Next Article in Journal
Microbiological and Sensory Characterization of an Artisanal Wine Made from Spondias purpurea L. and Fermented with Native Yeasts in Santa Elena, Ecuador
Previous Article in Journal
Ambrosia Beetles (Coleoptera: Curculionidae: Scolytinae) Attracted to Necrotraps: Insights into Their Diversity in the Sierra Norte De Puebla, Mexico
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Diversity
Volume 18
Issue 5
10.3390/d18050265
diversity-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Multi-Scale Variations in Understory Community Diversity and Their Driving Mechanisms Under Urbanization Pressure: A Case Study of Shanghai, China

by
Kang Xu
1,
Yeqian Chen
1,
Ruisen Lu
2,
Guiwu Zou
3,
Zhi Peng
4 and
Shanshan Xu
5,*
1
Jiangsu Key Laboratory for Recognition and Remediation of Emerging Pollutants in Taihu Basin, School of Environmental Science and Engineering, Wuxi University, Wuxi 214105, China
2
Jiangsu Key Laboratory for Conservation and Utilization of Plant Resources, Institute of Botany, Jiangsu Province and Chinese Academy of Sciences, Nanjing 210014, China
3
Jiangxi Provincial Key Laboratory of Conservation Biology, College of Forestry, Jiangxi Agricultural University, Nanchang 330045, China
4
Department of Silviculture, Shanghai Forestry Station, Shanghai 200072, China
5
Shanghai Natural History Museum (Branch of Shanghai Science and Technology Museum), Shanghai 200127, China
*
Author to whom correspondence should be addressed.
Diversity 2026, 18(5), 265; https://doi.org/10.3390/d18050265
Submission received: 25 February 2026 / Revised: 15 April 2026 / Accepted: 22 April 2026 / Published: 28 April 2026
(This article belongs to the Section Plant Diversity)
Browse Figures
Versions Notes

Abstract

Understory community diversity in urban forests is crucial for maintaining urban ecosystem functions and enhancing urban resilience, but it is threatened by rapid urbanization. Currently, there remains a knowledge gap regarding the multi-scale responses and driving mechanisms of understory community diversity along urbanization gradients, which hinders its scientific conservation and management. This study was conducted in Shanghai, a highly urbanized metropolis in China, at both plot and site scales. A total of 75 plots and 380 quadrats were established across 16 urban forest sites. Five key environmental factors were selected, including distance from the city center, visitor intensity, non-native species richness, overstory coverage, and forest area. Using taxonomic and phylogenetic diversity, regression models and null models were employed to analyze the multi-scale patterns and underlying assembly processes of understory plant communities. The results showed that the effects of environmental factors were scale-dependent, with environmental filtering as the core assembly mechanism. At the plot scale, the distance from the city center exhibited a U-shaped relationship with species richness (p = 0.005), while visitor intensity displayed a unimodal pattern with both species richness (p < 0.001) and Faith’s phylogenetic diversity (PD, p = 0.029). Increased non-native species richness intensified phylogenetic clustering (p < 0.05), and environmental filtering was the dominant process of community assembly. At the site scale, the β-diversity of non-native species drove the increase in phylogenetic the β-diversity of understory communities (p < 0.001); geographical distance had a significant positive effect on βMNTD (p = 0.002); and differences in non-native species could weaken biotic homogenization (p < 0.05). This study clarifies the multi-scale response patterns and driving mechanisms of understory community diversity and structure, providing a scientific basis for optimizing the conservation and management of understory vegetation in urban forests and enhancing urban ecosystem stability. Future work calls for long-term monitoring and broader environmental indicators.

