National Assessment of Tree City USA Participation According to Geography and Socioeconomic Characteristics

Characteristics of TCUSA Communities

TCUSA participation was positively associated with community population nationally (Figure 2) and at the regional scale (Table 3; Figure 3). Furthermore, long-term TCUSA participants had higher populations than those enrolling more recently, and the population of new enrollees appears to be leveling off over time (Figure 4). Taken together, these results indicate that while larger cities were early adopters of TCUSA and were more likely to participate overall, smaller communities are closing the participation gap slowly over time. These findings are not surprising, as research has long shown that larger cities are early adopters of innovative programs (e.g., Crain 1966) and are more likely to have municipal tree programs (Zhang and Zheng 2012). This may be attributable to multiple factors, such as higher staffing or preexisting management structures in larger municipalities that facilitated enrollment in TCUSA, or perhaps residents in larger municipalities more actively supported the implementation of community forestry programs.

Note that the analysis only considered current TCUSA participants, and thus did not account for communities that stopped participating in the program at some point. TCUSA recertification rates for 2013 were >90% for most states (Arbor Day Foundation 2014), but it is possible that losing small numbers of participants each year would have appreciable effects over time. Regardless of how the findings are influenced by discontinued enrollments, it is apparent that larger communities are more likely to participate (Figure 2) and remain in the program over time (Figure 4).

Logistic regression analyses with socioeconomic indicators provided further insights into the characteristics of TCUSA communities. In every region, owner occupancy rate was negatively associated with TCUSA participation at P < 0.01 (Table 3). This potentially can be explained by a coincidence of relatively high TCUSA participation rates and low owner occupancy rates in urban areas compared to suburban and rural areas. According to the U.S. Census Bureau (2014), national owner occupancy rates are approximately 60% in urban areas and 79% in rural areas, so the statistical finding may be indirectly indicating a higher level of TCUSA participation near urban centers. This is corroborated by positive associations between TCUSA participation and population in eight of nine U.S. regions (Table 3). While researchers did not explicitly test whether TCUSA participation was higher in metropolitan areas than rural areas, the findings regarding owner occupancy and population suggest this was indeed the case. The implication of this pattern may be that community forestry programs, in general, are encouraging tree planting and maintenance where the private citizenry may be less motivated to do so. Past research has suggested that tree planting is less common on renter-occupied property because absentee property owners (who do not occupy the home) and renters (who do not own the property and may move away before benefits are realized) are both less motivated to plant trees than owner occupants (Perkins et al. 2004). By promoting publicly maintained trees, TCUSA potentially compensates in part for reduced tree planting on private property in communities with low owner occupancy rates.

Education (high school graduation rate) and percent white population were positively associated with TCUSA participation in seven and five U.S. regions, respectively (Table 3). Past research has identified instances in which tree canopy cover was distributed unequally according to education (Heynen and Lindsey 2003) and race (Landry and Chakraborty 2009). It is possible that uneven participation in TCUSA will perpetuate such inequalities in canopy cover because TCUSA requires active management and funding of community forestry programs. More broadly, if one assumes that TCUSA delivers on its goals to yield social, economic, and environmental advantages for participant communities (Arbor Day Foundation 2014), then reduced participation rates in communities with lower education levels or higher minority populations may signal regional-scale patterns of environmental injustice [for recent articles on environmental justice in the context of urban forestry, see Berland et al. (2015) and Schwarz et al. (2015)]. In other words, residents of communities with lower educational attainment or higher minority populations may not have equal access to the advantages provided by TCUSA. More information on the realized benefits of TCUSA participation (as opposed to expected or anecdotal benefits) is needed to more fully address this issue. Additional research is needed to understand why these patterns are restricted to certain regions, and why municipal leaders do not pursue TCUSA status in communities with lower educational attainment or higher minority populations. For example, is this pattern driven by scarce municipal resources, prioritization of other programming instead of community forestry, or is it attributable to a lack of program awareness, perceived benefits, or citizen interest?

