Synthesis

INTRODUCTION

Street trees and the potential for mental health benefits

Cities around the world are facing significant and overlapping challenges around urbanization, climate change, and health and social inequalities (Grimm et al. 2008, Corburn 2017, Heaviside et al. 2017, Orimoloye et al. 2019). Street trees are a type of nature-based solution (NBS) with the potential to reduce the urban heat island effect, improve air quality, and increase physical activity (Roy et al. 2012, Andersson-Sköld et al. 2015, Mullaney et al. 2015, Salmond et al. 2016, Wolf et al. 2020). In selecting street trees as an NBS of focus, we were guided by stakeholders in a multicomponent EU-funded project (REGREEN: Fostering nature-based solutions for smart, green and healthy urban transitions in Europe and China, https://cordis.europa.eu/project/id/821016), who identified street trees as a priority NBS across the project’s three European Urban Living Labs (Aarhus, Denmark; Paris, France; Velika Gorica, Croatia).

We follow Salmond et al. in defining street trees broadly as “trees along streets,” including trees along the sides of major roads, in medians, and along residential streets but excluding trees in other urban spaces. Interest in the climate adaptation potential of street trees is growing, and local and national campaigns have been launched with the aim of preserving existing street trees and planting large numbers of new street trees (Rae et al. 2010, Salmond et al. 2016, Watkins et al. 2017, Werbin et al. 2020).

Street trees may impact mental health, for example, by reducing stress through contributing to a more pleasant living environment or by increasing Seasonal Affective Disorder through reducing ambient light in the winter (de Vries et al. 2013, Salmond et al. 2016). We define mental health as being “a state of well-being in which an individual realizes his or her own abilities, can cope with the normal stresses of life, can work productively, and is able to make a contribution to his or her community” (World Health Organization 2018). Street tree density has been shown to have a protective effect on mental health, especially among individuals with low socioeconomic status (SES) (Marselle et al. 2020). Given that the burden of disease related to mental ill-health is typically higher in cities (Vigo et al. 2016, Gruebner et al. 2017, Okkels et al. 2018), and projected to increase globally (Foreman et al. 2018), understanding how to maximize the benefits of street trees for mental health will enable city governments to make the most of this purported co-benefit of street trees.

Rationale for taking a systems thinking approach

Systems thinking provides a series of tools specifically designed to consider multiple outcomes, feedback loops, and unexpected consequences (Sterman 2000, Meadows and Wright 2009, Rutter et al. 2017, Petticrew et al. 2019, McGill et al. 2021). Within this perspective, an intervention (such as a street tree planting program) interacts with and impacts an underlying system, rather than producing a linear impact independent of the system (Petticrew et al. 2019). Instead of asking whether street trees “work” to improve mental health, a systems perspective leads us to ask a different question: How may street trees impact, and be impacted by, the underlying system around mental health over time (Petticrew et al. 2019, Skivington et al. 2021)? By adopting this approach, we also hope to be able to identify potential unexpected consequences of street trees for mental health, which in turn may enable municipal stakeholders to preempt any undesirable impacts and maximize potential benefits.

Causal Loop Diagrams (CLDs) are one systems thinking tool, which were developed to map and visualize hypothesized causal relationships and feedbacks between components of a system (Sterman 2000). CLDs provide a useful way of visually summarizing system structures and can be used in conjunction with systems archetypes to identify dynamic properties of an intervention within a system (Kim 1994).

Aim

We aim to develop systems thinking-informed hypotheses around the relationship between street trees and mental health, taking into account unintended consequences and feedback patterns over time.

METHODS

Figure 1 summarizes the approach we took, drawing on systems thinking best practices (Sterman 2000, Meadows and Wright 2009).

Table 1 provides a glossary of terms used.

Stage 1: Replicating a systems thinking depiction of mental health

We followed guidance to consider the underlying system before considering how an intervention may impact the system (Hawe et al. 2009, McGill et al. 2020). We conceptualized “mental health” as an outcome of a complex adaptive system comprised of multiple simultaneous interrelated determinants (e.g., socioeconomic status, physical health, stress, sleep problems, etc.) (Wittenborn et al. 2016, Langellier et al. 2019).

To describe this system, we drew on a systems thinking-informed CLD around the determinants of negative affect and major depressive disorder (MDD) developed by Wittenborn et al. based on the results of a systematic review (Fig. 2A for a simplified version) (Wittenborn et al. 2016). Negative affect entails feelings of “anxiety, sadness, fear, anger, guilt and shame, irritability, and other unpleasant emotions”(Stringer 2013) and is a key dimension of MDD when experienced over a period of two weeks or more (Gellman and Turner 2013). We were not able to identify an evidence-based CLD of mental health more broadly (Langellier et al. 2019), but many of the pathways in Wittenborn et al.’s model (e.g., economic status and stress, physical health, and physical inactivity, etc.) contribute to broader conceptions of mental health (Dolan et al. 2008), and a high ratio of positive to negative affect has been associated with flourishing mental health, while high levels of negative affect are associated with diminished mental health (Diehl et al. 2011).

Wittenborn et al. use the node “dysfunctional behaviors” to capture a wide range of behaviors, including medication non-adherence, poor diet, and perpetrating domestic abuse. However, the potential interactions and upstream determinants of this wide range of behaviors may be obscured through this aggregation. In addition, additional factors are absent, such as intergenerational trauma (Sangalang and Vang 2017, Barlow 2018), and a wider range of physiological factors. Acknowledging these limitations, we made the pragmatic decision to use the Wittenborn model as a proxy for broader determinants of mental health, given our hypothesis-generating research aim. We adapted their model with a focus on the social and economic dimensions. For example, while the Wittenborn model includes a pathway from physical inactivity to cortisol and then to sleep problems; we simplified this chain for our purposes, linking physical activity directly with sleep problems.

Stage 2: Identifying empirical studies

We were interested in identifying empirical studies which assessed the relationship between street trees (including determinants and impacts of street trees) and components of the system identified in Step 1 around mental health. Given our research aim to generate systems thinking-informed hypotheses, a traditional systematic review was neither appropriate nor feasible. Instead, we followed Lorenc et al. (Lorenc et al. 2012, 2014) and focused on identifying and synthesizing a wide range of relevant conceptual material, prioritizing the inclusion of evidence from a range of disciplines.

