key: cord-0994124-nh1gondh
authors: Lopez, Bianca; Kennedy, Christopher; Field, Christopher; Mcphearson, Timon
title: Who benefits from urban green spaces during times of crisis? Perception and use of urban green spaces in New York City during the COVID-19 pandemic
date: 2021-09-22
journal: Urban For Urban Green
DOI: 10.1016/j.ufug.2021.127354
sha: 51e236d4cc5d59bd0ae571ba645ab595de28448d
doc_id: 994124
cord_uid: nh1gondh

Urban green spaces (UGS) provide health benefits to city dwellers, which may be even more important during times of crisis such as the COVID-19 pandemic. However, lack of access to UGS or important UGS features, in addition to concerns about UGS safety or maintenance, could prevent people from receiving these benefits. We designed an online survey to understand how people were using and perceiving UGS during the COVID-19 pandemic in New York City during the spring of 2020. The survey included questions about how people’s visits to UGS and perceptions of the importance of UGS for health had changed since the start of the pandemic, as well as the concerns people had and features of UGS they considered important. Of the 1372 people who took the survey, most respondents were concerned about a lack of social distancing and crowded UGS, and respondents with these concerns were less likely to visit UGS and had visited UGS less often during than before the pandemic. In addition, generalized linear models showed differences in some concerns and important features of UGS across gender, race and ethnicity, demonstrating the importance of considering community needs in UGS design and management. Although concerns about lack of access were not common in our study population, these also appeared to prevent people from using UGS, and were more common in certain areas of the city that were also hard-hit by COVID-19 in the beginning of the pandemic. To ensure that people can get health benefits from UGS during times of crisis, cities must eliminate barriers by providing equitable access to UGS, considering what amenities communities need from UGS, and provide consistent communication about public health policies.

Urban green spaces (UGS), such as parks, gardens, and plazas, provide a number of potential benefits to city dwellers. Exposure to and use of UGS can improve physical and mental health through recreation and exercise (Kaczynski and Henderson, 2007) and decreased anxiety and stress (Cox et al., 2017; Hazer et al., 2018; Li et al., 2018; Tyrväinen et al., 2014) . These physical and mental health benefits may become even more critical during times of crisis (van den Berg et al., 2010) such as the current novel coronavirus (SARS-CoV-2) pandemic (McCunn, 2020; Samuelsson et al., 2020) , as residents face heightened health-related and economic stress, grief, isolation, and limited mobility during the implementation of social distancing policies J o u r n a l P r e -p r o o f (Brooks et al., 2020) . However, the use of UGS during the current pandemic may be limited by individuals' lack of access to UGS, in addition to temporary park and facilities closures and concerns about safety or overcrowding. This presents a unique challenge to city governments to decide how to safely and equitably manage UGS during a public health crisis.

Communities' use of UGS depends on a number of factors, including who has access to UGS, what amenities are provided, how well they match the needs of the community, and who feels safe and welcome to use them (Bedimo-Rung et al., 2005; Lee and Maheswaran, 2011; Rupprecht et al., 2015) . Access to UGS is not distributed equally in many cities, with less green space often found in low-income communities of color, qualifying it as an environmental justice issue rooted in a history of structural racism and disinvestment in UGS (Rigolon, 2016) . The same communities that have less UGS access also tend to have lower park "quality", including funding for parks, maintenance and cleanliness, and the amenities (e.g. recreational facilities) that are provided (Conedera et al., 2015) . Several recent studies have shown that perceptions of park quality, in addition to or even more than quantity of UGS, predict park usage and the health benefits that individuals can receive from parks (Akpinar, 2016; Brindley et al., 2019; Flowers et al., 2016) . Cultural and individual differences (e.g. affinity towards nature) can lead to different perceptions of and preferences for UGS among urban populations (Fischer et al., 2018; Flowers et al., 2016; Gobster, 2002; Madureira et al., 2015; Rupprecht et al., 2015) . Recognizing these differences is critical for providing the benefits of UGS to diverse residents (Frumkin et al., 2017; Hartig et al., 2014) .