The Geography of TCUSA Participation

TCUSA participation rates varied substantially among U.S. regions. The clearest national pattern was relatively low participation in the northeastern U.S. and high participation in the Midwest (compare Figure 1 and Table 1). Other patterns were less consistent. For instance, in the western U.S., participation rates were high in the Mountain region but low in the Pacific region (Table 1). The authors are unable to explain the variability in TCUSA participation rates among regions, but it is clear that opportunities exist to increase participation rates in all regions, and particularly in New England, Middle Atlantic, West South Central, and Pacific regions (Table 1).

Strategies to increase TCUSA participation likely need to be devised at regional or finer scales, because each region has unique population (Table 2; Figure 3) and socioeconomic (Table 3) profiles with respect to participation. For example, Middle Atlantic communities around 96,000 population have even odds of participating in TCUSA, whereas West North Central communities larger than 10,000 population are already more likely than not to participate (Table 2; Figure 3). As such, while it may be worthwhile to target Middle Atlantic communities around 100,000 population as new enrollees, participation may already be near saturation for larger communities in the West North Central, and efforts there may instead target smaller communities under 10,000 population. To improve the chances of success, efforts to increase participation in these two disparate regions would need to account for differences in institutional capacities between larger and smaller government units on top of other regional predilections. In light of regional differences, it may be appropriate to organize support for community forestry programs like TCUSA at the state level, perhaps through existing organizations (e.g., relevant state agencies or extension offices). State-level efforts would cast a relatively wide net while allowing for efforts to be tailored to the unique characteristics of each state.

Positive spatial autocorrelation in program participation suggests that communities are more likely to enroll in TCUSA if neighboring communities participate. This pattern could emerge spontaneously due to spatial clustering of communities with demographic profiles amenable to TCUSA participation (e.g., lower owner occupancy, higher education). Alternatively, the spatial autocorrelation could be explained by diffusion of ideas via professional networks, as municipal leaders would be more likely to share ideas with their counterparts in neighboring communities as opposed to more distant communities. Similarly, municipal leaders may feel pressure to participate in TCUSA as a means of upholding the community’s prestige in relation to its neighbors. Grove et al. (2006) described this phenomenon as an “ecology of prestige” for households within a neighborhood; here, researchers potentially observed an analogous desire to keep up with neighboring communities by enrolling in TCUSA.

Conclusions and Future Directions

In the U.S., TCUSA is the most prominent program recognizing a commitment to community forestry. The study was designed to provide a national overview of TCUSA participation according to geography, community size, and socioeconomic characteristics. In the analysis, nearly one-quarter of U.S. communities ≥2,500 population participated in TCUSA, and this represented over one-half of the total population in those communities. Program participation varied substantially among U.S. regions, from 11.5% of communities in the Middle Atlantic region to 44.4% in the West North Central region. TCUSA participation rates were higher in more populous communities, and long-term participants were more populous than recent enrollees. Across the majority of U.S. regions, TCUSA participation was negatively associated with owner occupancy rates, and positively associated with high school graduation rates and percent white population. These relationships could signal instances of environmental injustice if the presumed benefits of TCUSA participation are distributed unevenly with respect to educational attainment or racial composition. Patterns for income and median population age were less clear across regions. Finally, TCUSA participants exhibited positive spatial autocorrelation, as communities were more likely to participate if their neighbors participated.

This analysis represents the first national assessment of TCUSA participation, and as such, the findings provide novel insight into the characteristics of TCUSA communities. Continued research on TCUSA participation and attendant benefits will help improve community forest management, and the patterns described here can be used to frame future research questions. For example, identifying motivating factors and barriers to TCUSA participation could help gauge interest for TCUSA in non-participant communities, overcome barriers to participation, and reduce inequalities in program participation rates. In future analyses, additional factors, such as population density, existing tree canopy cover, and management responses to forest pests or changing municipal budgets, may provide insight into patterns of TCUSA participation over time and space. Finally, investigating the social (e.g., community education and engagement), economic (e.g., increased business and tourism activities), and environmental impacts (e.g., reduced stormwater runoff and urban heat island effects) of TCUSA could help quantify the advantages of participation, and may ultimately point to changes in the program that would help communities maximize the benefits associated with participation.