We began with several recent reviews of street tree impacts (Mullaney et al. 2015, Salmond et al. 2016, Wolf et al. 2020) and used these to identify relevant references, using a citation-chasing approach to identify primary empirical studies (both quantitative and qualitative) (Cooper et al. 2017). We supplemented this approach with targeted Google Scholar searches to identify evidence around intermediate links. For example, we found evidence linking street trees with changes in temperature and subsequently conducted a targeted search to identify evidence about the relationship between temperature and sleep problems (Rifkin et al. 2018). This mirrors a process used previously in the development of conceptual frameworks (Schram et al. 2017).

After several studies had been identified to provide evidence for a given hypothesized causal relationship, we focused attention on other relationships, prioritizing breadth and diversity of evidence. We also consulted with several senior academic researchers with expertise around street trees to identify additional relevant references.

Stage 3: Extracting data

For each included study, we extracted data on the following:

• Cause
• Polarity (+/-) of relationship
• Effect
• Delay (yes/no) (e.g., is there a 5+ year lag between the cause and effect?)
• Evidence type (empirical, review finding, assumption)
• Excerpt (to illustrate the hypothesized causal relationship)

The data extraction form was designed to enable translation into causal loop diagrams in Stage 4. Data were extracted using a standardized Excel template (Deegan 2009) by one member of the study team (MA), and every relationship was independently reviewed by a second member of the study team (RL, BW, TT, CG, JF). Disagreements were resolved through discussion. Following Lorenc et al., we broadly considered relevance and rigor, drawing on an approach informed by realist synthesis (Pawson et al. 2004, Lorenc et al. 2012, 2014).

Stage 4: Extending the mental health CLD

We conceptualized street trees as an exogenous intervention (i.e., an external influence) within the system described in Stage 1. We integrated the hypothesized causal relationships around street trees and their determinants/impacts with Wittenborn’s model. In some cases, there were direct connections (e.g., between street trees and stress, a key component of the mental health model) (de Vries et al. 2013), and in other cases, connections were indirect (e.g., the impact of street trees on temperature and of temperature on sleep problems, another key component of the mental health model).

Stage 5: Generating systems thinking-informed hypotheses

We used the final CLD and systems archetypes to develop systems thinking-informed hypotheses about the relationship between street trees and mental health over time. We asked ourselves the following questions: How would key variables in the CLD change over time? How would these changes over time vary under different starting conditions? We developed qualitative behavior over time graphs (Kim 1994) of key variables to reflect our hypotheses and then interrogated the drivers of these patterns.

We also used systems archetypes (Kim 1994, Wolstenholme 2004) to consider various aspects of the final CLD. Systems archetypes reflect patterns of feedback loops which can be found in many types of systems, and which can be used as diagnostic tools for understanding aspects of a complex system (Kim 1994). For example, we drew on the implication of a “fixes that fail” archetype, which suggests that over time “the problem symptom returns to its previous level or becomes worse” (Kim 1994).

Stage 6: Validation

Finally, we shared the database and CLDs with street tree experts and incorporated feedback in iterative revisions. Representatives (municipal planners, ecologists, and regional development staff) from each of the three European REGREEN Urban Living Labs were asked to reflect on the early results and identify ways in which they did or did not reflect local experiences with street trees in each city. This feedback was used to further revise the CLDs.

RESULTS

We summarize the process of developing the CLDs briefly and then focus on key findings based on the validated CLDs.

Stage 1: Replicating a systems thinking depiction of mental health

We present a simplified, adapted version of Wittenborn et al.’s causal loop diagram Figure 2, Panel A (Wittenborn et al. 2016). We take this model as a starting point and focus on extending it by integrating evidence around street tree determinants and impacts in Stages 2–4.

Stage 2: Identifying empirical studies

We identified 77 papers about street trees (and their impacts/determinants) and mental health (Appendix 1).

Stage 3: Extracting data

We extracted data on 94 hypothesized causal relationships (Appendix 2: Table 1). In addition, we made assumptions about a limited number (n=17) of relationships. For example, we assumed that tree vandalism reduces the attractiveness of a neighborhood given the well-established “broken windows theory,” (Wilson 1982) although we did not specifically find evidence on this. These relationships are clearly tagged as “assumption” within the database.

Stage 4: Extending the mental health CLD

We used this database to extend Wittenborn’s model and developed a detailed CLD with 128 links connecting 69 variables (94 links from the street tree evidence, 17 from Wittenborn et al.’s model, and 17 assumptions). The full CLD can be viewed and interrogated on the web using a standard browser (https://kumu.io/ecehh/street-trees-mental-health-overall-causal-loop-diagram-18d0) and has also been reproduced in Appendix 3, Figure 1.

Using this detailed CLD, we identified 20 interconnected pathways through which trees may have a beneficial effect on mental health (Appendix 4: Table 1). We categorized these broadly based on the three domains of the pathways by which nature contributes to health as identified by Markevych et al. as either 1) reducing harm (e.g., reducing exposure to environmental stressors), 2) restoring capacities (e.g., restoring attention or stress recovery), or 3) building capacities (e.g., encouraging physical activity or social cohesion) (Markevych et al. 2017).

We also identified at least eight pathways through which street trees may have a negative impact on mental health (Appendix 4: Table 2). We categorized these broadly, considering the inverse of the domains identified by Markevych et al. (i.e., increase harm, reduce capacities, prevent capacity building) (Markevych et al. 2017).

Stage 5: Generating systems thinking-informed hypotheses

For ease of interpretation, we present somewhat simplified CLDs here (Fig. 2, Panels A–D). Figure 2 Panel A reflects our adapted version of Wittenborn’s model and shows how multiple reinforcing loops contribute to increased negative affect. In Panel B, we introduce a simplified representation of a street tree system, highlighting that street tree canopy size is driven both by tree health (Blunt 2008, Ely 2010) and overall number of street trees. Both of these relationships occur over a longer time horizon (e.g., it takes time for trees to grow to maturity), which is indicated by the delay marks (double perpendicular bars) within the CLD (standard notation). Tree health is critical for both increasing the canopy size (e.g., stressed trees do not grow very well or quickly) and for maintaining the total tree stock (e.g., stressed trees often have shorter lifespans).

Resources for street trees are required to maintain street tree health. Unlike trees in natural settings, street trees require support to thrive in otherwise challenging urban conditions. As one forester summarized:

A lot of people think that planting a tree is simple - dig a hole in the ground and walk away - well then you’re doomed from the beginning. Urban trees have a tough life. They need planning and long-term care and commitment. (Shcheglovitova 2020)

Resources for street trees can also lead to the planting of additional trees, increasing the total number of trees (reflected in the simplified system in Panel B).

