spaces and parks, and perform screening and cleaning;

[0008] Specifically, you can obtain POI data of all urban green spaces and parks from the AutoNavi open platform; filter and clean out data that has been entered multiple times for a green space or park;

[0009] (3) Set the buffer range of green space exposure availability; calculate the number of POIs of urban green spaces and parks in each residential area within the buffer zone (called availability data); normalize the availability data to obtain the normalized availability value y1;

[0010] (4) Obtain data on major urban parks with recreational facilities and convenience facilities; combine with Amap to determine the geographical coordinates of major urban parks; filter and clean outliers to solve the problem of being unable to obtain geographical coordinates;

[0011] (5) Calculate the shortest path D from residential areas to all major city parks i ; Shortest path D for reachability data i Normalization is performed to obtain the reachability data D i Normalized value y2;

[0012] (6) Calculate the green space exposure inequality index GEII; perform a 1:1 weighted average of the availability and accessibility data to obtain y3; draw the Lorenz curve; calculate the Gini coefficient, that is, the green space exposure inequality index GEII.

[0013] Further:

[0014] The step (1) of obtaining the POI data of the residential communities in the study area from the AutoNavi open platform is to register an account on the AutoNavi open platform (https: / / lbs.amap.com) and apply for the key of the "web service API", splice the HTTP request URL according to the operation instructions, receive the data returned by the HTTP request, and parse the data; the screening and cleaning of the data entered multiple times in a community is achieved by using Python to apply the following rules: ① Eliminate redundant items with the same first four characters; ② Eliminate redundant items with the same first special character.

[0015] In step (2), the POI data of all urban green spaces and parks in the study area are obtained from the AutoNavi open platform by calling the data interface on the AutoNavi open platform (https: / / lbs.amap.com). The specific process is the same as that described in step (1). The data of a green space or park that has been entered multiple times is screened and cleaned. First, the data error problem is manually corrected and the data is cleaned according to the following two logics: ① Eliminate redundant items with the same first four characters; ② Eliminate redundant items with the same first special character. At the same time, manually screen and eliminate points that are not the target but are classified as green spaces or parks.

[0016] In step (3), the availability buffer range is set. Availability is assessed by creating a circular buffer with a preset radius around the geometric centroid of each residential area and calculating the number of urban green space and park POIs contained in the buffer. The preset buffer radius can be 500m or 1000m, etc., depending on the size of the assessment area and the density of residential areas and green spaces.

[0017] The calculation of the number of POIs of urban green spaces and parks within the buffer zone for each residential area is performed by using the Generate Neighbor Table in ArcGIS (Generate Neighbor Table (Analysis) - ArcMap | Documentation (arcgis.com)) to obtain a starting table of neighborhoods and parks within a specific buffer zone, and then using the summary tool to summarize the number of POI points.

[0018] The normalization of the availability data is to process the availability data according to the following formula:

[0019] y1=(xx min ) / (x max -x min ), (1)

[0020] Mapped to (0,1), where x refers to the number of urban green spaces and park POIs in each community within a specific buffer zone, x max Refers to the maximum number of urban green spaces and park POIs in a specific buffer zone for all communities, x min refers to the minimum value, and y1 represents the normalized availability value.

[0021] In step (4), the data of major urban parks with entertainment facilities and convenience facilities are obtained, specifically, the data of major urban parks with entertainment facilities and convenience facilities are obtained in combination with the latest urban park directory, such as comprehensive parks, community parks, sports parks, wetland parks, waterfront parks, heritage parks, forest parks, etc.

[0022] The determination of the geographical location coordinates corresponding to the major city parks in combination with Amap is based on a final list of major city parks, combined with Amap and a comprehensive analysis of road network data from OSM (Open Street Map) to obtain the geographical location coordinates corresponding to the city parks.

[0023] The screening and cleaning of outliers solves the problem of being unable to obtain geographic coordinates. Specifically, some city parks, such as "Shanghai Guyi Garden" and "Guqi Garden", have inconsistent names, resulting in the inability to obtain geographic coordinates. The outliers need to be verified and corrected one by one.