1. Introduction

Urbanization represents a pervasive and accelerating global trend. By 2025, the global urbanization rate had exceeded 81%, defined as the proportion of the population residing in cities and towns [1]. As urbanization continues to transform landscapes and societies, cities have emerged as complex socio-ecological systems that support the basic needs and livelihoods of rapidly growing human populations. Simultaneously, accelerating urbanization has generated a wide range of environmental pressures and challenges [2]. As one of the most significant near-natural ecosystems within urban environments, urban forests constitute a critical component of sustainable urban development, playing an irreplaceable role in enhancing native biodiversity, sequestering carbon, regulating microclimates, removing air pollutants, improving water quality, promoting public health, and enhancing aesthetic value [3]. In recognition of these multifaceted benefits, urban planners and policymakers increasingly prioritize the expansion and effective management of urban forests to enhance urban livability. As urbanization proceeds at an unprecedented pace, the demand for ecosystem services within cities is expected to intensify. Current projections indicate that 8.2–9.3 billion people, accounting for more than 85% of the global population, will reside in urban areas by 2050 [1]. The continued growth of the urban population is therefore expected to place unprecedented pressure on existing green infrastructure, rendering the conservation and sustainable management of urban forests not merely an ecological consideration but a fundamental requirement for urban resilience and human well-being.
While urban forests provide critical ecosystem services, these near-natural ecosystems are simultaneously threatened by the urban environments they sustain. Urbanization-driven processes, including non-native species introduction, habitat loss and fragmentation, microclimatic alteration, and soil compaction, can substantially impair the functioning and service provision of urban forest ecosystems [4]. Although understory vegetation contributes relatively little to total forest biomass, the maintenance of its diversity exerts a disproportionate influence on ecosystem functions and service delivery [5]. Notably, multiple ecosystem functions relevant to human well-being are contingent upon understory plant biodiversity, including phylogenetic and taxonomic diversity, which are closely associated with patterns of human use and accessibility [6]. Furthermore, understory vegetation diversity influences habitat quality for soil insects [7], avifauna [8,9], and soil microbial communities [10]. Urban forests characterized by structurally well-developed and diverse understory vegetation typically exhibit enhanced ecosystem functions and greater service provision [11]. Consequently, increasing research attention has been directed toward understanding the role of understory diversity in urban forest design and identifying the key determinants shaping understory vegetation diversity, particularly those associated with urbanization gradients.
Urbanization profoundly alters the species composition and structural characteristics of urban forest communities, while its impact mechanism on the understory remains unclear. In general, increasing levels of urbanization are associated with declines in native species diversity within urban forest communities [12]. Urban environmental conditions, anthropogenic preferences, and biological invasions are widely regarded as major drivers of species richness decline and biotic homogenization in urban ecosystems [13]. However, the relationship between urbanization intensity and biotic homogenization remains contested. For example, a study conducted in Baltimore demonstrated that urbanization primarily influenced the mean trait values of plant communities across locations, while exerting no systematic effect on intraspecific diversity [14]. Other studies have reported that urbanization-driven species loss does not necessarily generate significant compositional differences between urban and rural areas, nor does it invariably lead to vegetation homogenization [15,16,17].
In addition, limited attention has been devoted to examining understory community metrics along urbanization gradients across multiple spatial scales, which may exist various impacts [18]. Urbanization intensity exerts a deterministic influence on the genetic diversity and community structure of urban understory vegetation across spatial scales, thereby modulating ecosystem functions. At the β-diversity level, a declining trend with increasing urbanization has frequently been reported, whereas natural or semi-natural habitats within urban regions often exhibit elevated β-diversity [19]. At the α-diversity level, six distinct patterns of species richness responses along urbanization gradients have been identified, including no response, negative, punctuated, intermediate, positive, and bimodal patterns [20]. Therefore, resolving these inconsistencies requires case studies that integrate multiple spatial scales and different dimensions of biodiversity.
Moreover, few works focus on taxonomic and phylogenetic diversity, simultaneously. The high number of species in urban forests may be derived from introduced invasive species [21]. By considering evolutionary distances, phylogenetic diversity can help identify the ecological processes driving species coexistence and community assembly [22]. Phylogenetic analysis can reveal mechanisms of biotic homogenization and speciation [23]. Community phylogenetic diversity patterns serve as key evidence for distinguishing the relative importance of environmental filtering and interspecific competition in community assembly. Therefore, research should consider both phylogenetic and taxonomic diversity when focus on urbanization impact on native community structure and functions while few did it.
Consequently, inconsistent relationships between understory biodiversity and urbanization intensity, scale-dependent patterns, and discrepancies between phylogenetic and taxonomic diversity metrics constrain our understanding of understory community structure and functioning under urbanization. Such knowledge gaps also hinder evidence-based management, planning, and conservation of urban forest understory vegetation. This study focuses on Shanghai, a representative highly urbanized global metropolis, to examine patterns of understory communities along an urbanization gradient. The objectives of this study are to (1) characterize the multiscale distribution patterns of understory community diversity along the urbanization gradient; (2) assess the effects of environmental factors on native understory diversity and community structure; and (3) propose conservation and management strategies for urban forest understory communities. Based on previous studies, we proposed the following three hypotheses:
H1:
The taxonomic and phylogenetic diversity of understory communities exhibits similar scale-dependent non-linear responses to urbanization gradients;
H2:
Non-native species richness intensifies environmental filtering of understory communities at the plot scale, leading to phylogenetic clustering, while the β-diversity of non-native species drives the increase in phylogenetic β-diversity of native communities at the site scale and alleviates biotic homogenization;
H3:
Overstory coverage is a main factor in regulating understory communities at the plot scale, while forest area is a main factor at the site scale.
Five environmental factors, including non-native species richness, distance from the city center, visitor intensity, forest area, and overstory coverage, were evaluated as potential drivers of taxonomic and phylogenetic diversity of understory vegetation at both plot and site scales. Regression models and null models were used to test the potential impacts of the environmental factors on community diversity and structure.
This study fills knowledge gaps by jointly examining taxonomic and phylogenetic diversity of understory communities across plot and site scales in a megacity context. We provide empirical evidence for the multi-scale response patterns and driving mechanisms of understory diversity and community structure. The results advance understanding of urban community assembly and provide practical guidance for conserving and managing urban forest understory vegetation to enhance urban ecosystem stability.