Many of the benefits of street trees are linked to tree-canopy size (e.g., cooling effects (Rahman et al. 2011), noise reduction (McPherson et al. 2002), shade (Shashua-Bar et al. 2009), etc.). In Panel C, we highlight some of the direct and indirect pathways through which the street tree canopy size may impact negative affect and determinants of mental health in the Wittenborn et al. model. For example, street trees have been hypothesized to reduce negative affect directly, and also to improve cognitive performance, reduce sleep problems, and reduce perceived stress. The relationships shown here are only a subset of those represented in the detailed model (Appendix 3: Fig. 1)

In addition, street trees have been hypothesized to impact intermediary factors which then impact factors in the Wittenborn et al. model. For example, street trees may increase the attractiveness of a neighborhood, leading to increased walkability and thus increased physical activity. Street trees may increase views of nature, which in turn have been shown to improve physical health, reduce perceived stress, and improve cognitive performance. Finally, some impacts of street trees may also have a negative effect. For example, street trees may increase pollen, increasing allergies, and decreasing physical health and well-being. However, appropriate management can preempt many of the risks associated with street trees (e.g., planting appropriate species selection to reduce pollen, proactive pruning to reduce the risk of tree limb falls, etc.) (Brindal and Stringer 2009, Cariñanos and Casares-Porcel 2011, Trees and Design Action group 2014).

We have not shown all of the ways in which street trees may impact the Wittenborn et al. model here - for the complete representation, refer to Appendix 3, Figure 1.

Finally, in Panel D, we highlight the ways in which factors from the Wittenborn et al. model may feed into the availability of resources for street trees, in effect closing the loops between street trees and mental health. Resource availability for trees is in part determined by whether trees are seen as a valuable investment by residents and decision-makers. As people experience the benefits of street trees (not only in relation to negative affect, but also in terms of reductions in perceived stress, increased economic status through property values, increased interpersonal relationships through community cohesion, etc.), support for street trees may increase, justifying additional resource allocation. These impacts likely occur over a longer time horizon.

We analyzed the final CLD, drawing on systems archetypes to guide us in identifying deeper patterns. We identified or further explored the following hypotheses:

1. For street trees to impact mental health, the health of the trees themselves is critical.
2. Historical disparities in street trees are self-enforcing.
3. Uneven maintenance of street trees undermines tree-planting interventions, exacerbating mental health inequities.

We describe each hypothesis in more detail below.

Hypothesis 1: For street trees to impact mental health, the health of the trees themselves is critical.

As noted, for the positive impacts of street trees on mental health to be realized, the health of street trees is critical. As Widney et al. demonstrate, in a study of three U.S. cities, without intervention to increase the survival rate of street trees over time, premature tree mortality undermines the beneficial potential of street trees (Widney et al. 2016). Widney et al. inventoried 10%+ of all street trees planted between 2009–2011 in Detroit, Philadelphia, and Indianapolis, and calculated the annual and cumulative survival rates after 3–5 years (only 60–80% of trees were alive). Based on a modeling analysis, they highlighted that without improvement in the survival rate, only 40% of planted trees would be alive after another ten years, severely limiting the monetary benefits conferred by mature trees. While Widney et al. focused on benefits in terms of property value, energy savings, carbon, air quality, and stormwater effects, a similar pattern is likely to occur around mental health benefits.

As Shcheglovitova summarizes:

... planting trees means that tree needs must also be considered to some extent as well. Trees will not care for humans by attending to their needs if they are not alive. Tree needs and human needs are entangled as trees take on the role of active providers of care. (Shcheglovitova 2020)

Hypothesis 2: Historical disparities in street trees are self-enforcing

Since street trees provide value to neighborhoods, the historic stock of street trees may have contributed to the increasing affluence of tree-rich neighborhoods. In turn, more well-resourced neighborhoods are more effective at advocating for new trees. The reinforcing loops from street tree canopy to factors in Wittenborn et al.’s model, and then from these factors to resources for street trees (Fig. 2: Panel D) may help to explain the persistence of inequities in street tree distribution over time:

For example, a program that responds to resident requests for trees (an “opt-in” program), might actually result in more tree plantings (and subsequently higher future canopy cover) in wealthy neighborhoods where residents have access to information about and resources to take advantage of the program. (Watkins et al. 2017)

This echoes a “success to the successful” systems archetype in which one group has more access to resources initially, and this leads to higher performance or likelihood of succeeding, which in turn justifies additional resources (Kim 1994).

On the other hand, neighborhoods with historically low levels of street trees are less likely to have experienced their benefits and thus less likely to advocate for or invest in new trees. Where resources for street trees are low, tree health is likely to suffer, resulting in fewer trees and smaller canopies. The stressed trees are less likely to provide benefits, and residents are less likely to experience the value of street trees and in some cases even engage in, or fail to take active steps to prevent, tree vandalism. This has a knock-on effect: where tree vandalism is high, city governments tend to plant smaller (less expensive) trees, which again are less likely to produce experienced benefits (Pauleit et al. 2002). In the UK, residents cited not wanting trees because “they would be destroyed by vandals, wasting council tax payers' money for no benefit” (Hitchmough and Bonugli 1997). In the U.S., some residents have experienced an additional harm from street trees: their contribution to processes of gentrification of neighborhoods (Grove et al. 2018). This negative cost, combined with the lack of perceived or real benefits, makes street trees an unattractive intervention. This difficult to address reinforcing feedback loop may lead to low investments in street trees, with residents unlikely to perceive many benefits of street trees, further eroding future investments.

Differences in mental health and in the systems around mental health (e.g., economic status, perceived stress, dysfunctional behavior, interpersonal relationship quality) are the result of decades or centuries of complex power relations and reflect historic segregation, discrimination, and disenfranchisement. Patterns of street tree distribution and the reinforcing impact on perceived value and investment in trees continues (alongside many other factors) to further entrench these historic disparities (Landry and Chakraborty 2009, Grove et al. 2018).

Hypothesis 3: Uneven maintenance of street trees exacerbates mental health inequities

One apparent solution to address the stark differences in street tree coverage may be to invest in tree-planting programs in disadvantaged neighborhoods (setting aside the concerns about gentrification). However, without provisions for adequate maintenance, this intervention may only exacerbate inequities.

This can be summarized through the lens of a “fixes that fail” systems archetype, in which a problem is seemingly addressed by a solution in the short-term, only to be undermined in the long-term when the “solution” causes an unexpected impact, exacerbating the original problem (Kim 1994). While planting new trees in a deprived area may seem like a promising intervention, in the absence of appropriate resources for tree maintenance, tree health may suffer and eventually lead to increased tree decay/death. Ultimately, this apparent “fix” is temporary and may even exacerbate inequity in trees in low SES neighborhoods.