The COVID-19 (the disease caused by the SAR-CoV-2 virus) pandemic has changed people's perceptions and use of UGS (Honey-Rosés et al., 2020; McCunn, 2020) . Many cities, regions, and entire countries enacted temporary lockdowns (in addition to social distancing J o u r n a l P r e -p r o o f recommendations and mandates) to control the spread of the virus, leading to unprecedented reductions in mobility (Rutz et al., 2020) and temporary closure of public spaces (Slater et al., 2020) . Some studies have found increases in park and natural area use during the pandemic, as residents seek relief from rising stress and anxiety and try to limit indoor activities with higher virus transmission risk (Grima et al., 2020; Kleinschroth and Kowarik, 2020; Lu et al., 2020; Venter et al., 2020) . However, limited physical or perceived access to UGS, due to either lack of nearby greenspace, closed facilities, or perceptions of not being welcome in UGS, could prevent people from using UGS during this crisis. In addition, concerns about the safety of visiting public spaces could reduce people's ability or willingness to use UGS during the pandemic (Slater et al., 2020) , limiting people's access to potential benefits from visiting UGS.

To better understand these evolving concerns and issues, we designed an online survey to assess how people were using and perceiving UGS during the early stages of the COVID-19 pandemic (May 13 to June 15, 2020). Our study focuses on New York City (New York, NY, hereafter "NYC"), the largest U.S. city and one of the hardest-hit regions in the U.S. in the spring of 2020 (Centers for Disease Control and Prevention, 2020a). The survey included questions about changes in people's visits to UGS and perceptions of the importance of UGS for health since the start of the pandemic, as well as the concerns people had and features of UGS they considered important. We examined how responses to these questions varied across locations (city boroughs) and respondent demographics (gender and race/ethnicity) to understand how different groups of people were using and perceiving UGS. The results of this study have implications for funding, policy, and management of UGS in NYC, with potential applications to other cities, particularly during times of crisis.

J o u r n a l P r e -p r o o f

NYC is the largest city in the United States with a population of approximately 8.4 million people (US Census Bureau, 2019). The city comprises five boroughs: Manhattan, Brooklyn, Queens, the Bronx, and Staten Island. NYC has over 1,700 parks (NYC Parks 2020; https://www.nycgovparks.org/about/) and is considered one of the top large U.S. cities for providing walkable park access to its citizens (The Trust for Public Land 2020; https://www.tpl.org/city/new-york-new-york). NYC also has over 12,000 acres of natural areas, including wildlife refuges and eight public beaches (NYC Parks 2020; https://www.nycgovparks.org/about/), and approximately 12,600 acres of public or "passive" open space (1.5 acres per 1,000 residents) such as plazas and esplanades within the city limits (Harnik et al., 2016) .

In response to COVID-19 outbreaks in NYC and nearby New Rochelle, NY, New York state authorities created a shelter-in-place mandate ("PAUSE") in addition to social distancing policies that included restricted access or temporary closure of public parks, beaches, and other natural areas, beginning in March of 2020 (Angel et al. 2020 ; New York State Department of Health and Mental Hygiene 2020). Our survey was active during a time period following some of the worst health impacts of the crisis, but before NYC reopened many non-essential businesses (e.g. indoor dining).

We developed an online survey using Qualtrix to understand how people were using UGS during the early stages of the COVID-19 pandemic, whether people perceived UGS to be J o u r n a l P r e -p r o o f important for their health, and whether exposure to UGS was affecting mental health during this time. There were two versions of this survey: one which was sent out nationally and one that was specific to residents currently living in NYC. The NYC survey (analyzed in this study) was designed in partnership with the New York State Health Foundation, The Building Healthy Communities NYC program, and the Nature Conservancy of New York, who helped tailor questions and relevant language, including an emphasis on open spaces provided by the NYC Housing Authority (NYCHA). In the survey we used the terminology "parks and open spaces" (referred to here as UGS), which we defined as any public spaces with natural or managed vegetation, including parks, greenways, public gardens, and accessible wetlands, forest, prairies, and beaches.

The survey questions analyzed in this study came from three of the six sections of the survey (see Appendix A). The first set of questions (from Section 2) asked about participants' use of UGS during the pandemic, including 1) how often participants had visited UGS in the last week, 2) how their participation in visiting UGS (in addition to other activities) had changed since the start of the pandemic, 3) which features were considered important for a UGS experience, 4) whether the UGS they visited most often provided each of the features they identified as important, and 5) what concerns, if any, the participant had with visiting UGS.

Questions from Section 5 asked how important participants considered UGS to be for their mental and physical health, currently and before the COVID-19 crisis, respectively. Finally, in Section 6 we asked a series of demographic questions, including participants' zip code, NYC borough and neighborhood, gender, age, race and ethnicity, education level attained, and household income. All questions were multiple choice, with the option to select multiple responses for questions identifying concerns and important features of UGS and racial identity.