2. Materials and Methods

2.1. Study Site and Survey Protocol

The study was conducted in Shanghai (30°40′–31°53′ N, 120°52′–122°12′ E), a metropolitan city located at the estuary of the Yangtze River. Shanghai is one of the most rapidly urbanizing cities in China, with an urbanization rate of approximately 89% and a population density of about 3000 individuals/km2. Accelerated socio-economic development has driven extensive urban expansion, exerting substantial pressure on forest ecosystems within the metropolitan area. Therefore, Shanghai provides a representative case for examining the effects of urbanization on understory vegetation in urban forests.
The study was conducted at both plot and site scales. A total of 16 urban forest sites were selected using simple random sampling from all urban forests documented in the Shanghai Landscaping & City Appearance Almanac, with 75 plots surveyed across these sites in Shanghai (Figure 1). All plots were located within semi-natural forests that were initially planted and subsequently left unmanaged. Shanghai’s urban forests are dominated by semi-natural secondary broad-leaved forests with a typical two-layered structure (an overstory tree/shrub layer and an understory herb layer). These forests are mainly composed of native evergreen and deciduous broad-leaved tree species widely distributed in subtropical eastern China, with a small number of artificial afforestation species interspersed. The dominant overstory species include Cinnamomum camphora (Lauraceae, evergreen tree), Platanus acerifolia (Platanaceae, deciduous tree), Metasequoia glyptostroboides (Cupressaceae, deciduous conifer) and Sassafras tzumu (Lauraceae, deciduous tree), all of which are the most common constructive species in Shanghai’s urban green spaces. All forest sites exhibited a typical two-layered structure consisting of a tree/shrub layer (overstory) and an herbaceous layer (understory). Field surveys were conducted during the growing season (June–September) of 2022. Plots were established as 5 × 5 m (25 m2) squares and were separated by at least 10 m. Within each plot, five 1 × 1 m quadrats were established at the center and four corners. A total of 76 plots and 380 quadrats were surveyed. All understory individuals within each quadrat were recorded. For each quadrat, crown area, height, and abundance of each understory species were measured, respectively. Tree canopy structure governs shading, light availability, air mixing, and evapotranspiration, with canopy cover representing one of the most critical drivers of understory microclimate that strongly regulates the biodiversity and functioning of understory communities [11]. Accordingly, canopy coverage was selected as a key environmental variable in this study. Overstory canopy coverage (%) was defined as the proportion of ground area covered by the vertical projection of branches and foliage above 3 m in height and was calculated as the ratio of total canopy projection area to plot area (25 m2). In addition, site area (ha) was included as one of the environmental variables at the site scale as species–area relationships represent a pattern expected in natural ecosystem. Distance from the city center, non-native species richness and visitor intensity were treated as metrics representing urbanization. Data on site area and visitor volume (person yr−1) were obtained primarily from the Shanghai Landscaping & City Appearance Almanac, supplemented by field surveys to verify the accuracy of the recorded information. Visitor volume was calculated as the mean annual value from 2019 to 2022. Therefore, overstory coverage, site area, visitor volume, distance from the city center, and non-native species richness were considered as primary driving factors in this study. For urban forests that were not open to the public, a visitor volume of 1000 person yr−1 was assigned to maintain consistency in the dataset and avoid missing values in subsequent statistical analyses. Descriptive statistics for all key environmental factors are shown in Table 1.

2.2. Diversity Metrics

In this study, the taxonomic and phylogenetic diversity of native species were quantified at both plot and site scales. At the plot scale, species richness (SR) was used as the measure of taxonomic diversity, whereas Faith’s phylogenetic diversity (PD), mean pairwise phylogenetic distance (MPD), and mean nearest taxon distance (MNTD) were used as metrics of phylogenetic diversity. PD represents the sum of total phylogenetic branch length of species within each plot [24]. MPD is the mean value all intervening branch lengths between two taxa within each plot [25]. Lower MPD values indicate that species within a community are more closely related phylogenetically. MNTD represents the mean phylogenetic distance between each taxon and its nearest relative within a plot [25]. MNTD reflects patterns in recent evolutionary history, as it is derived primarily from terminal branch lengths of the phylogenetic tree [26]. The functions pd, mpd and mntd in R package picante were used to calculate the three metrics of phylogenetic diversity, respectively [27].
At the site scale, β-diversity was calculated based on pairwise-site dissimilarity between communities. Pairwise phylogenetic β-diversity matrices, including β mean pairwise phylogenetic distances (βMPD) and β mean nearest taxon distances (βMNTD), were generated using the functions comdist and comdistnt in R package picante, respectively. Euclidean distance matrices of explanatory variables (overstory coverage, distance from the city center, and non-native species richness) were calculated using the dist function.
The R package taxonstand was used to standardize species names according to The Plant List (https://www.worldfloraonline.org/; [28]). Native and non-native species were classified according to the checklist of naturalized plants in China [29]. The R package V.PhyloMaker was used to bind all the species into the backbone of mega-tree GBOTB.extended [30] and construct the phylogenetic tree of understory vegetation in Shanghai urban forests [31].

2.3. Phylogenetic Community Structure

To evaluate phylogenetic community assembly, the standardized effect sizes (SES) of MPD and MNTD (SES.MPD and SES.MNTD) were calculated using the null model “taxa.labels”. The observed MPD and MNTD values for each plot were compared with those generated from 999 randomized communities produced by the null model. SES values were calculated as follows:
SES.MPD = (MPDobs − MPDnull)/SDMPD
SES.MNTD = (MNTDobs − MNTDnull)/SDMNTD
where MPDobs and MNTDobs is the observed values of each plot, MPDnull and MNTDnull are means of the 999 null communities, and SDMPD and SDMNTD are the standard deviations of the 999 simulated values. Positive SES.MPD or SES.MNTD values indicate phylogenetic overdispersion, whereas negative values indicate phylogenetic clustering relative to the null expectation [26]. At the site scale, the standardized effect sizes of βMPD and βMNTD (SES.βMPD and SES.βMNTD) were calculated using the same procedure (Equations (1) and (2)). Calculations were performed using ses.mpd and ses.mntd functions in R package picante [27]. The function taxaShuffle was used to create 999 null communities for calculating βMPDnull and βMNTDnull. All phylogenetic metrics were abundance-weighted and subsequently used in further analyses.