[0024] In step (5), the shortest path D from the residential area to all major city parks is calculated. i , is to use the formula:

[0025] D i =min j d ij , (2)

[0026] Select the shortest path from the community to all city parks, where D i represents the shortest distance from residential area i to the main city park, d ij It represents the distance between the main urban park j and the residential area i. The shortest path analysis was implemented in ArcGIS, and the key tool used was OD cost analysis (OD cost matrix analysis—ArcMap | Documentation (arcgis.com)).

[0027] For points on roads, you can first break the line shp at the intersection: Convert features to lines. Then, create a network dataset in ArcCatalog. Load the network dataset with the shp into the table of contents. In Network Analyst, click Create New OD Cost Matrix and load the shps for the origin and destination points. Then, import the shps for the origin and destination points into the Network Analyst window. After solving, the shortest path can be found in the line attribute table of the OD matrix.

[0028] For the case where the point is not on the road, use the Generate Neighbor Table and XY Line Conversion functions to generate the connection between the community and the nearest road, and then follow the general process to finally obtain the result in the line attribute table.

[0029] The shortest path D for the reachability data i Normalization is done using the formula:

[0030]

[0031] To D i Normalization is performed, where x represents D i, y2 represents the accessibility data D i The normalized value is in the range of (0,1).

[0032] In step (6), the availability and accessibility data are weighted averaged in a ratio of 1:1, which is obtained by taking a weighted average of the availability y1 and accessibility y2 corresponding to each community in a ratio of 1:1:

[0033]

[0034] The Lorenz curve is drawn by performing data calculation in Excel and using a two-dimensional line graph to draw the Lorenz curve, where the horizontal axis X represents the cumulative percentage of the number of residential communities, and the vertical axis Y represents the cumulative percentage of the resident green space exposure y3;

[0035] The calculation of the Gini coefficient, namely the Green Space Exposure Inequality Index GEII, is to first calculate the area under the Lorenz curve:

[0036]

[0037] Among them, n is the total number of POIs in residential areas, X i represents the cumulative percentage corresponding to the i-th residential area, Y i represents the cumulative percentage corresponding to the green space exposure of residents in the i-th community;

[0038] Further calculate the Gini coefficient:

[0039] G=1-2*S B , (6)

[0040] This is the value of the Green Space Exposure Inequality Index GEII. Figure 1 The concept of Gini coefficient is explained in the formula It is equal to formula (6). In actual calculation, it is more convenient to calculate the area of ​​part B first and then calculate the Gini coefficient. Figure 1 The calculation formula in is mainly to explain the meaning of the Gini coefficient.

[0041] Beneficial effects of the present invention

[0042] The present invention provides a simple but effective and universal method for assessing residents' green space exposure inequality. This method has three major advantages: serving practical planning as its purpose, mapping human potential green space exposure as its core, and scientific expansion and improvement as its development direction. It abandons the complex evaluation procedures of numerous indicators, overcomes the shortcomings of existing evaluation systems in serving practical planning, and better promotes the study of exposure inequality from theoretical research to practical urban planning guidance. It has important theoretical and practical significance for guiding scientific and humane urban planning and reducing population green space exposure differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a conceptual diagram of the Greenland Exposure Inequality Index.

[0044] Figure 2 A technical flow chart for operating the indicators of the present invention.

[0045] Figure 3 Calculate the shortest path diagram from the community to the city park for ArcGIS.

[0046] Figure 4 This is the spatial distribution of POIs in residential areas and communities in Shanghai in 2012, 2015, 2018 and 2021.

[0047] Figure 5 This is the spatial distribution of POIs in all urban green spaces and parks in Shanghai in 2012, 2015, 2018 and 2021.

[0048] Figure 6 This is the number of POIs in green spaces and parks within a 1000m buffer zone in communities in Shanghai in 2012, 2015, 2018, and 2021.

[0049] Figure 7 This is the normalized spatial distribution of residential green space accessibility in Shanghai in 2012, 2015, 2018 and 2021.