There was also an option to provide additional comments at the end of the survey. See Appendix A for exact wording of questions used in this study.

We recruited survey participants using a combination of a convenience sample and snowball approach. We advertised the survey on social media (Twitter and Instagram) and distributed a link to the survey over email to listservs and networks, including those of our partner organizations (The New York State Health Foundation and The Nature Conservancy of New York). The survey was open to anyone over the age of 18 who resided in NYC.

We analyzed data from survey responses that answered at least 70% of the survey questions to capture responses to the questions analyzed in this study (Appendix B Table 1 ). We compared the demographics of survey participants to data from the U.S. Census American Community Survey 5-year estimates 2015-2019 (U.S. Census Bureau, 2020) using Fisher's tests. We examined the distributions of responses to identify the most commonly selected concerns about and important features of UGS. We also coded write-in "other" responses about concerns and important features into themes using an emergent, inductive coding approach (Ryan and Bernard, 2003) . To determine whether respondents perceived that they had access to their desired UGS features, we calculated the proportion of features that each respondent selected as important that they also said their most-visited UGS provides.

We were interested in how people's concerns with visiting UGS, important features for UGS to provide, access to those important features, perceptions of UGS's importance for their J o u r n a l P r e -p r o o f health, and frequency of UGS visits differed across populations. We therefore used generalized linear models to examine differences in responses to these questions across locations and demographics, focusing on NYC borough (Manhattan, Bronx, Queens, Brooklyn, and Staten Island), gender (female, male, and nonbinary), and non-exclusive, census-based race and ethnicity categories (Asian, Black or African American, Hispanic/Latinx, Native American [including American Indian or Alaska Native and Native Hawaiian or Pacific Islander], and white), all included as fixed effects. We were also interested in how perceptions of UGS (concerns, how well UGS provided important features, and perceived importance of UGS for health) affected how often people visited UGS and how this frequency changed during the pandemic. We thus included the following in models predicting UGS visit frequency and change in frequency: dummy variables coding whether respondents had selected particular concerns (lack of social distancing, crowded UGS, lack of access, and lack of safety); the proportion of important features provided by the most visited UGS; and an indicator of perceived importance of UGS, calculated as the mean of the numeric values (1-5 on a Likert scale from "Not at all important" to "Extremely important") of responses to the four questions about the importance of UGS for health (Appendix B Table 1 ).

All data analysis was performed in R (R Core Team, 2020). We used the "polr" function in the "MASS" package (Venables and Ripley, 2002) to fit ordinal regressions for responses with ordered categories (importance of UGS for health, frequency and change in frequency of UGS visits) and the "glm" function using the "family = binomial" argument to fit logit regressions for binary responses (selected concerns and important features) and the proportion of features provided by the most visited UGS. Log ratios that did not overlap one (with confidence intervals calculated using the "confint" function) were considered significant. We tested for J o u r n a l P r e -p r o o f multicollinearity in models using the variance inflation factor (VIF) with the "vif" function in the "car" package (Fox and Weisberg, 2019) . We found no evidence of collinearity in models, with all VIFs < 2.5. Some binomial models showed evidence of overdispersion, so we reran them using a "quasibinomial" family. Results are shown for binomial models only, since results did not differ substantially between any binomial and quasibinomial models. Figures were created using the "ggplot2" (Wickham, 2016) and "ggalluvial" (Brunson, 2017) packages in R.

In total, 1372 people took the NYC survey, and 1145 people completed over 70% of the survey questions. Of these, over 70% identified as female and 23% as male, with ~2% (19 individuals) identifying as nonbinary; females were significantly overrepresented in the dataset compared to estimates from the U.S. Census (Fisher's test simulated p-value < 0.001; Appendix C Figure 1 ).

Over 75% of survey respondents identified as white (72% as white only), 8% as Black, 7% as Asian, and 1% (11 individuals) as Native American. About 10% of respondents identified as Latinx. This demographic makeup differs significantly from NYC residents (Fisher's test simulated p-value < 0.001; Appendix B Figure 2 ), of whom about 45% identify as white, 26%

identify as Black, 15% as Asian, 1% as American Indian or Pacific Islander, and 29% as Latinx (U.S. Census Bureau, 2020).