2.4. Data Analysis

The best-fitting multiple linear regression models and key explanatory variables were selected based on the corrected Akaike Information Criterion (AICc) (Table S1). The top three environmental variables retained in the best-supported models (ΔAICc < 2) were subsequently used to construct single-variable ordinary least squares (OLS) models to examine the relationships between taxonomic and phylogenetic diversity metrics and each explanatory variable. First-, second-, and third-order polynomial regressions were fitted to evaluate linear, quadratic, and cubic relationships between diversity metrics and selected explanatory variables (Table S2). Model selection across combinations of fixed effects was conducted using the function dredge in the R package MuMIn. Model averaging and comparison among polynomial models were performed using the function model.avg in MuMIn. All the analysis process was performed in R version 4.4.2.

3. Results

3.1. Responses of Diversity to Environmental Factors at the Plot Scale

A total of 116 native understory species were recorded, with species richness per plot ranging from 4 to 23 species. At the plot scale, SR showed no significant relationship with increasing non-native species richness (p = 0.838, R2 < 0.01; Figure 2a). Distance from the city center and visitor volume had significant but contrasting effects on SR (p = 0.005, R2 = 0.13; p < 0.001, R2 = 0.19; Figure 2b,c). SR exhibited a U-shaped relationship with distance from the city center and a unimodal pattern with visitor volume. At distances < 30 km, SR averaged approximately 13 species per plot, declined to approximately 11 species at 30–40 km, and then increased slightly at 40–50 km. A similar pattern was observed for PD, with the highest values within 10 km of the city center and the lowest values at approximately 35 km (Figure 2e). However, the relationship between distance and PD was not statistically significant (p = 0.055). Visitor volume was the only significant predictor of PD, exhibiting a similar unimodal pattern to that observed for SR (p = 0.029, R2 = 0.09; Figure 2f).
Non-native species richness and distance from the city center had negative and positive effects on MPD and MNTD, respectively. Linear regression analyses revealed that MPD increased significantly with distance from the urban core, rising by an average of 15.3 million years for every 10 km increase (p = 0.013, R2 = 0.08; Figure 2h). MNTD exhibited a significant linear decline with increasing non-native species richness, decreasing by an average of 12.9 million years per additional non-native species (p = 0.010, R2 = 0.11; Figure 2j). Visitor volume exerted a significant U-shaped effect on MNTD (p = 0.020, R2 = 0.10; Figure 2l).

3.2. Effects of Environmental Factors on Community Structure

SES.MPD values were generally negative (mean = −0.32), indicating prevalent phylogenetic clustering of understory communities across most plots. Regression analyses revealed a significant decline in SES.MPD with increasing non-native species richness (p = 0.025, R2 = 0.09; Figure 3a) and a significant increase in SES.MPD with greater distance from the city center (p = 0.041, R2 = 0.06; Figure 3b). These results indicate that understory communities tend to become more phylogenetically clustered when introducing more non-native species and at a closer distance from the city center. Similar effects were found on SES.MNTD (Figure 3d–f). SES.MNTD declined sharply with increasing non-native species richness (p = 0.004, R2 = 0.09; Figure 3d). No significant relationship was detected between SES.MNTD and distance from the city center (p = 0.185, R2 = 0.05; Figure 3e), as well as the visitor volume (p = 0.661, R2 = 0.01; Figure 3f).

3.3. Responses of β-Diversity to Environmental Factors at the Site Scale

At the site scale, βMPD increased nonlinearly with non-native β-diversity, as indicated by a significant quadratic regression (βMPD = 218 + 0.182x + 0.171x2, where x represents non-native β-diversity; p < 0.001, R2 = 0.18; Figure 4a). Neither geographic distance nor visitor volume significantly affected βMPD of native communities (p ≥ 0.05). Both non-native β-diversity and geographic distance exerted significant positive effects on βMNTD (p < 0.01; Figure 4e,f). Specifically, βMNTD increased by an average of 1.7 million years per unit increase in non-native β-diversity (p = 0.001, R2 = 0.10; Figure 4d) and by 5.2 million years for every 10 km increase in geographic distance (p = 0.002, R2 = 0.08; Figure 4e). The relationship between visitor volume and βMNTD was best described by a significant cubic polynomial regression (βMNTD = 58.8 + 4.17 × 105x + 3.24 × 1011x2 − 5.61 × 10−18x3, where x represents visitor volume; p = 0.008, R2 = 0.10; Figure 4f). The fitted curve exhibited a non-monotonic pattern, initially increasing, then decreasing, and finally increasing again with higher visitor volume.

3.4. Effects of Environmental Factors on Community Structure at the Site Scale

A total of 87.5% SES.βMPD and 75.0% SES.βMNTD values were negative, indicating prevalent phylogenetic clustering among sites. SES.βMPD showed a continuous increase with increasing non-native β-diversity (p < 0.001, R2 = 0.22; Figure 5a). Based on the quadratic regression results (SES.βMPD = −0.865 − 0.0133x + 0.00395x2, where x represents non-native β-diversity), SES.βMPD exceeded zero when non-native β-diversity between two communities surpassed 16. In contrast to the plot-scale results, variation in overstory coverage had a significant negative effect on SES.βMPD (p = 0.010, R2 = 0.06; Figure 5b). Differences in site area had no significant effect on SES.βMPD (p = 0.687, R2 < 0.01; Figure 5c). For SES.βMNTD, non-native β-diversity also exerted a significant positive effect (p = 0.016, R2 = 0.05; Figure 5d), and SES.βMNTD reached zero when non-native β-diversity exceeded 20. In contrast, site area had a significant linear effect on SES.βMNTD (p = 0.012, R2 = 0.05; Figure 5f).