For example, in Baltimore the official housing strategy is to green “stressed” neighborhoods while prioritizing the provision of ongoing municipal services in “choice” neighborhoods to protect their value (Shcheglovitova 2020). However, this short-term “fix” of planting new trees in deprived neighborhoods backfired without appropriate maintenance:

...the effect of a dead tree on a city block is felt by the people who must live with them, especially in those neighborhoods that have experienced other forms of disinvestment. Ms Avery [...] remarks on this experience when she tells me about the trees that were planted in front of her home: ‘‘They planted them and they’re dead. They’re dead. You want to be making the neighborhood look better not worse. (Shcheglovitova 2020)

Far from improving mental health, dead or decaying trees may increase stress, worsen social cohesion, make neighborhoods less attractive, and have a dampening impact on many of the other pathways through which street trees may benefit neighborhoods (Appendix 4: Table 1).

Stage 6: Validation

We validated the CLDs and hypotheses with two prominent street tree experts through one-to-one meetings, during which we shared a database with all the identified cause–effect relationships, the detailed CLD, the final CLDs, and the systems thinking-informed hypotheses. We received feedback during these calls and in follow-up correspondence.

We also validated the CLDs with 14 municipal stakeholders from the three European Urban Living Labs (Aarhus, Denmark; Paris, France; Velika Gorica, Croatia), associated with the broader REGREEN project. We presented our early findings in a series of online workshops (one workshop with municipal stakeholders from each setting) and asked for general feedback (e.g., “What do you think?) and more specific feedback (e.g., “Are any factors missing which are important in your setting?”). These discussions led to the inclusion of several additional contextual factors (e.g., differences in temperature, public perception of street trees) and further refinements to the CLDs and hypotheses.

DISCUSSION

Statement of principal findings

We integrated quantitative and qualitative evidence around street trees and mental health to develop hypotheses about the relationship between street trees and mental health over time. These hypotheses have implications for maximizing the potential co-benefits of street trees for mental health, addressing entrenched inequities in street tree distribution, and highlighting the importance of investments in maintenance.

To develop evidence-based systems thinking-informed hypotheses, we integrated evidence from 77 empirical studies with a pre-existing model of MDD. We developed a detailed CLD (reflecting 128 hypothesized causal relationships) as an interim output intended to capture a series of hypothesized causal relationships rather than presenting an objective description of proven causal relationships. We presented a simplified version, and used this final CLD to develop three corresponding systems thinking-informed hypotheses.

First, to both optimize benefits and mitigate harms, the health of street trees themselves is critical. Resources for appropriate maintenance are essential for tree health, enabling trees to reach the age and size when many of the benefits to mental health are realized. Second, communities which have benefited from street trees in the past may be best-positioned to advocate for additional street trees and engage in tree stewardship in the present. At the same time, communities with scarce or poorly maintained street trees may not experience any benefits of street trees and limit future investment in trees. This dynamic further entrenches inequities in street tree coverage. Third, targeting low SES neighborhoods for new tree planting may fail if maintenance issues are ignored, potentially exacerbating inequities in mental health and beyond.

Strengths and weaknesses of the study

We used an innovative approach to synthesize insights across a wide range of conceptual material and develop systems thinking-informed theory around the relationship between street trees and mental health.

However, we also faced several limitations. First, we used a wide range of types of material (theoretical, quantitative and qualitative empirical research, case studies, expert opinion, etc.) and the validity and generalizability of findings across this evidence base are likely to vary widely. Rather than using this diversity of evidence to inform assessments of causality, we focused on developing causal hypotheses: proposed claims that remain to be further warranted by future empirical work. The process of identifying and selecting studies and hypothesized causal relationships for inclusion was systematic but necessarily subjective. These are common limitations faced by other attempts to synthesize a wide range of evidence in order to develop systems thinking-informed theory (Lorenc et al. 2012, 2014). Importantly, these CLDs are not intended to make robust empirical predictions, but rather to generate systems thinking-informed hypotheses, which can then be tested empirically and refined or refuted. We extracted excerpts from the studies which underpin each hypothesized causal relationship and report these in detail in Appendix 2, Table 1, in an effort to transparently demonstrate the range of evidence included in the development of these CLDs.

Second, these CLDs do not reflect population or community differences, (such as those between children/adults or across neighborhoods with different area-level SES), in tree type (e.g., between deciduous/coniferous), or according to local climatic or geographical conditions. It is likely that the relationships between trees and mental health vary across these factors in ways in which we were not able to capture (Collier 2020). For instance, tree species vary considerably, with implications for impacts (e.g., temperature, water availability, local climate, cultural acceptability, etc.). Instead of trying to tease apart differences by tree species, we included a broad category “appropriate tree planting and maintenance activities,” to acknowledge the importance of locally-tailored species selection. The ways in which street trees produce impacts may also vary seasonally, in ways which may affect various aspects of the Wittenborn et al. model differentially. However, we were not able to explicitly reflect this seasonality in the final CLD.

Third, we used the Wittenborn et al. model as a starting point for conceptualizing the system that produces mental health. However, limitations in this initial model (e.g., the exclusion of factors such as intergenerational trauma and the aggregation of diverse factors into several more generic nodes, such as “dysfunctional behavior”) suggest that there may be additional links between street trees and mental health which we did not identify and which may be important in developing holistic, systems-informed hypotheses in this area. Community-based system dynamics (Hovmand 2014) and group model building (Vennix 1999) are examples of systems-informed participatory community-oriented approaches which may help to address these gaps in the future.

Finally, most of the evidence that we drew on comes from North America, Northern and Western Europe, and Australia, and it is unclear to what extent these findings may be relevant in other settings. Even among the regions for which we found more evidence, there are likely to be contextual differences, which we did not identify or reflect in these CLDs. To begin exploring contextual differences, we validated the CLDs with local stakeholders in three European cities. This led to the inclusion of several contextual factors (e.g., differences in temperature, public perception of street trees, etc.), but it was challenging to represent nuanced context-specific pathways without developing an overwhelming, unusable CLD. Reassuringly, the overall themes captured in the CLDs appeared consistent across the three cities considered.

In relation to previous studies

The links between nature and mental health (Kuo 2015, Bratman et al. 2019), urban trees and mental health (Roy et al. 2012, Wolf et al. 2020), and overall benefits of street trees (Mullaney et al. 2015, Salmond et al. 2016) have been documented in a number of reviews. Many of the pathways summarized in Appendix 4, Tables 1 and 2, have been captured previously, albeit not with a specific street tree/mental health focus. However, to our knowledge this is the first systems thinking-informed assessment of how street trees may impact mental health.