Respondents were also biased towards young age (45% in their twenties and thirties but only 7% over 65 years old; Appendix C Figure Most responses came from Brooklyn (44%) and Manhattan (32%), with fewer from Queens (12%), the Bronx (6%) and Staten Island (1%). Over 85% of respondents stated that they use UGS.

The most common concerns about visiting UGS selected by respondents were that "People are not practicing social distancing" (59%) and that UGS were "Too crowded" (57%) (Figure 1 ).

The most common write-in concerns (all <2% of respondents) were that people were not wearing masks, that public restrooms were lacking or closed, and that there was no available UGS close by or that nearby UGS were closed during the shelter-in-place order.

There were some differences in concerns across gender, race/ethnicity, and borough ( Figure 1 ). Men responded that UGS did not meet their needs more often than women, and nonbinary respondents were more concerned than women about police presence and lack of adequate lighting in UGS. Latinx respondents were more likely than others to select a lack of park staff and "Not child-friendly" as concerns. Access to UGS was a greater concern in Brooklyn and Queens than in Manhattan, and concerns about crowded UGS and a lack of park staff also differed across boroughs (Appendix D Table 1 ).

J o u r n a l P r e -p r o o f "Places to walk / trails" and "Trees, shade" were selected as important features of UGS visits by over 85% of respondents, and over half of respondents selected "Places to sit", "Landscaping / maintained gardens, and/or "Water feature (e.g. river, lake, pond, ocean, fountain, sprinkler, pool)" (Figure 1 ). The most common write-in features (all <2% of respondents) were public restrooms, nature or wildlife (e.g. "Birds and squirrels", "Natural environment to observe cycles of nature"), bicycle paths, and public art.

The features of UGS that people considered important varied across gender, with men less frequently selecting trails, water features, places for children to play, or "Educational opportunities (e.g. informational signage, guided trails)" while nonbinary respondents more frequently selected "Places to BBQ, cook food" and "Dog-friendly" than women (Figure 1 ).

Latinx respondents more frequently selected "Places to BBQ, cook food" and "Places to sit", and Black respondents were less likely to select "Opportunities to view wildlife". There were differences across boroughs in the number of respondents who selected "Landscaping / maintained gardens'', "Trees, shade", "Places to sit", and "Places to exercise / play sports" (Figure 1 ; Appendix D Table 2 ).

On average, respondents stated that most of the features they deemed important for a UGS visit experience were provided by their most frequently visited UGS (mean 83±24%, median 100%). However, respondents from Queens reported a significantly lower proportion of the features they consider important that were provided (Appendix D Table 2 ).

Over eighty percent of respondents reported that UGS were either "extremely important" or "very important" for mental and physical health, respectively, but overall, UGS were considered J o u r n a l P r e -p r o o f more important for mental than physical health (Figure 2 ). Men ranked UGS as less important for both mental and physical health than did women and respondents from Queens ranked UGS as less important for mental and physical health than those from Manhattan. Black respondents ranked UGS as less important for mental and physical health than others and Asian respondents ranked UGS as less important than others for physical health only (Figure 2 ).

On average, respondents reported that they consider UGS to be slightly more important for their mental and physical health now than they did before the start of the COVID-19 crisis.

Although more than half of respondents reported no change in their assessment of UGS importance for mental or physical health (most of whom selected "extremely important" for both time periods), 39% ranked UGS as more important for mental health now than before the pandemic and 36% ranked UGS as more important for physical health now than before ( Figure   2 ).

Approximately equal numbers of respondents reported that they increased (15%) and decreased (14%) their visitation of UGS during the pandemic, with fewer reporting that they started or stopped visiting UGS (Figure 3 ; Appendix D Table 3 ). Asian and white respondents were more likely than others to increase their visitation of UGS during the pandemic. Respondents were less likely to increase their UGS visitation during the pandemic if they reported concerns about social distancing, access to UGS, or the safety of UGS (but not crowded UGS; Figure 3 ).

Over 20% of survey respondents reported that they had visited a UGS 3-4 times during the preceding week and another 20% had visited about once per day (Figure 3 ). Nonbinary respondents visited UGS more frequently than women and Latinx respondents visited UGS less J o u r n a l P r e -p r o o f frequently than others. People who considered UGS to be more important for their health and those whose most-visited UGS provides more of the features they deem important visited UGS more frequently, while those concerned about social distancing, access to UGS, the safety of UGS, or crowded UGS visited UGS less frequently (Figure 3 ).

