Massachusetts Urban Forest Diversity: Insights from an Undergraduate Service Learning Course

Urban forests are the trees, plants, and associated ecosystems on streetscapes as well as in public parks, private residential properties, and forested natural patches in cities, towns, and suburbs (Miller et al. 2015; Johnson et al. 2020). These ecosystems provide a suite of environmental, social, and economic benefits (Endreny 2018; Mei et al. 2021). However, urban trees face many challenges, including soil compaction, exposure to pollution, lack of growing space, diseases and insect pests, and construction (Scharenbroch et al. 2017; Roman et al. 2022). As a result, urban tree loss in the United States (US) has been occurring at a rate of 36 million urban trees annually (the equivalent of 175,000 acres), with a total urban tree canopy cover decline of about 1.0 percentage point from 2009 to 2014 (Nowak and Greenfield 2018).

Urban tree inventories are critical components for the successful management of urban forests (Bond 2013; Klobucar et al. 2020). An urban tree inventory is the systematic collection of measurements of individual urban trees that typically includes attributes like stem diameter, species, height, crown width, location, and health condition rating (Miller et al. 2015) and may be used to help estimate tree growth or determinants of tree removal (Esperon-Rodriguez et al. 2023). Data about individual trees can be aggregated across larger geographies to assess conditions related to urban forest systems, like overall tree species composition and size class. Without current data from inventories and assessments, urban and community forest managers and other decision-makers may be challenged to effectively monitor variables that maximize ecosystem service provision, mitigate hazards, and develop informed policies and management strategies, such as achieving desirable future tree species diversity goals (Klobucar et al. 2020).

A variety of stakeholders engage in contemporary urban and community forest assessment and inventory activities, including researchers, private contractors, government agencies, and citizen or civic volunteers (e.g., Bloniarz and Ryan 1996; Fazio 2015; Elton et al. 2022). The involvement of volunteer or other less formally trained personnel represents an important paradigm shift from traditional approaches to urban forest assessments and inventories, which were typically conducted by agency specialists (i.e., staff) and professional contractors. Volunteers now account for a small but important (5%) amount of municipal tree care performed in the US (Hauer et al. 2018), and specific initiatives have included high profile volunteer street tree inventory initiatives in large cities (Bloniarz and Ryan 1996; Johnson et al. 2018). These individuals are often motivated by intrinsic and extrinsic factors including altruism, environmental stewardship, the opportunity for social interaction, and the opportunity to gain new skills (Vogt and Fischer 2017; Roman et al. 2020; Elton et al. 2022). While conventional volunteer pools may generate desirable candidates for citizen science activities, ample opportunity may also exist to involve and incorporate student populations through service learning initiatives (Bringle and Hatcher 2007; Cowett and Bassuk 2012).

Service learning is a pedagogical approach usually involving student populations that incorporates community-based experience, academic learning objectives, and intentional reflection into the learning process (Gelmon et al. 2018). At colleges and universities, service learning typically involves a 3-way partnership between the academic unit, the community (e.g., government agency, civic association, nongovernmental organization), and the student. Service learning and citizen science have the potential to enhance undergraduate education through inquiry-based learning and data collection, research opportunities, and class projects (Oberhauser and LeBuhn 2012). Students in forestry and other natural resource or environmental studies programs have shown to benefit from experience-based learning activities with increased motivation and confidence, development of professional skills, integrated forms of knowledge, and enhanced sense of place (Ward 1999; Hix 2015; Watkins and Poudyal 2021). Undergraduate students at the University of Massachusetts Amherst indicated that they appreciated the “immersive, hands-on experience” of urban forestry field work service learning projects (Harper et al. 2021). Studentled data collection has been shown to enhance learning while also serving broader management needs (Gelmon et al. 2018).

Integrating ecological monitoring into undergraduate courses is not new (e.g., Hitchcock et al. 2021), and service learning also has the potential to help address important gaps in urban forest data. Though the importance of an urban forest inventory is widely recognized—83% of US communities have an inventory—only 41% of urban tree inventories are current (Hauer and Peterson 2016). Many barriers exist for communities to conduct an urban tree inventory and include the costs associated with data collection and management as well as the technical expertise to operate and maintain inventory software (Berland et al. 2019). Urban forest managers in under-resourced cities and towns are often tasked with making management and operational decisions with substantial limitations related to inventory data, personnel, and a paucity of scientific knowledge (Harper et al. 2017; Lass and Harper 2023).

In this study, we bridge the opportunity to assess connections between college student learning with management needs in urban and community forestry. The goal of this research is to assess tree diversity in Massachusetts, USA, by comparing statewide urban tree data collected by undergraduate university student researchers to professional statewide assessments (Cumming et al. 2006; Cowett and Bassuk 2020) and using that comparison as a reflection to discuss the potential broader application of student-led inventories, particularly at land grant universities (LGU).

We asked the following research questions:

What is the relative taxonomic composition, diversity, and size class distribution of urban forest trees across Massachusetts?

