Change of urban park usage as a response to the COVID-19 global … – Nature.com

The types of parks chosen for analysis

The total number of park visits determined from SafeGraph points of interest (POIs) located in NYC parks was 20,913,290 in 2019, but only 10,279,798 in 2020, representing a decrease of 49.2 percent.

There are 18 types of parks listed in the NYC Open Space Parks Data29. However, the top eight types of parks accounted for 91.35% of total park visits in 2019 and 92.17% in 2020, respectively, and are thus the focus of this study (Fig.1). These parks are classified by the NYC Department of Parks and Recreation29 as (1) Community Park, (2) Flagship Park, (3) Jointly Operated Playground, (4) Nature Area, (5) Neighborhood Park, (6) Playground, (7) Recreation Field/Courts, and (8) Triangle/Plaza. The detailed classification standard can be found in Table S-7.

Locations of parks in New York City. Each point represents an individual park. The number of parks and the median park area are shown for each park type.

We calculated four metrics associated with the number of park visits and the number of park visitors, namely (1) all visits: the total number of visits from all visitors; From the SafeGraph documentation, the duration of a visit must last at least 4min, and there could be multiple visits from a single visitor during the time period when the data were collected; (2) all visitors: the total number of unique visitors, regardless of their origin; (3) US visitors: the total number of unique visitors whose home locations are within the US; (4) NYC local visitors: the total number of unique visitors whose home locations are within NYC. Since temperature has been reported as a vital factor influencing park visitation30,31, we corrected the data for the effects of temperature, as described in the Data and methods section. The total numbers of these four types of visit/visitor counts by park type, and after the temperature correction, are summarized in Table S-1.

We examined the park visits and visitors change rate in each NYC borough by computing the total number of park visits or visitors in a month in that borough, then calculating the percentage change in 2020 visits (or visitors) compared to 2019 (i.e., (the visits in 2020the visits in 2019)/the visits in 2019). Manhattan was divided into lower Manhattan and upper Manhattan using 86th street as a divide. Results for total visits, total visitors, US visitors, and NYC visitors are shown in Fig.2.

Park visits and visitors change rate by borough, which is calculated as the percent change of total monthly park visits/visitors in 2020 compared to 2019. The letters on the right of each figure are the Tukey HSD multi-group comparison results, the same letters indicate the boroughs belong to the same group.

Starting from March 2020, the parks in all boroughs experienced decreased total visits (Fig.2a). April 2020 was the month with the greatest percentage decrease in visits compared to 2019, then visits slowly increased as the months progressed. Lower Manhattan had the greatest decrease in park visits from March to December (overall 61.1%, with a maximum of 86.6% in April 2020), while Staten Island experienced the smallest decrease (overall 20.3%, with a maximum of 57.5% in April 2020). All other boroughs experienced similar changes in visits, and shared a similar trend through time. All visitors (Fig.2b), U.S. visitors (Fig.2b), and NYC visitors (Fig.2d) had the same pattern with the greatest decrease in April, followed by a slow rebound with progressing time. Again, lower Manhattan had the largest decrease in unique visitors while Staten Island the smallest (Fig.2bd).

We examined the park visits and visitors change rate across the eight selected park types, by computing the total number of visits or unique visitors in a month to each park type and then calculating the percentage change of visits/visitors in 2020 compared to 2019 (Fig.3). There was a decrease in all types of visits and visitors to all eight park types across the city when comparing 2019 to 2020 for the months of March to June (Fig.3ad). For NYC local visitors, Triangle/Plazas (overall 62.9%, with a maximum of 82.9% in April 2020) and Flagship Parks (overall 57.0%, with a maximum of 78.7% in April 2020) had the largest decrease, followed by Jointly Operated Playground, Playground, Community Park, Neighborhood Park, then Recreation Field/Courts. Nature Areas had the smallest decrease in the number of NYC local visitors (overall 3.6%, with a maximum of 44.5% in April 2020) with some months even showing an increase. Beginning in June, the number of NYC local visitors to Nature Areas returned to about the same level as 2019 and even increased in some months (with a maximum increase of 29.0% in July 2020) (Fig.3d). The other three types of visits/visitors shared similar trends as NYC local visitors (Fig.3ac).

Park visits and visitors change rate by park type, which is calculated as the percent change of total monthly park visits/visitors in 2020 compared to 2019. The letters on the right of each figure are the Tukey HSD multi-group comparison results, any common letter shared by two park types indicates that the two park types were found to belong to the same group.

In order to better understand the needs and park usage of local urban residents, we focused our remaining analyses on data for NYC residents (also subsequently referred to as NYC local visitors) only. We also defined the visitor census block group (visitor CBG) as the home census block group where a visitor lived; and defined the park census block group (park CBG) as the census block group that a park was in, or was the closest to.

The CBGs (neighborhoods) that surround parks were divided into three income groups: lower, middle and upper, based on per capita income. The results for park visits change rate between 2019 and 2020 for each of the analyzed park types are provided in Fig.4.

NYC park visitors change rate by park type and by income level of park CBGs. The letters before the income groups are the Tukey HSD multi-group comparison results, any common letter shared by two income level groups indicates that the two groups were found to belong to the same group.