4. Discussion

4.1. The Main Driving Factors on Native Species Diversity

Our results, based on in situ measurements, showed considerable variability, yet statistically significant relationships were evident for several key parameters linking urbanization metrics to forest community attributes. At both plot and site scales, the diversity of native species—both taxonomic and phylogenetic—is primarily influenced by direct human activities, particularly distance from the city center and visitor volume. Unlike natural ecosystems, the area of urban forests does not appear to be a major determinant of species diversity. It is noted that distance from the city center and visitor volume affect native species diversity in distinct ways: the relationship follows a unimodal curve for distance, while it exhibits a U-shaped curve for visitor volume. Moderate levels of disturbance enhance habitat heterogeneity, thereby promoting species diversity [32,33]. High species richness is also related to the quality of urban life [34], i.e., the moderate distance forests. These findings differ from a Shanghai urban–rural transect study, which reported a nonsignificant relationship between native species richness and urbanization [35], possibly due to taxon-specific responses (e.g., annual versus perennial herbs). Urban forests in city centers, characterized by diverse landscapes, experience intensive and highly refined artificial management, leading to structurally differentiated communities designed to support human well-being. In contrast, more distant suburban forests rely on natural processes to maintain diversity. These results are remarkably consistent with previous studies reporting a U-shaped relationship between plant diversity and urbanization [19].
At the site scale, in contrast to plot-scale findings, increases in β-diversity of non-native species emerge as the dominant driver of higher βMPD and βMNTD values. This suggests that greater differences in non-native species richness correspond to higher phylogenetic diversity among native species. Consequently, at the site scale, urbanization, reflected in increased invasive species presence and greater distance from urban centers, significantly reduces biotic homogenization. This finding aligns with previous research on natural or semi-natural urban habitats [36,37] and suggests that management practices of urban forests in Shanghai more closely resemble those applied in near-natural settings.
These results do not support Hypothesis 1, which assumed that the taxonomic and phylogenetic diversity of understory communities would exhibit similar scale-dependent non-linear responses to urbanization gradients. The inconsistent response patterns between plot and site scales fully reflect the strong scale dependence of the effects of urbanization gradients on understory community diversity, which is mainly due to the different ecological processes dominating community assembly at plot and site scales.

4.2. Mechanisms Driving Diversity Patterns

Urbanization is widely recognized as a primary driver of biotic homogenization [38]. Building upon evidence that environmental filtering predominantly shapes understory communities in urban forests, our study further demonstrates the multi-scale regulatory effects of invasive species on community assembly, with these effects exhibiting strong scale dependence. At the local (plot) scale, invasive species intensify environmental filtering. Their establishment often alters soil physicochemical properties, increases litter accumulation, and modifies light availability, thereby exacerbating local environmental selection pressures [39,40,41,42]. Plots with higher invasive species abundance exhibited lower SES.MPD and SES.MNTD values among the remaining native communities, indicating stronger phylogenetic clustering. This suggests that invasion preferentially eliminates native species unable to tolerate novel environmental conditions, thereby enhancing convergence in community phylogenetic structure. At the site scale, the filtering environments created by invasion vary from one plot to another. This heterogeneous environmental filtering drives the phylogenetic differentiation of native species communities in each plot along distinct evolutionary paths. Thus, the different impacts of invasion observed at the two scales support Hypothesis 2.
Beyond the regulatory role of invasive species turnover, environment variables associated with urbanization also independently influenced phylogenetic assembly processes. Distance from city center modulated the strength of environmental filtering: communities in peri-urban forests exhibited SES.MPD values approaching zero (Figure 3b), indicating a shift towards more stochastic assembly and a near-natural state as urbanization pressure relaxes. Forest area enhanced habitat heterogeneity—larger patches encompass greater microhabitat diversity, creating ecological opportunities for the coexistence of phylogenetically distantly related species [43]. Canopy coverage acted as a direct environmental filter regulating understory light availability [11]; importantly, variation in canopy coverage among sites contributed to β-scale community convergence, as similar closed-canopy conditions across different locations select for the same subset of shade-adapted lineages. This pattern does not support Hypothesis 3, which posited overstory coverage and forest area as dominant plot- and site-scale regulators respectively. Neither factor affected plot-scale community structure, with both only exerting significant site-scale impacts on phylogenetic assembly. Their plot-scale effects are likely masked by the overwhelming influence of local human disturbance and non-native species invasion, alongside relatively homogeneous microenvironments within individual plots. Visitor volume exhibited a U-shaped relationship with SES.MPD and SES.MNTD at the plot scale, suggesting that moderate human disturbance functions as the strongest environmental filter, consistently selecting for specific disturbance-adapted lineages, while both near-natural conditions and extreme disturbance may obscure phylogenetic signals—the former through increased stochasticity, the latter by creating such harsh conditions that only a phylogenetically disparate set of highly tolerant species persist [44,45].
Predominantly negative SES values at both plot and site scales indicate that environmental filtering is the principal process structuring understory communities across the study region, which is a widely documented ecological pattern in urban ecosystems around the world [19,46,47]. The effects of invasive species, distance from city center, forest area, canopy coverage, and visitor volume all manifest through their modulation of this fundamental assembly mechanism, either intensifying or relaxing the strength of environmental selection depending on scale and context.

