Beyond canopy targets: a performance- and risk-aware framework for urban tree species selection

Introduction

Cities are adopting ambitious tree canopy targets and large-scale planting programs to mitigate urban heat, air pollution, biodiversity loss, stormwater runoff, and climate-related risks (Jim and Chen, 2009; Pincetl et al., 2013; Yao et al., 2019). Canopy cover targets are attractive policy instruments because they are measurable, reportable, and verifiable, easily communicated, and facilitate the monitoring of urban greening commitments. However, canopy expansion is fundamentally a quantitative measure and does not necessarily indicate tree survival, long-term functionality under changing climatic conditions, or sustained net benefits for urban residents (Kenney et al., 2011). Consequently, canopy expansion alone does not ensure appropriate species composition or adaptation to local environmental conditions.

Urban forestry research has recognized these limitations through concepts such as the right tree, right place principle (Langenheim et al., 2020), climate-ready tree selection (Tabassum et al., 2023), functional and taxonomic diversity (Paquette et al., 2021), and ecosystem service–disservice assessments (Escobedo et al., 2011). Studies have demonstrated that inappropriate species selection can increase exposure to allergenic pollen and biogenic volatile organic compounds (BVOCs) (Calfapietra et al., 2013; Escobedo et al., 2023), intensify infrastructure conflicts (Randrup et al., 2001), while inadequate post-planting stewardship can reduce establishment success (Roman et al., 2015), and low diversity can increase vulnerability to pests and diseases (Raupp et al., 2006). Trait-based species selection and species–site matching can improve the likelihood of successful establishment and performance under challenging urban conditions (Sjöman et al., 2018), whereas functional diversity can reduce vulnerability to future disturbances (Paquette et al., 2021). Nevertheless, existing approaches differ in their primary emphasis, decision scale, and treatment of interactions among site suitability, species traits, ecosystem functions, risks, diversity, and long-term performance.

Hence, the transition in urban forestry should be from canopy quantity to urban forest performance. The current article proposes a performance- and risk-aware framework for urban tree species selection. The framework treats species selection as a multi-objective decision process in which candidate species are evaluated against local environmental constraints, desired ecosystem functions, avoidance of potential disservices, city-scale diversity, and long-term performance. The framework is organized around

three propositions for new urban forestry: (i) urban greening should progress from quantification of canopy to qualitative assessment of canopy with long-term performance; (ii) selecting suitable individual trees does not necessarily produce a resilient urban forest, because repeated selection of locally optimal species may promote taxonomic and functional homogenization at the city scale; and (iii) tree selection should be embedded within an adaptive management process that monitors performance and emerging risks. The contribution is not the individual consideration of these established concepts, but their integration into an explicit sequential and adaptive decision process. The framework links site feasibility and candidate-species screening with functional priorities and context-dependent risk and disservice evaluation, followed by a city-scale check of diversity and shared vulnerability. Subsequent monitoring of performance and emerging risks provides feedback for management, replacement where necessary, and future species-selection decisions, thereby linking individual planting choices with longer-term urban forest outcomes.

Why does tree selection matter?

Expanding the urban tree canopy does not necessarily result in healthier or more resilient urban forests. The benefits of trees depend on species traits, site conditions, spatial configuration, and interactions with infrastructure and human populations (Kenney et al., 2011; Sharmin et al., 2023). Poor species–site matching can undermine intended ecosystem benefits and increase the likelihood of ecological, public-health, and economic disservices. Highly allergenic species may increase pollen exposure, whereas high-BVOC-emitting species can contribute to ozone formation under polluted conditions (Escobedo et al., 2023; Calfapietra et al., 2013). Similarly, water-demanding plantings may increase irrigation requirements in drought-prone conditions, while dense canopies can restrict ventilation and nocturnal radiative cooling in some configurations (Roman et al., 2021; Wujeska-Klause and Pfautsch, 2020). Moreover, low taxonomic and functional diversity can increase vulnerability to pests, pathogens, and climatic disturbances (Raupp et al., 2006; Paquette et al., 2021). Repeated selection of a limited number of species can homogenize urban forests, whereas poor matching of mature tree dimensions and root architecture with infrastructure can create conflicts (Martin and Conway, 2025; McKinney, 2006; Randrup et al., 2001). Thus, tree selection requires a broader consideration of species suitability, desired functions, potential disservices, city-scale diversity, and long-term performance (Langenheim et al., 2020; Speak et al., 2018).

Performance- and risk-aware framework for urban tree selection

Urban tree selection is commonly guided by species suitability, environmental tolerance, ecosystem service potential, and the right tree–right place principle (Langenheim et al., 2020; Sjöman et al., 2018). Existing approaches (Supplementary Table S1) integrate several of these considerations but differ in emphasis, decision scale, and treatment of interactions among suitability, traits, functions, risks, diversity, and long-term performance. The proposed framework builds on these approaches by organizing them into an explicit decision sequence for new urban plantings. The framework uses five interconnected dimensions: space, traits, function, risk, and time (Figure 1; Supplementary Table S2) within a sequential and adaptive process. Site conditions constrain the candidate pool, traits inform expected performance, functions establish priorities, risks filter unsuitable alternatives, city-scale diversity and vulnerability modify site-level choices, and monitoring informs management, replacement, and future selection. Thus, the framework links each planting decision to long-term urban-forest performance through a sequential adaptive cycle. The framework is conceptual rather than a fully operational decision-support model. Implementation would require locally appropriate species and performance data, quantitative thresholds for exclusion and trade-offs, and context-specific information on site conditions, exposure, and city-scale species composition; these parameters are not prescribed universally because they depend on local conditions and management objectives.