The COVID-19 pandemic has led to historic economic impacts, loss of life (Centers for Disease Control and Prevention, 2020b), long-term morbidity, and rates of unemployment not seen since The Great Depression, all of which have led to quantifiable stress among affected populations (Martin et al., 2020; Nicola et al., 2020) . Indeed, health professionals have warned that the pandemic and associated stress and isolation are creating a mental as well as a physical health crisis (Twenge and Joiner, 2020; Vindegaard and Benros, 2020) . During this time of crisis, UGS may be increasingly important for providing health benefits to city dwellers (McCunn, 2020). Our survey of over one thousand New Yorkers demonstrated that many people recognize the importance of UGS for their health, and that this recognition has increased since the start of the pandemic. The survey also showed that many New Yorkers who use UGS have continued to use them, even during the deadliest period of the pandemic for NYC thus far.

However, we found that concerns about the safety of visiting UGS, a lack of UGS access, and a paucity of desired features can discourage people from using UGS and gaining associated health benefits. This may potentially further exacerbate the long-term negative health effects of the pandemic and disproportionate effects on Black, Latinx, and other often-marginalized groups (Andrasfay and Goldman, 2020) . Our results point to some practical strategies for providing opportunities for receiving health benefits from UGS during times of crisis.

We found that most survey participants considered UGS to be very or extremely important for their health across all groups (gender, race/ethnicity, and borough) analyzed in this study. Notably, UGS were considered to be more important for mental than physical health. Some reviews and synthesis on the effects of green space on health have suggested stronger relationships with mental health impacts, including stress restoration, mood, and mental health disorders, than on physiological stress, morbidity, and other physical health measures (Kondo et al., 2018; Lee and Maheswaran, 2011) . Access to UGS can encourage physical exercise, but UGS also have a variety of other uses that contribute to health and wellbeing, including socializing, community building, and fostering connections to nature (Lee and Maheswaran, 2011) . In NYC in particular, parks are a critical social space contributing to perceptions of community quality and cohesion (Auyeung et al., 2016) . With recognition that outdoor social interactions produce lower risk of disease transmission than indoor gatherings, UGS have become increasingly important venues for relieving social isolation and stress.

Several studies have found that people have used UGS more since the pandemic started (Grima et al., 2020; Kleinschroth and Kowarik, 2020; Lu et al., 2020; Venter et al., 2020) . Our survey respondents visited UGS multiple times per week on average, but respondents were equally likely to increase or decrease their UGS visits during the pandemic. Our results suggest that concerns about COVID-19 were a factor, as these were the most frequently selected concerns about visiting UGS and respondents who were concerned about social distancing in UGS were more likely to decrease their visits to UGS. Lack of mask wearing was also a concern, and some respondents commented that they would like to see more clearly stated policies and more enforcement of social distancing and mask wearing in parks. Here, the specific time and place of our study may have had an impact on the results. NYC had stricter lockdown policies J o u r n a l P r e -p r o o f than many U.S. cities, but less strict than many other regions, such as parts of Europe. New Yorkers may have been more wary, on average, of visiting UGS than people in less affected areas of the country or at later stages of reopening and more likely to wear masks than in other parts of the county where mask-wearing has been eschewed by political leaders.

Other factors also led to reduced UGS visits (compared to other respondents or pre-COVID-19 behaviors), including concerns about UGS access and safety and UGS that do not adequately meet people's needs. Access to UGS and the perceived safety of UGS were likely worsened by COVID-19, with park closures and potential virus spread from surfaces and strangers, but may also have been chronic issues unrelated to lack of neighborhood green space) that became prohibitive in the face of new stresses. Even when parks and other UGS have remained open, they have often had some areas or facilities (e.g. public restrooms, sports facilities, seating areas) that are closed. Partially opened or small parks, or large iconic parks that serve large populations, are likely to become crowded and cause stress for people looking for a reprieve, rather than a risk of infection, from going outside. Closed facilities could also make a UGS visit more difficult (e.g. visiting a park without a restroom or playground with small children) or simply less worthwhile. Addressing these factors could help encourage UGS users to continue to visit UGS during the pandemic; however, since few people who do not use UGS took our survey, we are limited in our ability to say that these actions would promote UGS use by people who are otherwise not inclined or able to use them. In addition, other factors that we did not specifically ask about may have been important deterrents from UGS use.

We found some key differences in perception and use of UGS across groups of people, as well as some inequalities that should be addressed in future UGS planning and management.