4.3. Implications for Understory Conservation in Urban Forests

Monitoring and controlling invasive species must be a central component of understory vegetation management, as they fundamentally modify the assembly processes of local communities. Developing strategies that simultaneously address invasion control and biodiversity conservation is imperative, particularly regarding non-native species intentionally introduced by humans. Management strategies should not only target the removal of invasive species but also mitigate the environmental filtering effects they reinforce, including alterations in soil chemistry and litter layer structure. Such interventions are crucial for restoring both taxonomic and phylogenetic diversity in understory vegetation.
At the plot scale, regulating overstory canopy coverage provides a direct and controllable means of modulating understory environmental filters. Our results indicate that canopy coverage functions as a strong environmental filter favoring shade-adapted lineages, suggesting that creating a mosaic of light conditions may enhance understory diversity. Targeted thinning or gap creation can alleviate excessive filtering, promoting the coexistence of shade-tolerant and light-demanding lineages. This approach is supported by empirical evidence showing that thinning increases understory biodiversity [48].
Although forest area is influenced by a complex interplay of environmental and economic factors, this study identifies a positive correlation between forest size and phylogenetic β-diversity, emphasizing the critical value of large, contiguous green spaces. Where expanding the spatial footprint of urban green space is unfeasible, conservation efforts should focus on maximizing habitat heterogeneity within existing forests. This can be accomplished by creating and maintaining diverse habitat types—such as wetlands, water bodies, sunlit gaps, and shaded woodlands—to emulate the heterogeneity of large green spaces, thereby supporting more distantly related species with greater phylogenetic divergence [49,50].
Visitor volume exhibited complex with understory communities. Species richness and phylogenetic diversity peaked under intermediate disturbance levels; however, phylogenetic structure (MNTD, SES.MPD) displayed the strongest clustering at these same levels. Moderate visitor pressure increases species numbers; however, these species are predominantly close relatives from a few disturbance-adapted lineages, masking underlying phylogenetic homogenization. Management should therefore focus on moderate-disturbance forests not merely by reducing visitor numbers, but by actively creating heterogeneous microhabitats (e.g., light gaps, soil aeration) to mitigate strong environmental filtering, thereby reducing evolutionary redundancy and promoting genuine phylogenetic diversity.

4.4. Limitations

The study focused on data collected within a single metropolitan region and a single year, thus it did not account for long-term dynamic changes. The process of non-native species invasion relies on long-term monitoring; a more complete understanding of the invasion process is essential for gaining deeper insights into its impacts. This is the main reason for the nonsignificant results for null models (i.e., 95% SES values were between −1.96 and +1.96). Furthermore, additional metrics related to urban microenvironments or green space characteristics should be incorporated into future research, such as soil hydrothermal conditions, the urban heat island effect, and green space connectivity [51,52,53]. This would help clarify the mechanisms of environmental filtering effects and elucidate the pathways and modes through which invasive species influence these environmental filters.

5. Conclusions

This study demonstrates that urbanization profoundly reshapes the phylogenetic assembly of understory communities in urban forests, with effects differing across spatial scales. At the plot (local) scale, native species richness and phylogenetic diversity displayed contrasting non-linear responses to environmental gradients: a U-shaped relationship with distance from the city center and a unimodal response to visitor volume. However, phylogenetic structure (SES.MPD/SES.MNTD) revealed that moderate disturbance, while maximizing species richness, produced the strongest phylogenetic clustering, indicating that high taxonomic diversity under intermediate disturbance obscures underlying evolutionary redundancy. At this local scale, invasive species intensified environmental filtering, thereby promoting further phylogenetic convergence. At the site scale, this pattern was reversed. Dissimilarity among non-native species emerged as the primary driver increasing phylogenetic β-diversity of native communities, counteracting biotic homogenization through differential invasion along environmental gradients. Overstory canopy coverage promoted phylogenetic convergence across sites, whereas larger forest areas fostered divergence by enhancing habitat heterogeneity. These scale-dependent findings have important management implications: (1) targeted invasive species control should mitigate their environmental filtering effects, rather than solely focus on removal; (2) canopy regulation (e.g., creating light gaps) can sustain understory habitat heterogeneity; (3) large green spaces should be prioritized for conservation, whereas smaller parks require active creation of diverse microhabitats to emulate heterogeneity effects; and (4) moderate-disturbance zones represent critical intervention targets—not for merely reducing visitor numbers, but for actively establishing heterogeneous microhabitats to alleviate intense environmental filtering and reduce evolutionary redundancy.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/d18050265/s1. Table S1: Fitting results of all environmental variable combination models based on the corrected Akaike Information Criterion (AICc). Table S2: Screening outcomes of first-, second-, and third-order polynomial regressions generated by the dredge function in the R package MuMIn.

Author Contributions

Conceptualization, K.X., S.X. and Y.C.; methodology, K.X. and Y.C.; software, K.X.; analysis, K.X., Y.C., R.L. and G.Z.; writing—original draft preparation, K.X. and S.X.; writing—review and editing, K.X., Y.C., R.L., G.Z., Z.P. and S.X.; visualization, K.X.; field sampling, K.X., Z.P. and S.X.; project administration, K.X. and S.X.; funding acquisition, K.X. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the National Natural Science Foundation of China (NSFC) (Grant No. 42401326), the Shanghai Science and Technology Museum Research Special Program (SSTM/SOPZY-04-R2KZ-2026030600010) and the Wuxi University Research Start-up Fund for Introduced Talents (2023r031). The APC was funded by the National Natural Science Foundation of China (NSFC) (Grant No. 42401326).