The framework distinguishes exclusionary criteria from comparative preferences. Fundamental site incompatibility, severe infrastructure conflicts, unacceptable health or safety risks, and unsuitable water requirements may exclude candidates from further consideration (Randrup et al., 2001; Escobedo et al., 2023; Roman et al., 2021). Among the feasible candidates, relevant traits, expected ecosystem functions, maintenance requirements, and contributions to urban forest diversity are compared according to site conditions, management objectives, and locally acceptable risks rather than a universal weighting scheme. Therefore, a highly suitable species may be deprioritized when repeated use increases city-scale concentration or shared vulnerability, provided that suitable alternatives meet site requirements and functional objectives. Similarly, conflicting benefits and risks should be considered simultaneously, with the exclusion of species with unacceptable risk.

These five dimensions are functionally interconnected and conceptually distinguishable. Site conditions determine the environmental and exposure in which species perform and interact with infrastructure and urban populations. Traits describe species attributes that influence performance or potential effects, whereas functions refer to the ecosystem outcomes sought from planting. For example, pollution tolerance is a species attribute, whereas pollution removal and air quality improvement are intended ecosystem functions influenced by traits such as leaf area and physiological characteristics. Potential disservices are unintended negative effects associated with species traits and their interaction with site conditions and urban contexts. Risk refers to potential adverse consequences that arise when such hazards or disservices occur under specific exposure and socio-ecological conditions. For example, allergenicity and BVOC emissions are species attributes whose significance as disservices or risks depends on factors such as abundance, environmental conditions, atmospheric interactions, and exposure of susceptible populations.

This pathway operates across three nested decision contexts. The site is the space i.e., the immediate planting location and has environmental and physical constraints, including rooting space, soil, microclimate, water availability, and infrastructure (Randrup et al., 2001; Sjöman et al., 2018). The land-use context is the broader setting in which planting objectives and environmental conditions influence selection, including streets, healthcare facilities, residential areas, schools, and parks (Bourne and Conway, 2014). The city context refers to the cumulative urban tree portfolio, where species composition, functional diversity, and shared vulnerability are considered across planting decisions (Paquette et al., 2021). Site-level choices contribute to land-use assemblages and cumulatively, to the city-wide portfolio, while information on existing diversity and vulnerability can guide subsequent site-level selection. Species and assemblages are considered collectively to identify excessive taxonomic or functional concentration (Raupp et al., 2006; Paquette et al., 2021). A species that performs well at an individual site may therefore be deprioritized when repeated use would increase city-scale vulnerability, provided suitable alternatives are available. The framework emphasizes the consideration of the distribution of species, functions, and vulnerabilities across the urban forest; specific microclimate and shade models can evaluate localized performance (Langenheim et al., 2020), while spatial statistics and mapping methods can be incorporated to identify systemic clustering, hotspots, or unequal distribution among urban populations. Moreover, city-scale assessments should consider community preferences, governance and ownership responsibilities, procurement and nursery availability, maintenance capacity and life-cycle costs, and environmental equity, as these factors can constrain functional priorities, acceptable risks, implementation feasibility, and the distribution of benefits and risks among urban populations.

Future climate, pest pressure, and species performance are uncertain. Hence, selection should consider tolerance across plausible environmental conditions rather than a single expected future state (Tabassum et al., 2023). Monitoring survival, growth, health, and emerging risks can inform management adjustments, species replacement, and subsequent planting decisions (Roman et al., 2015; Edmondson et al., 2025). A hypothetical example illustrates how the sequence can operate in practice: consider drought-prone urban streets with restricted rooting space, where thermal comfort is prioritized. Tree-A has high drought tolerance and cooling potential but is already widely planted across the city, whereas Tree-B has adequate drought tolerance, lower infrastructure conflict, and a distinct functional profile. Although Tree-A performs well at the site scale, Tree-B is preferred because it meets the site requirements while reducing the city-scale concentration. After planting, survival, growth, health, and emerging risks are monitored, and poor performance or unacceptable risks can trigger management adjustments or species replacement, informing subsequent planting decisions.

Conclusion

Urban greening should evolve from increasing tree numbers and canopy cover toward urban forests that provide sustained benefits under changing environmental and societal conditions. The performance- and risk-aware framework for urban tree species selection integrates site conditions, species traits, desired functions, potential risks, city-scale diversity, and temporal performance within an adaptive decision-making process. Applying the framework requires context-specific trait information, assessment of service–disservice trade-offs, and monitoring of the survival, growth, health, emerging risks, and climate adaptability of species. Monitoring can guide management adjustments, species replacement, and future-planting decisions. The framework should complement, rather than substitute, the retention and protection of existing mature trees and the effective long-term management of urban forests. Overall, it connects individual planting choices with land-use priorities, city-scale diversity and vulnerability, and long-term adaptive performance, with the potential for measurements using remote sensing.

Generative AI statement

The author(s) declared that Generative AI was used in the creation of this manuscript. Artificial Intelligence (AI)-based language tools were used solely to paraphrase and improve the English language of the manuscript. Generative AI was used to generate individual illustrative icons incorporated into Figure 1. The overall conceptual framework, decision pathway, figure composition, labels, and integration of visual elements were developed and assembled by the author(s). The author(s) reviewed and edited the AI-assisted language and visual elements and remain fully responsible for the content and presentation of the manuscript and figure.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frsc.2026.1861981/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Daniel Jato-Espino, Valencian International University, Spain

Reviewed by: Osman Zeybek, Bursa Uludag Universitesi, Türkiye