Despite a strong bias in our sample population towards relatively high-income, educated, white women and the use of coarse census-based race and ethnicity categories that ignore cultural differences, we were able to detect some significant differences in concerns (e.g. police presence) and important features of UGS (e.g. educational opportunities, wildlife, places to cook food) across groups. These differences demonstrate the importance of communication with and involvement of communities to find out what their UGS needs are (Gobster, 2002) . Participatory planning processes and procedural justice reforms could help to increase feelings of inclusion and safety in UGS, particularly for historically marginalized communities such as Black Americans who are often assumed to be disinterested in nature and the environment, creating a cycle of exclusion from public spaces (Finney, 2014) . Inequality in access, amenities, or inclusion in UGS could further exacerbate health inequities by limiting opportunities for outdoor exercise and stress restoration in nature (Hartig et al., 2014) . Indeed, in our study, respondents from Queens reported more concerns about a lack of UGS access and a lack of the features they deemed important being provided by local UGS. This is consistent with recent research that found that neighborhoods in Queens were disproportionately impacted by COVID-19, in part because of increased social vulnerability in Western Queens (Choi et al., 2020) .

For people to receive benefits from UGS during a public health crisis, the benefits of visiting a UGS need to outweigh the cost. The first barrier that cities can try to remove from this equation is to ensure that there is adequate and equitable access to UGS in the city. Even in NYC, a city with higher-than-average access to parks per capita compared to other U.S. cities (over 90% of residents living within a 10-minute walk of a park; The Trust for Public Land, 2020), concerns J o u r n a l P r e -p r o o f about a lack of safe and easy access prevented people from visiting UGS. Access may be even more prohibitive in other cities. Although NYC has UGS distributed throughout the city, there are differences across neighborhoods in the size and quality of UGS (Miyake et al., 2010; Weiss et al., 2011) . Access to larger parks or connected networks of parks may be particularly beneficial during the current pandemic because they offer greater ability to maintain social distance. In general, larger parks that are more accessible by public transportation receive more visitors (Hamstead et al., 2018) , but during the COVID-19 pandemic some typical methods of travel to UGS (e.g. subways) are considered high-risk, and traveling long distances to UGS is less feasible. Maintaining UGS access during the pandemic requires keeping UGS open and working to make public transportation systems safe (Slater et al., 2020) . Further, ensuring equitable access likely requires increasing available open space in some areas (even in NYC). In the short term, cities can use open and slow streets to increase outdoor recreation areas in neighborhoods that need them, but in the long term, equitable access will require prioritizing underserved neighborhoods for creating new UGS (Slater et al., 2020) .

A second barrier that may prevent people from visiting UGS in times of crisis is not getting what they need or expect from them. Many UGS provide some of the important features most commonly recognized by our respondents (i.e. trees, planted areas, places to sit and walk)

just by being open, but other features that are important for some groups, including restrooms, playgrounds, and places to cook food, have been closed due to concerns about disease spread. In addition, budget constraints or concerns about staff safety can lead to a lack of UGS maintenance during times of crisis, leading to complaints about cleanliness and aesthetics. Maintaining facilities and keeping them open is important, to the extent that the risk it presents to staff and the public is deemed acceptable by public health experts (Slater et al., 2020) . Such tradeoffs J o u r n a l P r e -p r o o f between providing benefits to UGS visitors and keeping people safe will be ongoing and potentially difficult to assess, especially when cities need to make decisions with incomplete information about the pandemic or other emergency situations. However, a real understanding of what communities need and want from UGS could be highly informative for determining priorities.

Finally, ensuring that people can get benefits from UGS requires that they are perceived as safe. Perceptions of safety are a major contributor to individuals' use of UGS, both in NYC (Weiss et al., 2011) and more broadly (Jansson et al., 2013) . During the COVID-19 pandemic, (Slater et al., 2020) . In comments, several survey respondents expressed a need for clear communication (e.g. signage at parks) for what behaviors were expected, particularly as information around the novel coronavirus evolved. Some respondents also expressed a desire for park staff to enforce recommended behaviors (e.g. by handing out masks). Public health agencies will need to determine what level of engagement with the public is safe for park staff, but there is a clear need for financial and informational support for UGS staff and for development of comprehensive and adaptable policy to address evolving public health crises in public spaces.