Institutional Review Board Statement

Not applicable.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. United Nations. World Urbanization Prospects 2025: Summary of Results; UN DESA/POP/2025/TR/ NO. 12; United Nations: New York, NY, USA, 2025. [Google Scholar]
  2. Lafortezza, R.; Sanesi, G. Nature-based solutions: Settling the issue of sustainable urbanization. Environ. Res. 2019, 172, 394–398. [Google Scholar] [CrossRef]
  3. Singh, A.K.; Shukla, S.; Singh, M.K.; Singh, A.; Singh, R.; Singh, B.K. Urban Forests: Importance, Challenges and Opportunities. In Urban Forests, Climate Change and Environmental Pollution: Physio-Biochemical and Molecular Perspectives to Enhance Urban Resilience; Singh, H., Ed.; Springer Nature: Cham, Switzerland, 2024; pp. 23–45. [Google Scholar]
  4. Zhang, B.; Ren, Z.; Miao, Z.; Wang, L.; Wang, C.; Zhang, P.; Hong, S.; Wang, X.; Meng, F.; Huang, B. Strong increase in coverage but accelerated fragmentation in China’s urban forests under rapid urbanization. Appl. Geogr. 2025, 185, 103793. [Google Scholar] [CrossRef]
  5. Mata, L.; Andersen, A.N.; Morán-Ordóñez, A.; Hahs, A.K.; Backstrom, A.; Ives, C.D.; Bickel, D.; Duncan, D.; Palma, E.; Thomas, F.; et al. Indigenous plants promote insect biodiversity in urban greenspaces. Ecol. Appl. 2021, 31, e02309. [Google Scholar] [CrossRef] [PubMed]
  6. Haritika; Negi, A.K. The underestimated role of understory vegetation dynamics for forest ecosystem resilience: A review. Plant Ecol. 2025, 226, 763–787. [Google Scholar] [CrossRef]
  7. Biryol, C.; Baldy, V.; Prevosto, B.; Trap, J.; Forey, E.; Perez-Izquierdo, L.; Ballini, C.; Gauquelin, T.; Santonja, M. Influence of forest thinning on the soil fauna: A systematic review of current knowledge and research gaps. Ann. For. Sci. 2026, 83, 6. [Google Scholar] [CrossRef]
  8. Huang, Y.; Zhao, Y.; Li, S.; von Gadow, K. The Effects of habitat area, vegetation structure and insect richness on breeding bird populations in Beijing urban parks. Urban For. Urban Green. 2015, 14, 1027–1039. [Google Scholar] [CrossRef]
  9. Rush, S.A.; Romito, T.; Robison, T.L. Avian diversity in a suburban park system: Current conditions and strategies for dealing with anticipated change. Urban Ecosyst. 2014, 17, 45–60. [Google Scholar] [CrossRef]
  10. Wu, X.; Ouyang, S.; Tan, X.; Bose, A.K.; Cheng, W.; Tie, L. Effects of understory vegetation and climate change on forest litter decomposition: Implications for plant and soil management. Plant Soil 2025, 515, 51–75. [Google Scholar] [CrossRef]
  11. Xu, S.; Xu, K.; Zou, G.; Yan, J.; Peng, Z.; Zhang, W.; Zhang, Y.; Han, Y.; Wang, J.; Chang, J. Density management strategy for overstory and understory of urban woodland based on ecological size-density allometry. Urban For. Urban Green. 2021, 66, 127379. [Google Scholar] [CrossRef]
  12. Trentanovi, G.; von der Lippe, M.; Sitzia, T.; Ziechmann, U.; Kowarik, I.; Cierjacks, A.; Pyšek, P. Biotic homogenization at the community scale: Disentangling the roles of urbanization and plant invasion. Divers. Distrib. 2013, 19, 738–748. [Google Scholar] [CrossRef]
  13. La Sorte, F.A.; McKinney, M.L.; PyŠEk, P. Compositional similarity among urban floras within and across continents: Biogeographical consequences of human-mediated biotic interchange. Glob. Change Biol. 2007, 13, 913–921. [Google Scholar] [CrossRef]
  14. Kim, J.S.; Noto, A.E. Intraspecific diversity of multiple plant species shows no change across an urbanization gradient. Urban Ecosyst. 2024, 28, 61. [Google Scholar] [CrossRef]
  15. du Toit, M.J.; Kotze, D.J.; Cilliers, S.S. Quantifying Long-Term Urban Grassland Dynamics: Biotic Homogenization and Extinction Debts. Sustainability 2020, 12, 1989. [Google Scholar] [CrossRef]
  16. Aronson, M.F.J.; Handel, S.N.; La Puma, I.P.; Clemants, S.E. Urbanization promotes non-native woody species and diverse plant assemblages in the New York metropolitan region. Urban Ecosyst. 2014, 18, 31–45. [Google Scholar] [CrossRef]
  17. Cameron, G.N.; Culley, T.M.; Kolbe, S.E.; Miller, A.I.; Matter, S.F. Effects of urbanization on herbaceous forest vegetation: The relative impacts of soil, geography, forest composition, human access, and an invasive shrub. Urban Ecosyst. 2015, 18, 1051–1069. [Google Scholar] [CrossRef]