This assessment led us to consider the politics of street trees: who benefits, who has experienced benefits in the past, whose trees are maintained, etc., and to contextualize street trees within wider and historical systems, all of which have links with mental health. For us, this highlights the value of taking this wide-ranging, open, and systems-inspired approach. We were surprised by the direction that formulating these hypotheses took us in, and perhaps this reflects the possibilities of bringing together insights across disciplines within a systems thinking framework.

Implications/meaning of the study

The hypotheses developed here have implications for how we design, monitor, and evaluate interventions related to street trees and mental health in urban environments.

For example, given the importance of tree health and tree-canopy size in realizing benefits, for mental health and for many of the other co-benefits associated with street trees, it would be important to develop an indicator that incentivizes investments in tree health (instead of a focus only on absolute numbers of new trees). From a systems thinking perspective, such an indicator may represent a powerful leverage point as a change in the “structure of information flows,” one of the more effective leverage points based on Donella Meadow’s classification (Meadows and Wright 2009). Such changes are especially powerful because they create information feedback loops, whereby city officials are either commended for being “at the top of the list” or motivated to make policy changes if they are shown to fall behind other similar cities. Along these lines, the Lancet Countdown series on health and climate change introduced a new environmental indicator in 2020: using remote sensing to measure green vegetation in large cities and ranking them according to levels of greenness (Watts et al. 2021). While not specifically about street trees, this measure is influenced by tree-canopy coverage more than by the number of new trees, and similar measures may create incentives to align action with the evidence base more closely.

Given the potential benefits of street trees for mental health, there is a strong mandate to find ways to effectively reduce street tree inequities without unintentionally causing other forms of inequity, such as that resulting from gentrification. Lower SES residents may experience the protective effects of street trees on mental health more than their higher SES counterparts (Marselle et al. 2020), and neighborhoods, which have been historically marginalized, may benefit from urban trees more than more privileged neighborhoods (e.g., based on different experiences of industrial pollution and potential remedial benefits of trees) (Vogt and Abood 2020). However, interventions to address differential access to street trees may exacerbate existing inequities if they do not address, at a minimum, street tree maintenance issues. Guidance around a “fixes that fail” archetype suggests that breaking the cycle requires “acknowledging up front that the fix is merely alleviating a symptom, and making a commitment to solve the real problem now” (Kim 1994). This may require deeply considering issues of disenfranchisement, historical legacies, community-based and community-led development, and a collective-impact approach (Vogt and Abood 2020). This is surely a much more ambitious project, but without addressing these root problems, short-term solutions may fail or even compound existing problems.

Finally, identifying ways in which street trees may contribute to the goals of other sectors (e.g., public health, transportation) will be key and may facilitate access to cross-sector resources in line with the expected benefits of street trees (Trees and Design Action group 2014). There is a compelling argument for public health authorities to consider street trees as a health intervention. Street trees may provide an “unintentional daily contact to nature” (Marselle et al. 2020), and their impact on mental health through pathways such as stress reduction or attention restoration may require less agency on behalf of the individual than other types of greenspace (e.g., parks) (Taylor et al. 2015). There have been increasing calls for low-agency population health interventions, which have the potential to prevent illness and minimize inequalities (Adams et al. 2016). With appropriate tree maintenance, street trees may represent such a low-agency intervention with the potential to reduce inequities in mental health. Finally, the restrictions of movement associated with COVID-19 may have increased the impact of street trees on local residents’ mental health, highlighting the need to understand the role trees can play in resilience building.

Future research

All of the hypotheses that we have proposed here can be empirically tested and further refined. For example, tree health, canopy size, or tree quality can be assessed quantitatively as a potential mediator between street tree quantity and mental health impacts. City-level street tree maintenance expenses could be compared or assessed over time and between jurisdictions. Case studies around municipal experiences with advocating for increased maintenance funding would be particularly relevant. A number of disciplinary perspectives and methodological approaches could test and build on the hypotheses put forth here.

CONCLUSION

Overall, we found that street trees have the potential to impact mental health in many ways, including by reducing harm, restoring capacities, and building capacities. Although street trees may also have negative impacts on mental health, many of these negative impacts can be mediated by appropriate tree management practices.

However, many of the beneficial impacts that street trees may have on mental health are contingent on the health of the trees themselves, and relatedly, on the availability of resources for appropriate tree maintenance. There are several systemic structures which we hypothesize produce the patterns we have described here, for example, around inequities in street tree coverage and underfunding of street tree maintenance. We used systems archetypes to develop systems thinking-informed hypotheses around these complex relationships.

In an era when urban trees are increasingly under pressure from densification, prioritization of other infrastructure, and increased negative impacts of climate change, it is all the more important that efforts to promote street trees take into account unintended consequences, feedback loops, and delayed effects. The use of causal loop diagrams and systems archetypes has enabled us to integrate quantitative and qualitative evidence to generate hypotheses that take these dynamics into account, identifying promising and potentially high-impact systems structures and intervention points and offering a critical analysis of some of the bottlenecks and potential leverage points which can be acted upon.

LITERATURE CITED

Adams, J., O. Mytton, M. White, and P. Monsivais. 2016. Why are some population interventions for diet and obesity more equitable and effective than others? The role of individual agency. PLOS Medicine 13(4):e1001990. https://doi.org/10.1371/journal.pmed.1001990

Andersson-Sköld, Y., S. Thorsson, D. Rayner, F. Lindberg, S. Janhäll, A. Jonsson, U. Moback, R. Bergman, and M. Granberg. 2015. An integrated method for assessing climate-related risks and adaptation alternatives in urban areas. Climate Risk Management 7:31-50. https://doi.org/10.1016/j.crm.2015.01.003

Barlow, J. N. 2018. Restoring optimal Black mental health and reversing intergenerational trauma in an era of Black Lives Matter. Biography 41(4):895-908. https://doi.org/10.1353/bio.2018.0084

Blunt, S. M. 2008. Trees and pavements—are they compatible? Arboricultural Journal 31(2):73-80. https://doi.org/10.1080/03071375.2008.9747522

Bratman, G. N., C. B. Anderson, M. G. Berman, B. Cochran, S. de Vries, J. Flanders, C. Folke, H. Frumkin, J. J. Gross, T. Hartig, P. H. Kahn, M. Kuo, J. J. Lawler, P. S. Levin, T. Lindahl, A. Meyer-Lindenberg, R. Mitchell, Z. Ouyang, J. Roe, L. Scarlett, J. R. Smith, M. van den Bosch, B. W. Wheeler, M. P. White, H. Zheng, and G. C. Daily. 2019. Nature and mental health: an ecosystem service perspective. Science Advances 5(7):eaax0903. https://doi.org/10.1126/sciadv.aax0903

Brindal, M., and R. Stringer. 2009. The value of urban trees: environmental factors and economic efficiency. The 10th National Street Tree Symposium.