This study has several important limitations, partly due to the short time frame of study planning and implementation that was necessary to address the urgent question of how people were using UGS during the early stages of the COVID-19 pandemic; the timeliness of these results enabled us to share them with the NYC Mayor's Office of Climate Resiliency (MOCR) Rapid Research and Assessment (RRA) series (Kyrkjebo et al., 2021) for consideration in COVID-19 response planning. Our recruitment method was employed to receive responses rapidly, with the tradeoff of obtaining a relatively large but biased sample in a short time period. Many groups of people (including Black, Latinx, Asian, and other communities that have been most affected by the pandemic) were underrepresented in our dataset, limiting the generalizability of our results and our insights into these communities' needs. Indeed, our overall results may have differed if we had been able to reach more participants from these groups. More research is needed to fully understand how to promote UGS use and its associated benefits for all city dwellers, including more comprehensive surveys and methods that elicit more detail on users' perceptions and needs, such as in-depth interviews and participatory planning. Future spatially explicit analysis could also help identify physical and geographic factors that influence UGS use and perceptions, examine the effects of individual park closures on overall park use, and identify important management considerations for individual UGS in NYC and other U.S. (Spotswood et al., 2021) .

UGS can provide health benefits that may be all the more important for short-term relief during times of crisis and long-term preventative health. But people can only get these benefits if they have adequate access to UGS, are able to get what they need from them, and feel safe there (Bedimo-Rung et al., 2005) . During the COVID-19 pandemic, city and state governments have (Slater et al., 2020) ; these strategies will also be important to consider for future pandemics. Our findings also highlight the importance of determining the needs of diverse communities to guide UGS management, particularly during times of crisis when UGS play an important role in health and wellbeing. 

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Figure 1. Percent of respondents who selected each concern about visiting urban green space (UGS) (A) and important feature of UGS (G), and coefficients of binomial regressions predicting differences in these responses across gender, race/ethnicity, and NYC borough (B-F and H-L). Plots show the odds ratios of the given variables compared to their associated baseline group (e.g. Male : Female is comparing the relative likelihood of selecting the response for people who identified as male, compared to the baseline group Female). If a variable's coefficient is not sufficiently different from the baseline group, the odds ratio takes a value of one (shown as a vertical black line for reference). Statistically significant differences (those where a variable's 95% confidence bounds do not overlap one) are shown in darker shades. Upper limits of confidence intervals are capped at 8 for ease of viewing; confidence intervals that reach 8 have higher upper limits (see Appendix D Table 1 ). Model responses are shown for the five most frequently selected concerns (B-F) and features (H-L). Results for other responses can be found in Appendix D Table 1. Figure 2 . Perceived importance of urban green space (UGS) for mental (A) and physical (B) health before (left-hand bars) and since (right-hand bars) the start of the COVID-19 pandemic and results of ordinal regressions predicting differences in current (since COVID-19) perceived importance across gender, race/ethnicity, and NYC borough (C). A and B show the number of respondents who selected each category of importance of UGS for health (ranging from "Not at all important" to "Extremely important"). Shaded areas between bars are made up of lines connecting individuals' responses regarding the two time periods. Thicker shaded areas represent a larger proportion of respondents who made the same selections; for example, the largest proportion of respondents selected that parks were "Extremely important" both before and during the pandemic (darkest shaded area connecting the two top bars) for both mental and physical health. Model coefficients (C) are odds ratios comparing the relative likelihoods of respondents selecting a given response compared to a baseline group (e.g. Male : Female is comparing the relative likelihood of selecting the response for people who identified as male, compared to the baseline group Female). Coefficient estimates are shown as white dots and colored bars represent confidence intervals. Figure 3 . Change in urban green space (UGS) visits since the start of the COVID-19 pandemic (A) and frequency of UGS visits within the week preceding survey participation (B). We compared UGS visit frequency and change in UGS visits during the pandemic across gender, race/ethnicity, and NYC borough and in relation to respondents' perceived importance of UGS for health, features provided by their most visited UGS, and concerns about visiting UGS (C-D). Model coefficients from ordinal regressions (C) are odds ratios comparing the relative likelihoods of respondents selecting a given response compared to a baseline group (e.g. Male :

Female is comparing the relative likelihood of selecting the response for people who identified as male, compared to the baseline group Female). Coefficient estimates are shown as white dots and colored bars represent confidence intervals. Confidence intervals that do not overlap 1 (colored in darker shades) are considered significant differences between groups. Upper limits of confidence intervals are capped at 8 for ease of viewing; confidence intervals that reach 8 have higher upper limits (Appendix D Table 3 ).