[0050] Figure 8 These are the Lorenz curves and inequality index GEII of community green space exposure in Shanghai in 2012, 2015, 2018 and 2021. DETAILED DESCRIPTION

[0051] The method of the present invention is further described below with reference to the accompanying drawings and examples.

[0052] Taking Shanghai as an example, the green space exposure inequality index GEII of the present invention is used to evaluate the inequality of residents' green space exposure in Shanghai in 2012, 2015, 2018 and 2021.

[0053] Obtain POI data of residential areas and communities in Shanghai from 2012 to 2021 on the AutoNavi open platform ( Figure 4 ), use Python to filter and clean out the multiple entries for a community. Get POI data of all urban green spaces and parks in Shanghai from the AutoNavi open platform ( Figure 5 ), filter and clean out data that are entered multiple times for a green space or park, as well as data that are misclassified but do not belong to the target, set the buffer radius to 1000 meters, and calculate the number of POIs of urban green spaces and parks in each residential area within the buffer zone ( Figure 6) as the availability value and use the formula:

[0054] y1=(xx min ) / (x max -x min ),

[0055] The number of POIs is normalized and the availability value is mapped to (0, 1).

[0056] According to the 2021 Shanghai City Park List and the annual list adjustment notices of the Shanghai Greening and Urban Appearance Administration, data on major parks in Shanghai from 2012 to 2021 were obtained. The corresponding geographic coordinates were obtained by comprehensive analysis based on Amap and road network data from OSM. Outliers were filtered and cleaned, and the formula D was used to calculate the park's location. i =min j d ij Select the shortest path from the community to all city parks and use the formula:

[0057]

[0058] Normalize it ( Figure 7 ).

[0059] Take the 1:1 weighted average of availability y1 and accessibility y2 to get y3;

[0060] The Lorenz curve was drawn in Excel, where the horizontal axis X represents the cumulative percentage of the number of residential communities and the vertical axis Y represents the cumulative percentage of the residents’ green space exposure y3.

[0061] Compute the area under the Lorenz curve:

[0062]

[0063] Calculate the Gini coefficient G = 1-2*S B , which is the value of the green space exposure inequality index GEII.

[0064] The Lorenz curve and inequality index GEII of community green space exposure in Shanghai in 2012, 2015, 2018 and 2021 are as follows: Figure 8 As shown, this case also verifies the feasibility of the present invention.

Claims

1. A method for assessing inequality in residents’ exposure to green space, characterized by: The specific steps are as follows: (1) Obtaining POI data of residential communities, and screening and cleaning them; Specifically, we obtain POI data of residential communities in the study area from the AutoNavi open platform; filter and clean out data that has been entered multiple times in a community; (2) Obtain POI data of urban green spaces and parks, and perform screening and cleaning; Specifically, we obtain POI data of all urban green spaces and parks from the AutoNavi open platform; filter and clean out data that has been entered multiple times for a green space or park; (3) Set the buffer range of green space exposure availability; calculate the number of POIs of urban green spaces and parks in each residential area within the buffer zone, which is called availability data; normalize the availability data to obtain the normalized availability value y1; (4) Obtain data on major urban parks with recreational facilities and convenience facilities; combine with Amap to determine the geographical coordinates of major urban parks; filter and clean outliers to solve the problem of being unable to obtain geographical coordinates; (5) Calculate the shortest path D from residential areas to all major city parks i ; Shortest path D for reachability data i Perform normalization to obtain the reachability data D i Normalized value y2; (6) Calculate the green space exposure inequality index GEII; perform a 1:1 weighted average of the availability and accessibility data to obtain y3; draw the Lorenz curve; calculate the Gini coefficient, that is, the green space exposure inequality index GEII; In step (5); The shortest path D from the residential area to all major city parks is calculated. i , is to use the formula: D i =min j the ij , (2) Select the shortest path from the community to all city parks, where D i represents the shortest distance from residential area i to the main city park, d ij It represents the distance between the main urban park j and the residential area i. The shortest path analysis is implemented in ArcGIS using the OD cost analysis tool. For points on roads, first break the line shp at the intersection: Convert features to lines; then create a network dataset in ArcCatalog; then load the network dataset shp into the table of contents, click New OD Cost Matrix in Network Analyst, and load the shp of the start and destination points. Then, import the shp of the start and destination points into the Network Analyst window; after solving, obtain the shortest path in the line attribute table of the OD matrix; For the case where the point is not on the road, use the Generate Neighbor Table and XY Line Function to generate the line connecting the cell and the nearest road, and then follow the general process to finally obtain the result in the line attribute table; The shortest path D for the reachability data i Normalization is done using the formula: To D i Normalization is performed, where x represents D i , y2 represents the accessibility data D i The normalized value range is (0,1); In step (6): The availability and accessibility data are weighted averaged at a ratio of 1:1, which is obtained by taking a weighted average of the availability y1 and accessibility y2 corresponding to each community at a ratio of 1:1: The Lorenz curve is drawn by performing data calculation in Excel and using a two-dimensional line graph to draw the Lorenz curve, where the horizontal axis X represents the cumulative percentage of the number of residential communities, and the vertical axis Y represents the cumulative percentage of the resident green space exposure y3; The calculation of the Gini coefficient, namely the Green Space Exposure Inequality Index GEII, is specifically to calculate the area under the Lorenz curve: Calculate the Gini coefficient: G=1-2*S B , (6) This is the value of the green space exposure inequality index GEII.