Cariñanos, P., and M. Casares-Porcel. 2011. Urban green zones and related pollen allergy: a review. Some guidelines for designing spaces with low allergy impact. Landscape and Urban Planning 101(3):205-214. https://doi.org/10.1016/j.landurbplan.2011.03.006

Collier, B. 2020. There’s a trauma in loss of connection to nature, we need to stop saying that it’s not for us. March 10. http://www.bethcollier.co.uk/theres-a-trauma-in-loss-of-connection-to-nature-we-need-to-stop-saying-that-its-not-for-us/

Cooper, C., A. Booth, N. Britten, and R. Garside. 2017. A comparison of results of empirical studies of supplementary search techniques and recommendations in review methodology handbooks: a methodological review. Systematic Reviews 6(1):234. https://doi.org/10.1186/s13643-017-0625-1

Corburn, J. 2017. Urban place and health equity: critical issues and practices. International Journal of Environmental Research and Public Health 14(2):117. https://doi.org/10.3390/ijerph14020117

Deegan, M. A. 2009. Developing causal map codebooks to analyze policy recommendations: a preliminary content analysis of floodplain management recommendations following the 1993 midwest floods.

de Vries, S., S. M. E. van Dillen, P. P. Groenewegen, and P. Spreeuwenberg. 2013. Streetscape greenery and health: stress, social cohesion and physical activity as mediators. Social Science & Medicine 94:26-33. https://doi.org/10.1016/j.socscimed.2013.06.030

Diehl, M., E. L. Hay, and K. M. Berg. 2011. The ratio between positive and negative affect and flourishing mental health across adulthood. Aging & Mental Health 15(7):882-893. https://doi.org/10.1080/13607863.2011.569488

Dolan, P., T. Peasgood, and M. White. 2008. Do we really know what makes us happy? A review of the economic literature on the factors associated with subjective well-being. Journal of Economic Psychology 29(1):94-122. https://doi.org/10.1016/j.joep.2007.09.001

Ely, M. 2010. Integrating trees into the design of the city. The University of Adelaide. Adelaide, Australia.

Foreman, K. J., N. Marquez, A. Dolgert, K. Fukutaki, N. Fullman, M. McGaughey, M. A. Pletcher, A. E. Smith, K. Tang, C.-W. Yuan, J. C. Brown, J. Friedman, J. He, K. R. Heuton, M. Holmberg, D. J. Patel, P. Reidy, A. Carter, K. Cercy, A. Chapin, D. Douwes-Schultz, T. Frank, F. Goettsch, P. Y. Liu, V. Nandakumar, M. B. Reitsma, V. Reuter, N. Sadat, R. J. D. Sorensen, V. Srinivasan, R. L. Updike, H. York, A. D. Lopez, R. Lozano, S. S. Lim, A. H. Mokdad, S. E. Vollset, and C. J. L. Murray. 2018. Forecasting life expectancy, years of life lost, and all-cause and cause-specific mortality for 250 causes of death: reference and alternative scenarios for 2016-40 for 195 countries and territories. Lancet 392(10159):2052-2090. https://doi.org/10.1016/S0140-6736(18)31694-5

Gellman, M. D., and J. R. Turner, editors. 2013. Major depressive disorder. Pages 1187-1187. Encyclopedia of behavioral medicine. Springer, New York, New York, USA. https://doi.org/10.1007/978-3-030-39903-0_301110

Grimm, N. B., S. H. Faeth, N. E. Golubiewski, C. L. Redman, J. Wu, X. Bai, and J. M. Briggs. 2008. Global change and the ecology of cities. Science 319(5864):756-760. https://doi.org/10.1126/science.1150195

Grove, M., L. Ogden, S. Pickett, C. Boone, G. Buckley, D. H. Locke, C. Lord, and B. Hall. 2018. The legacy effect: understanding how segregation and environmental injustice unfold over time in Baltimore. Annals of the American Association of Geographers 108(2):524-537. https://doi.org/10.1080/24694452.2017.1365585

Gruebner, O., M. A. Rapp, M. Adli, U. Kluge, S. Galea, and A. Heinz. 2017. Cities and mental health. Deutsches Ärzteblatt International 114(8):121-127. https://doi.org/10.3238/arztebl.2017.0121

Hawe, P., A. Shiell, and T. Riley. 2009. Theorising interventions as events in systems. American Journal of Community Psychology 43(3-4):267-276. https://doi.org/10.1007/s10464-009-9229-9

Heaviside, C., H. Macintyre, and S. Vardoulakis. 2017. The urban heat island: implications for health in a changing environment. Current Environmental Health Reports 4(3):296-305. https://doi.org/10.1007/s40572-017-0150-3

Hitchmough, J. D., and A. M. Bonugli. 1997. Attitudes of residents of a medium-sized town in South West Scotland to street trees. Landscape Research 22(3):327-337. https://doi.org/10.1080/01426399708706518

Hovmand, P. S. 2014. Introduction to community-based system dynamics. Pages 1-16. Community based system dynamics. Springer, New York, New York, USA. https://doi.org/10.1007/978-1-4614-8763-0_1

Kim, D. H. 1994. Systems archetypes. Pegasus Communications, Cambridge, Massachusetts, USA.

Kuo, M. 2015. How might contact with nature promote human health? Promising mechanisms and a possible central pathway. Frontiers in Psychology 6:1093. https://doi.org/10.3389/fpsyg.2015.01093

Landry, S. M., and J. Chakraborty. 2009. Street trees and equity: evaluating the spatial distribution of an urban amenity. Environment and Planning A: Economy and Space 41(11):2651-2670. https://doi.org/10.1068/a41236

Langellier, B. A., Y. Yang, J. Purtle, K. L. Nelson, I. Stankov, and A. V. Diez Roux. 2019. Complex systems approaches to understand drivers of mental health and inform mental health policy: a systematic review. Administration and Policy in Mental Health and Mental Health Services Research 46(2):128-144. https://doi.org/10.1007/s10488-018-0887-5

Lorenc, T., S. Clayton, D. Neary, M. Whitehead, M. Petticrew, H. Thomson, S. Cummins, A. Sowden, and A. Renton. 2012. Crime, fear of crime, environment, and mental health and well-being: mapping review of theories and causal pathways. Health & Place 18(4):757-765. https://doi.org/10.1016/j.healthplace.2012.04.001