How is quality of urban green spaces associated with physical activity and health? Urban Forestry & Urban Greening

Reductions in 2020 US life expectancy due to COVID-19 and the disproportionate impact on the Black and Latino populations

Coronavirus and the Cities: Explaining Variations in the Onset of Infection and in the Number of Reported Cases and Deaths in US Metropolitan Areas as of 27

Reading the landscape: Citywide social assessment of New York City parks and natural areas in 2013-2014. Social Assessment White Paper No

The significance of parks to physical activity and public health: a conceptual model

Is more always better? Exploring field survey and social media indicators of quality of urban greenspace, in relation to health

The psychological impact of quarantine and how to reduce it: rapid review of the evidence

ggalluvial: Layered Grammar for Alluvial Plots

Geographic differences in COVID-19 cases, deaths, and incidence-United States

Excess Deaths Associated with COVID-19

Queens Neighborhoods Hardest Hit by Virus Home to Many Service Workers [WWW Document

Residents' preferences and use of urban and peri-urban green spaces in a Swiss mountainous region of the Southern Alps

Doses of Neighborhood Nature: The Benefits for Mental Health of Living with Nature

Black faces, white spaces: Reimagining the relationship of African Americans to the great outdoors

Recreational ecosystem services in European cities: Sociocultural and geographical contexts matter for park use

A cross-sectional study examining predictors of visit frequency to local green space and the impact this has on physical activity levels

An {R} Companion to Applied Regression, Third Edition

Managing Urban Parks for a Racially and Ethnically Diverse Clientele

The importance of urban natural areas and urban ecosystem services during the COVID-19 pandemic

Geolocated social media as a rapid indicator of park visitation and equitable park access

City Park Facts

The relationship between selfreported exposure to greenspace and human stress in

The impact of COVID-19 on public space: an early review of the emerging questions -design, perceptions and inequities

Perceived personal safety in relation to urban woodland vegetation-A review

Environmental correlates of physical activity: a review of evidence about parks and recreation

COVID-19 crisis demonstrates the urgent need for urban greenspaces

Urban Green Space and Its Impact on Human Health

Rapid Research and Assessment on COVID-19 and Climate

The health benefits of urban green spaces: a review of the evidence

Moving beyond the neighborhood: Daily exposure to nature and adolescents' mood

Escaping to nature in pandemic: a natural experiment of COVID-19 in Asian cities

Urban residents' beliefs concerning green space benefits in four cities in France and Portugal

Socio-economic impacts of COVID-19 on household consumption and poverty

The importance of nature to city living during the COVID-19 pandemic: Considerations and goals from environmental psychology

Not Just a Walk in the Park: Methodological Improvements for Determining Environmental Justice Implications of Park Access in New York City for the Promotion of Physical Activity

Frequently Asked Questions : NYC Parks [WWW Document

New York State on PAUSE

The socio-economic implications of the coronavirus and COVID-19 pandemic: A review

R: A language and environment for statistical computing. R Foundation for Statistical Computing

A complex landscape of inequity in access to urban parks: A literature review

It's real, not fake like a park': Residents' perception and use of informal urban green-space in

COVID-19 lockdown allows researchers to quantify the effects of human activity on wildlife

Techniques to identify themes

Urban nature as a source of resilience during social distancing amidst the coronavirus pandemic

Recommendations for Keeping Parks and Green Space Accessible for Mental and Physical Health During COVID-19 and Other Pandemics

Who has nature during the pandemic? COVID-19 cases track widespread inequity in nature access across the United States

New York Park Score [WWW Document

US Census Bureau-assessed prevalence of anxiety and depressive symptoms in 2019 and during the 2020 COVID-19 pandemic

The influence of urban green environments on stress relief measures: A field experiment

American Community Survey 5-Year Data Profile

US Census Bureau 2019 National and State Population Estimates [WWW Document

Green space as a buffer between stressful life events and health

Modern Applied Statistics with S, Fourth Edition

Urban nature in a time of crisis: recreational use of green space increases during the COVID-19 outbreak in Oslo

COVID-19 pandemic and mental health consequences: Systematic review of the current evidence

Reconsidering access: park facilities and neighborhood disamenities in New York City

ggplot2: Elegant Graphics for Data Analysis

We would like to acknowledge input on the survey instrument and policy and