2. The method for assessing inequality in residents’ green space exposure according to claim 1, characterized in that: In step (1): To obtain the POI data of residential communities in the study area from the AutoNavi open platform, you need to register an account on the AutoNavi open platform and apply for a "web service API" key, follow the instructions to assemble the HTTP request URL, receive the data returned by the HTTP request, and parse the data; The screening and cleaning of data entered multiple times by a community is performed by using Python using the following rules: ① Eliminate redundant items with the same first four characters; ② Eliminate redundant items that are consistent before the first special character.

3. The method for assessing inequality in residents’ green space exposure according to claim 2, characterized in that: In step (2): The POI data of all urban green spaces and parks in the study area are obtained from the AutoNavi open platform by calling the data interface on the AutoNavi open platform. The specific process is the same as that described in step (1). The method of screening and cleaning the data that has been entered multiple times for a green space or park first manually corrects the data row errors and performs data cleaning according to the following two logics: ① Eliminate redundant items with the same first four characters; ② Eliminate redundant items with the same first four characters; and at the same time, manually screen out points that do not belong to the target but are classified as green spaces or parks.

4. The method for assessing inequality in residents’ green space exposure according to claim 1, characterized in that: In step (3): The accessibility buffer range is set by creating a circular buffer with a preset radius around the geometric centroid of each residential area and calculating the number of urban green space and park POIs contained in the buffer to evaluate accessibility data. The preset buffer radius is 500m-1000m, which can be determined based on the scale of the assessment area and the density of residential areas and green spaces. The number of POIs of urban green spaces and parks in each residential area within the buffer zone is calculated by generating a neighbor table in ArcGIS to obtain a starting table of residential areas and parks within a specific buffer zone, and using a summary tool to summarize the number of POI points; The normalization of the availability data is to process the availability data according to the following formula: y1=(xx min ) / (x max -x min ), (1) Mapped to (0,1), where x refers to the number of urban green spaces and park POIs in each community within a specific buffer zone, x max Refers to the maximum number of urban green spaces and park POIs in a specific buffer zone for all communities, x min Refers to the minimum value, and y1 represents the normalized availability value, which ranges from (0,1).

5. The method for assessing inequality in residents' green space exposure according to claim 4, characterized in that: In step (4); The acquisition of data on major urban parks with recreational facilities and convenience facilities is specifically to acquire data on major urban parks with recreational facilities and convenience facilities in combination with the latest urban park directory; Determining the geographical coordinates of major city parks by combining with Amap is based on the final list of major city parks, combined with Amap and comprehensive analysis of road network data from OSM to obtain the geographical coordinates of the city parks; The screening and cleaning of outliers solves the problem of being unable to obtain geographic coordinates. Specifically, the names of some city parks are inconsistent, resulting in the inability to obtain geographic coordinates, and the outliers need to be verified and corrected one by one.

Images