Lorenc, T., M. Petticrew, M. Whitehead, D. Neary, S. Clayton, K. Wright, H. Thomson, S. Cummins, A. Sowden, and A. Renton. 2014. Crime, fear of crime and mental health: synthesis of theory and systematic reviews of interventions and qualitative evidence. Public Health Research 2(2). https://doi.org/10.3310/phr02020

Markevych, I., J. Schoierer, T. Hartig, A. Chudnovsky, P. Hystad, A. M. Dzhambov, S. de Vries, M. Triguero-Mas, M. Brauer, M. J. Nieuwenhuijsen, G. Lupp, E. A. Richardson, T. Astell-Burt, D. Dimitrova, X. Feng, M. Sadeh, M. Standl, J. Heinrich, and E. Fuertes. 2017. Exploring pathways linking greenspace to health: theoretical and methodological guidance. Environmental Research 158:301-317. https://doi.org/10.1016/j.envres.2017.06.028

Marselle, M. R., D. E. Bowler, J. Watzema, D. Eichenberg, T. Kirsten, and A. Bonn. 2020. Urban street tree biodiversity and antidepressant prescriptions. Scientific Reports 10(1):22445. https://doi.org/10.1038/s41598-020-79924-5

McGill, E., V. Er, T. Penney, M. Egan, M. White, P. Meier, M. Whitehead, K. Lock, R. Anderson de Cuevas, R. Smith, N. Savona, H. Rutter, D. Marks, F. de Vocht, S. Cummins, J. Popay, and M. Petticrew. 2021. Evaluation of public health interventions from a complex systems perspective: a research methods review. Social Science & Medicine 272:113697. https://doi.org/10.1016/j.socscimed.2021.113697

McGill, E., D. Marks, V. Er, T. Penney, M. Petticrew, and M. Egan. 2020. Qualitative process evaluation from a complex systems perspective: a systematic review and framework for public health evaluators. PLOS Medicine 17(11):e1003368. https://doi.org/10.1371/journal.pmed.1003368

McPherson, E. G., S. E. Maco, J. R. Simpson, P. J. Peper, Q. Xiao, A. M. VanDerZanden, and N. Bell. 2002. Western Washington and Oregon community tree guide: benefits, costs and strategic planting. International Society of Arboriculture, Pacific Northwest Chapter, Silverton, Oregon, USA.

Meadows, D. H., and D. Wright. 2009. Thinking in systems: a primer. Earthscan, London, UK.

Mullaney, J., T. Lucke, and S. J. Trueman. 2015. A review of benefits and challenges in growing street trees in paved urban environments. Landscape and Urban Planning 134:157-166. https://doi.org/10.1016/j.landurbplan.2014.10.013

Okkels, N., C. B. Kristiansen, P. Munk-Jørgensen, and N. Sartorius. 2018. Urban mental health: challenges and perspectives. Current Opinion in Psychiatry 31(3):258-264. https://doi.org/10.1097/YCO.0000000000000413

Orimoloye, I. R., S. P. Mazinyo, A. M. Kalumba, O. Y. Ekundayo, and W. Nel. 2019. Implications of climate variability and change on urban and human health: a review. Cities 91:213-223. https://doi.org/10.1016/j.cities.2019.01.009

Pauleit, S., N. Jones, G. Garcia-Martin, J. L. Garcia-Valdecantos, L. M. Rivière, L. Vidal-Beaudet, M. Bodson, and T. B. Randrup. 2002. Tree establishment practice in towns and cities - results from a European survey. Urban Forestry & Urban Greening 1(2):83-96. https://doi.org/10.1078/1618-8667-00009

Pawson, R., T. Greenhalgh, G. Harvey, and K. Walshe. 2004. Realist synthesis: an introduction. ESRC Research Methods Programme Working Paper Series.

Petticrew, M., C. Knai, J. Thomas, E. A. Rehfuess, J. Noyes, A. Gerhardus, J. M. Grimshaw, H. Rutter, and E. McGill. 2019. Implications of a complexity perspective for systematic reviews and guideline development in health decision making. BMJ Global Health 4(Suppl 1):e000899. https://doi.org/10.1136/bmjgh-2018-000899

Rae, R. A., G. Simon, and J. Braden. 2010. Public reactions to new street tree planting. Cities and the Environment 3:21. https://doi.org/10.15365/cate.31102010

Rahman, M. A., J. G. Smith, P. Stringer, and A. R. Ennos. 2011. Effect of rooting conditions on the growth and cooling ability of Pyrus calleryana. Urban Forestry & Urban Greening 10(3):185-192. https://doi.org/10.1016/j.ufug.2011.05.003

Rifkin, D. I., M. W. Long, and M. J. Perry. 2018. Climate change and sleep: a systematic review of the literature and conceptual framework. Sleep Medicine Reviews 42:3-9. https://doi.org/10.1016/j.smrv.2018.07.007

Roy, S., J. Byrne, and C. Pickering. 2012. A systematic quantitative review of urban tree benefits, costs, and assessment methods across cities in different climatic zones. Urban Forestry & Urban Greening 11(4):351-363. https://doi.org/10.1016/j.ufug.2012.06.006

Rutter, H., N. Savona, K. Glonti, J. Bibby, S. Cummins, D. T. Finegood, F. Greaves, L. Harper, P. Hawe, L. Moore, M. Petticrew, E. Rehfuess, A. Shiell, J. Thomas, and M. White. 2017. The need for a complex systems model of evidence for public health. Lancet 390:10112 https://doi.org/10.1016/S0140-6736(17)31267-9

Salmond, J. A., M. Tadaki, S. Vardoulakis, K. Arbuthnott, A. Coutts, M. Demuzere, K. N. Dirks, C. Heaviside, S. Lim, H. Macintyre, R. N. McInnes, and B. W. Wheeler. 2016. Health and climate related ecosystem services provided by street trees in the urban environment. Environmental Health 15(S1):S36. https://doi.org/10.1186/s12940-016-0103-6

Sangalang, C. C., and C. Vang. 2017. Intergenerational trauma in refugee families: a systematic review. Journal of Immigrant and Minority Health 19(3):745-754. https://doi.org/10.1007/s10903-016-0499-7

Schram, A., A. Ruckert, J. A. VanDuzer, S. Friel, D. Gleeson, A.-M. Thow, D. Stuckler, and R. Labonte. 2017. A conceptual framework for investigating the impacts of international trade and investment agreements on noncommunicable disease risk factors. Health Policy and Planning 33(1):123-136. https://doi.org/10.1093/heapol/czx133