What is the extent of management-relevant regional variation across relative taxonomic composition, diversity, and size class distribution?

How do these student-collected data—and the affiliated findings—compare to conclusions based on other data collected by professionals?

As demonstrated by the findings in this undergraduate-led data collection initiative, American LGUs and the Cooperative Extension system are uniquely positioned to leverage and collaborate with undergraduate students in environmental monitoring and formal research. Systematic methodologies exist to inventory and monitor urban trees over time (Roman et al. 2020) and there is an expansive need to survey, record, and collect consistent data with standardized methods in urban and community forestry (Morgenroth and Östberg 2017).

We see the present undergraduate university studentled initiative from the University of Massachusetts as a complement to a model piloted at Cornell University, which connects skills-based training with larger data collection needs. In 2002, Cornell University piloted a program titled the Student Weekend Arborist Team (SWAT)(Cowett and Bassuk 2012). The SWAT program was designed specifically to meet the capacity needs of smaller, inadequately resourced communities that often lacked personnel, funding, and the baseline data needed to generate community forest management plans. Student participants were paid an $80 stipend for each day worked and also earned one academic credit to complete a half-day training session that included both classroom and hands-on instruction. The SWAT program has not only generated urban forest management-related capacity for underserved municipalities, but participating students also reported a greater confidence in their ability to identify trees, as well as to work as part of a team (Cowett and Bassuk 2012).

The formal involvement of undergraduate students in urban forestry and experience-based, service learning activities—like that demonstrated by the methodology of this study—has the potential to broaden engagement and to include greater numbers of historically marginalized groups in STEM degree programs. It may also help to focus the professional trajectory and long-term stewardship behaviors of students during the developmental stages of their career and more broadly support social and economic diversity in the urban and community forestry sector (Kuhns et al. 2004; Postles and Bartlett 2018; Westphal et al. 2022).

This study also highlights an important opportunity for urban forestry academic researchers and the broader communities of practice to learn more about student learning of technical arboriculture and urban forestry skillsets. While volunteer groups remain an important component of many urban forestry programs (e.g., Harper et al. 2018), sustaining a workforce requires informal and formal education opportunities for wage-earning professionals who are sometimes, though inconsistently, introduced to urban forestry and arboriculture during their undergraduate education (O’Herrin et al. 2018).

Undergraduate student learning may be explored through different modes of literacy evaluation, focusing on bioliteracy (the understanding and application of scientific topics), data literacy (understanding data collection, analysis, and visualization), and numeracy (understanding numeric measurements, scales, and units). Precedential studies in urban forestry education may be worth revisiting for new research undertakings; for example, McPherson (1984) found that 70% of industry professionals believe that undergraduate arboriculture and urban forestry students need to have at least 6 months of supervised field experience prior to entering the workforce, in spite of the varying skills expected of these separate professions (Table 3). Each of these skills represent different areas of knowledge learning outcomes from university courses and may be taught through different pedagogical strategies that optimize student learning—and the combinations and depth of research questions that consider the needs of different learners, differing student backgrounds, different teaching methods, and different topics could yield important information for the productivity of these industries and the well-being of individuals in the workforce.

In addition to student-related impacts, our findings also point to important natural resource management considerations collected via citizen or civic science. The dominance of the maple family, genus, and species in municipalities of the Northeastern US is not surprising (e.g., Ma et al. 2020), since maples have been a common replacement for urban elm trees (Ulmus spp.) following the introduction and proliferation of Dutch elm disease in previous decades (Cumming et al. 2006). This finding does, however, reinforce that the installation of new urban (nonmaple) trees may positively contribute to the diversity and resilience of tree populations within Massachusetts’ municipalities (Elton et al. 2020), statewide management regions (Cumming et al. 2006), and multistate regions like the Northeastern US (Doroski et al. 2020). The abundance of white pine (Pinus strobus) is also of note as it is susceptible to both pest pressure and structural failure (Wyka et al. 2018; Mcintire et al. 2021), and efforts to diversify urban forests could include a broader palette of coniferous species (Clapp et al. 2014). Though Santamour’s 10-20-30 rule is a widely recognized, simple metric through which to view urban forest diversity, many researchers continue to advocate more stringent or nuanced planting guidelines (Laćan and McBride 2008; Clapp et al. 2014; Ball and Tyo 2016). Advancing both understanding and potential inclusion of other guidelines is an important urban forest management consideration.

Our results suggest that volunteer data collection efforts alongside trained, university undergraduate students have the potential to generate accurate results and to supplement a more intensive urban tree inventory protocol at minimal cost to the local community. Numerous scholars and urban forestry practitioners have identified and discussed the importance of performing an urban forest inventory at the community level (e.g., Fischer et al. 2007; Ma et al. 2020). Urban forest inventories are a common management and assessment gap, and in the absence of current urban tree-related data, urban foresters may be forced to make management decisions about their urban natural resources with substantial knowledge limitations (Harper et al. 2017). While national and state-wide funding may provide capacity to manage urban forest resources at the regional and statewide-level, budget limitations, personnel costs, and inflation have continued to adversely impact urban forest management and operations at the local level (Healy et al. 2023). This study also contributes to the utility of civic or citizen science for cataloging components of urban nature (Hawthorne et al. 2015; Duchesneau et al. 2021).