All eight park types saw decreased NYC local visitors regardless of the park CBG income level (Fig.4). Overall, parks in lower-income neighborhoods experienced statistically greater decreases in NYC local visitors than those in upper-income neighborhoods. No trend in visits change rate with income level was observed for Jointly Operated Playground, Neighborhood Park, Playground, and Triangle/Plaza. Community Parks and Nature Areas showed greater reductions in NYC local visitors in lower-income neighborhoods but showed no difference between middle- and upper-income neighborhoods. Flagship Parks showed greater reductions in NYC local visitors in lower- and middle-income neighborhoods. The outlier to the overall trend is Recreational Field/Courts, which showed greater reductions in NYC local visitors in upper-income neighborhoods than in lower-income neighborhoods.

The travel distance of visitors was used to examine how the travel behavior of NYC residents to parks changed during the early stages of the COVID-19 pandemic. In this section, the mean travel distances were computed for the time period from March to December in 2019 and 2020, as the major outbreak of the pandemic and the associated travel restrictions began in March 2020.

Overall, the mean travel distance of NYC residents to all parks reduced from 5.9km in 2019 to 5.1km in 2020 over March to December, representing a change of 13.2% (95% CI 13.4%, 13.1%). In 2020, there was a significant decrease in travel distance compared to 2019 for all study park types except for the Jointly Operated Playground, Playground, and Nature Area park types (Fig.5).

(a) Mean travel distance by park type. (b) Percentage change of mean travel distance by park type, with 95% CI error bars.

Before the pandemic, the mean travel distances to the Triangle/Plaza and Flagship Park types were the longest, both averaging 7.1kmover March to December in 2019; while the travel distances to the Playground and Jointly Operated Playground were the shortest, averaging 5.0km and 4.5km, respectively. The Nature Area, Jointly Operated Playground and Playground park types experienced a smaller decrease than average or even a slight increase, which were 1.2% (95% CI 2.0%, 0.3%), 1.9% (95% CI 2.5%, 1.3%) and 1.1% (95% CI 0.6%, 1.7%), respectively. All other types of parks experienced a greater reduction in travel distance (Table S-4).

The overall mean travel distances of NYC residents from lower-, middle- and upper-income level CBGs were 5.3km, 6.5km, and 6.0km, respectively, from March to December in 2019; and were 4.7km, 5.6km, and 5.0km, respectively, in the same period in 2020 (Fig.6a,b). In general, people from lower-income CBGs traveled a statistically shorter distance to parks than those from middle-income and upper-income CBGs in both 2019 and 2020. This pattern was common across all types of parks, except for Nature Areas and Triangle/Plazas, to which visitors from upper-income CBGs traveled the shortest distance.

(a,b) Mean travel distance by park type and by income level of visitor CBGs; The letters to the right of the mean travel distances are the Tukey HSD multi-group comparison results between income groups for each park type, any common letter shared by two income groups indicates that the two groups were found to belong to the same group. (c) Percentage change of mean travel distance by park type and by income level of visitor CBGs, with 95% CI error bars.

Overall, visitors from higher income CBGs had the greatest reduction in travel distance (Fig.6c). The percentage change of travel distance for visitors from lower-income, middle-income and upper-income CBGs are 10.7% (95% CI 11.0%, 10.4%), 13.9% (95% CI 14.2%, 13.7%), and 15.8% (95% CI 16.2%, 15.5%), respectively (Table S-5).

The specific changes varied by park type. For Community Park, Flagship Park, Jointly Operated Playground, Nature Area and Triangle/Plaza park types, the visitors from upper income level CBGs experienced the greatest percentage reduction in travel distance. While for Recreation Field/Courts, the visitors from upper income level CBGs had the smallest percentage reduction in travel distance.

The mean travel distances to parks located in lower-, middle- and upper-income level CBGs were 5.7km, 5.6km, and 6.3km, respectively, from March to December in 2019, and were 4.9km, 4.7km, and 5.6km, respectively, during the pandemic from March to December in 2020 (Fig.7a,b). In general, people tended to travel a statistically longer distance to parks in upper-income CBGs than to parks in middle-income and lower-income CBGs in both 2019 and 2020, with the exception of Community Park and Flagship Park in 2019.

(a,b) Mean travel distance by park type and by income level of park CBGs; The letters to the right of the mean travel distances are the Tukey HSD multi-group comparison results between income groups for each park type, any common letter shared by two income groups indicates that the two groups were found to belong to the same group. (c) Percentage change of mean travel distance by park type and by income level of park CBGs, with 95% CI error bars.

The travel distance to parks located in upper-income level CBGs had the smallest percentage decrease (Fig.7c), which was 10.6% (95% CI 10.9%, 10.3%), while it was 16.0% (95% CI 16.3%, 15.7%) for parks located in middle-income CBGs, and was 14.1% (95% CI 14.4%, 13.8%) for parks located in lower-income CBGs (Table S-6).

Examining by park type, for Community Park, Flagship Park, Recreational Field/Courts and Triangle/Plaza park types, the parks located in upper-income level CBGs experienced the smallest percentage reduction in travel distance. While for Nature Area, Jointly Operated Playground and Playground park types, the parks located in lower income level CBGs had an increase in travel distance.

Go here to read the rest:

Change of urban park usage as a response to the COVID-19 global ... - Nature.com