Shashua-Bar, L., D. Pearlmutter, and E. Erell. 2009. The cooling efficiency of urban landscape strategies in a hot dry climate. Landscape and Urban Planning 92(3-4):179-186. https://doi.org/10.1016/j.landurbplan.2009.04.005

Shcheglovitova, M. 2020. Valuing plants in devalued spaces: caring for Baltimore’s Street trees. Environment and Planning E: Nature and Space 3(1):228-245. https://doi.org/10.1177/2514848619854375

Skivington, K., L. Matthews, S. A. Simpson, P. Craig, J. Baird, J. M. Blazeby, K. A. Boyd, N. Craig, D. P. French, E. McIntosh, M. Petticrew, J. Rycroft-Malone, M. White, and L. Moore. 2021. A new framework for developing and evaluating complex interventions: update of Medical Research Council guidance. BMJ 374:n2061. https://doi.org/10.1136/bmj.n2061

Sterman, J. 2000. Business dynamics: systems thinking and modeling for a complex world. Irwin/McGraw-Hill, Homewood, Illinois, USA.

Stringer, D. M. 2013. Negative affect. Pages 1303-1304 in M. D. Gellman and J. R. Turner, editors. Encyclopedia of behavioral medicine. Springer, New York, New York, USA. https://doi.org/10.1007/978-3-030-39903-0_606

Taylor, M. S., B. W. Wheeler, M. P. White, T. Economou, and N. J. Osborne. 2015. Research note: urban street tree density and antidepressant prescription rates—a cross-sectional study in London, UK. Landscape and Urban Planning 136:174-179. https://doi.org/10.1016/j.landurbplan.2014.12.005

Trees and Design Action group. 2014. Trees in hard landscapes: a guide for delivery.

Vennix, J. A. M. 1999. Group model-building: tackling messy problems. System Dynamics Review 15(4):379-401. https://doi.org/10.1002/(SICI)1099-1727(199924)15:4%3C379::AID-SDR179%3E3.0.CO;2-E

Vigo, D., G. Thornicroft, and R. Atun. 2016. Estimating the true global burden of mental illness. Lancet Psychiatry 3(2):171-178. https://doi.org/10.1016/S2215-0366(15)00505-2

Vogt, J., and M. Abood. 2020. A transdisciplinary, mixed methods research agenda for evaluating the collective impact approach for tree planting: the CommuniTree initiative in northwest Indiana, U.S. Urban Forestry & Urban Greening 53:126735. https://doi.org/10.1016/j.ufug.2020.126735

Watkins, S. L., S. K. Mincey, J. Vogt, and S. P. Sweeney. 2017. Is planting equitable? An examination of the spatial distribution of nonprofit urban tree-planting programs by canopy cover, income, race, and ethnicity. Environment and Behavior 49(4):452-482. https://doi.org/10.1177/0013916516636423

Watts, N., M. Amann, N. Arnell, S. Ayeb-Karlsson, J. Beagley, K. Belesova, M. Boykoff, P. Byass, W. Cai, D. Campbell-Lendrum, S. Capstick, J. Chambers, S. Coleman, C. Dalin, M. Daly, N. Dasandi, S. Dasgupta, M. Davies, C. D. Napoli, P. Dominguez-Salas, P. Drummond, R. Dubrow, K. L. Ebi, M. Eckelman, P. Ekins, L. E. Escobar, L. Georgeson, S. Golder, D. Grace, H. Graham, P. Haggar, I. Hamilton, S. Hartinger, J. Hess, S.-C. Hsu, N. Hughes, S. J. Mikhaylov, M. P. Jimenez, I. Kelman, H. Kennard, G. Kiesewetter, P. L. Kinney, T. Kjellstrom, D. Kniveton, P. Lampard, B. Lemke, Y. Liu, Z. Liu, M. Lott, R. Lowe, J. Martinez-Urtaza, M. Maslin, L. McAllister, A. McGushin, C. McMichael, J. Milner, M. Moradi-Lakeh, K. Morrissey, S. Munzert, K. A. Murray, T. Neville, M. Nilsson, M. O. Sewe, T. Oreszczyn, M. Otto, F. Owfi, O. Pearman, D. Pencheon, R. Quinn, M. Rabbaniha, E. Robinson, J. Rocklöv, M. Romanello, J. C. Semenza, J. Sherman, L. Shi, M. Springmann, M. Tabatabaei, J. Taylor, J. Triñanes, J. Shumake-Guillemot, B. Vu, P. Wilkinson, M. Winning, P. Gong, H. Montgomery, and A. Costello. 2021. The 2020 report of The Lancet Countdown on health and climate change: responding to converging crises. Lancet 397(10269):129-170. https://doi.org/10.1016/S0140-6736(20)32290-X

Werbin, Z. R., L. Heidari, S. Buckley, P. Brochu, L. J. Butler, C. Connolly, L. H. Bloemendaal, T. D. McCabe, T. K. Miller, and L. R. Hutyra. 2020. A tree-planting decision support tool for urban heat mitigation. PLoS ONE 15(10):e0224959. https://doi.org/10.1371/journal.pone.0224959

Widney, S., B. C. Fischer, and J. Vogt. 2016. Tree mortality undercuts ability of tree-planting programs to provide benefits: results of a three-city study. Forests 7(3):65. https://doi.org/10.3390/f7030065

Wilson, J. Q., and G. L. Kelling. 1982. Broken windows. The Atlantic. March. https://www.theatlantic.com/magazine/archive/1982/03/broken-windows/304465/

Wittenborn, A. K., H. Rahmandad, J. Rick, and N. Hosseinichimeh. 2016. Depression as a systemic syndrome: mapping the feedback loops of major depressive disorder. Psychological Medicine 46(3):551-562. https://doi.org/10.1017/S0033291715002044

Wolf, K. L., S. T. Lam, J. K. McKeen, G. R. A. Richardson, M. van den Bosch, and A. C. Bardekjian. 2020. Urban trees and human health: a scoping review. International Journal of Environmental Research and Public Health 17(12):4371. https://doi.org/10.3390/ijerph17124371

Wolstenholme, E. 2004. Using generic system archetypes to support thinking and modelling. System Dynamics Review 20(4):341-356. https://doi.org/10.1002/sdr.302

World Health Organization. 2018. Mental health: strengthening our response. https://www.who.int/news-room/fact-sheets/detail/mental-health-strengthening-our-response

Copyright © 2023 by the author(s). Published here under license by The Resilience Alliance. This article is under a Creative Commons Attribution 4.0 International License. You may share and adapt the work provided the original author and source are credited, you indicate whether any changes were made, and you include a link to the license.