This study is not without limitations. Researchers have conducted detailed evaluations to assess student learning outcomes from civic engagement and service learning initiatives in natural resources. Carr et al. (2011) studied the geospatial student learning outcomes of undergraduate forestry and natural resource students, finding that geospatial learning was below intended outcomes and that the assessment informed curricula improvement. Thompson and Licklider (2011) found student-generated concepts maps, illustrating conceptual hierarchies and connections among course topics, to be an effective model of assessment in an undergraduate urban forestry class. Though the undergraduate student-led urban forest inventory initiative was not designed to evaluate student learning beyond successful completion of the tree inventory, it could be considered a contributory citizen science undertaking, by which data was collected using systematic protocols that allow a high degree of student agency; however, the present study did not investigate how participants utilize and experience and make meaning of their involvement (e.g., Diprose et al. 2022)—an ample arena for future research given the pressing need to broaden (Lass and Harper 2023) and diversify (Chhin and Dahle 2024) the urban forestry workforce and the opportunity to accelerate student appeal to urban forestry professions (O’Herrin et al. 2018). Additional student assessments from the urban forest inventory initiative would build understanding and further inform updates to course curriculum.

While data collected from the present study was qualitatively screened by the course instructor, previous studies from urban and community forestry citizen science literature have explicitly engaged in data quality assessment. Volunteer geographic information has, for example, used validated citizen science data (e.g., controlled and approved on the basis of evidence, expert judgment, or knowledge rules) to supplement tree inventories and enable the mapping of allergenic tree species abundance (Dujardin et al. 2022). Other urban forestry research has evaluated data quality errors between samples collected by experts and less experienced personnel, like volunteers (Bloniarz and Ryan 1996), field crews (Roman et al. 2017), and boy scouts (Hallett and Hallett 2018). Long-term monitoring and tree inspections have also been studied and developed with volunteers and citizen scientists in mind (e.g., Vogt and Fischer 2017; Roman et al. 2020). Findings have consistently revealed a moderate-high consensus between expert and less experienced personnel, especially if course-level determinations are acceptable. Similar assessments that formally examine the quality of data from the undergraduate student-led urban forest inventory initiative would inform assignment (and course-related) curriculum updates. Findings may also apply beyond the classroom and inform data collection and tree monitoring programs in other municipalities.

Considerations of the nonrandom sample of municipalities selected by the students is also a limitation of this study. Since multicity analyses are subject to the willingness of collaborative partners for data sharing, and the existence of recent inventory data, it is not uncommon to have a sampling bias in comparable urban forest inventory analyses (e.g., Cowett and Bassuk 2017; Koch et al. 2018; Love et al. 2022). In other words, the nonrandom sampling that occurred in our study is a widespread challenge in urban forestry research. In the context of a university course, an approach to overcome this limitation would be to randomly assign a municipality to each student.

It is important to consider that this undergraduate university student-led sample urban forest inventory is not designed to formally inform some of the more nuanced aspects of urban forest management potentially included in a more expansive, professional urban forest inventory report (Morgenroth and Östberg 2017). For instance, student-derived recommendations related to the inspection and mitigation of urban trees for risk (of failure) should be reviewed with a tree risk assessment qualified (TRAQ) arborist. This is a clear example as to how an undergraduate student researcher cannot replace a trained, skilled arborist, or a professionally conducted urban tree inventory.

Additionally, data collected via convenience sample is not without limitation, including bias, error, and validity. Though there is precedence of informing useful conclusions from data that has been derived from a convenience sample in the natural resources sector (Day 1994; Etikan et al. 2016; Lass and Harper 2023), other data collection protocols might also be considered. These may include deliberately randomized samples that would better statistically reflect the nature of the community’s urban forest and sampling beyond current limitations (i.e., 100 trees) to include greater numbers of trees per community as well as the collection of more attributes. The incorporation of virtual data-collection methods with undergraduate students may also be worthy of further exploration.

Undergraduate student-led urban forest data collection efforts yielded findings that largely align with conclusions from previous literature in relation to statewide urban tree taxa abundance, diversity, and size class distributions. Management-relevant differences between species and genus-level diversity as well as disease and pest-prone taxa emerged, especially in subregions across the state. Key areas for future research might emphasize experimentation with approaches to data collection, formal assessment of student learning outcomes, and formal data quality validation. With an ever-increasing demand on local budgets and the continuous need for current urban tree-related data, the incorporation of trained undergraduate natural resource student volunteers into urban forest management practices—namely urban tree inventory efforts—may yield useful data that will inform conclusions and supplement more intensive urban tree inventory protocols at broader scales.

The authors reported no conflicts of interest.