  18. Dylewski, Ł.; Banaszak-Cibicka, W.; Maćkowiak, Ł.; Dyderski, M.K. How do urbanization and alien species affect the plant taxonomic, functional, and phylogenetic diversity in different types of urban green areas? Environ. Sci. Pollut. Res. 2023, 30, 92390–92403. [Google Scholar] [CrossRef] [PubMed]
  19. Blouin, D.; Pellerin, S.; Poulin, M. Increase in non-native species richness leads to biotic homogenization in vacant lots of a highly urbanized landscape. Urban Ecosyst. 2019, 22, 879–892. [Google Scholar] [CrossRef]
  20. McDonnell, M.J.; Hahs, A.K. The use of gradient analysis studies in advancing our understanding of the ecology of urbanizing landscapes: Current status and future directions. Landsc. Ecol. 2008, 23, 1143–1155. [Google Scholar] [CrossRef]
  21. Borden, J.B.; Flory, S.L. Urban evolution of invasive species. Front. Ecol. Environ. 2021, 19, 184–191. [Google Scholar] [CrossRef]
  22. Shi, W.; Wang, Y.Q.; Xiang, W.S.; Li, X.K.; Cao, K.F. Environmental filtering and dispersal limitation jointly shaped the taxonomic and phylogenetic beta diversity of natural forests in southern China. Ecol. Evol. 2021, 11, 8783–8794. [Google Scholar] [CrossRef]
  23. Tretyakova, A.S.; Yakimov, B.N.; Kondratkov, P.V.; Grudanov, N.Y.; Cadotte, M.W. Phylogenetic Diversity of Urban Floras in the Central Urals. Front. Ecol. Evol. 2021, 9, 663244. [Google Scholar] [CrossRef]
  24. Faith, D.P. Conservation evaluation and phylogenetic diversity. Biol. Conserv. 1992, 61, 1–10. [Google Scholar] [CrossRef]
  25. Webb, C.O. Exploring the Phylogenetic Structure of Ecological Communities: An Example for Rain Forest Trees. Am. Nat. 2000, 156, 145–155. [Google Scholar] [CrossRef] [PubMed]
  26. Webb, C.O.; Ackerly, D.D.; McPeek, M.A.; Donoghue, M.J. Phylogenies and Community Ecology. Annu. Rev. Ecol. Syst. 2002, 33, 475–505. [Google Scholar] [CrossRef]
  27. Kembel, S.W.; Cowan, P.D.; Helmus, M.R.; Cornwell, W.K.; Morlon, H.; Ackerly, D.D.; Blomberg, S.P.; Webb, C.O. Picante: R tools for integrating phylogenies and ecology. Bioinformatics 2010, 26, 1463–1464. [Google Scholar] [CrossRef]
  28. Cayuela, L.; Granzow-de la Cerda, Í.; Albuquerque, F.S.; Golicher, D.J. taxonstand: An r package for species names standardisation in vegetation databases. Methods Ecol. Evol. 2012, 3, 1078–1083. [Google Scholar] [CrossRef]
  29. Yan, X.; Wang, Z.; Ma, J. The Checklist of the Naturalized Plants in China; Shanghai Scientific & Technical Publishers: Shanghai, China, 2019. [Google Scholar]
  30. Smith, S.A.; Brown, J.W. Constructing a broadly inclusive seed plant phylogeny. Am. J. Bot. 2018, 105, 302–314. [Google Scholar] [CrossRef]
  31. Jin, Y.; Qian, H.V. PhyloMaker: An R package that can generate very large phylogenies for vascular plants. Ecography 2019, 42, 1353–1359. [Google Scholar] [CrossRef]
  32. Abd El-Wahab, R.H. Plant assemblage and diversity variation with human disturbances in coastal habitats of the western Arabian Gulf. J. Arid Land 2016, 8, 787–798. [Google Scholar] [CrossRef]
  33. Biswas, S.R.; Mallik, A.U. Species diversity and functional diversity relationship varies with disturbance intensity. Ecosphere 2011, 2, 1–10. [Google Scholar] [CrossRef]
  34. Jogan, N.; Küzmič, F.; Šilc, U. Urban structure and environment impact plant species richness and floristic composition in a Central European city. Urban Ecosyst. 2021, 25, 149–163. [Google Scholar] [CrossRef]
  35. Wang, M.; Li, J.; Kuang, S.; He, Y.; Chen, G.; Huang, Y.; Song, C.; Anderson, P.; Łowicki, D. Plant Diversity Along the Urban–Rural Gradient and Its Relationship with Urbanization Degree in Shanghai, China. Forests 2020, 11, 171. [Google Scholar] [CrossRef]
  36. Lososová, Z.; Chytrý, M.; Tichý, L.; Danihelka, J.; Fajmon, K.; Hájek, O.; Kintrová, K.; Láníková, D.; Otýpková, Z.; Řehořek, V. Biotic homogenization of Central European urban floras depends on residence time of alien species and habitat types. Biol. Conserv. 2012, 145, 179–184. [Google Scholar] [CrossRef]
  37. Gong, C.; Chen, J.; Yu, S. Biotic homogenization and differentiation of the flora in artificial and near-natural habitats across urban green spaces. Landsc. Urban Plan. 2013, 120, 158–169. [Google Scholar] [CrossRef]
  38. McKinney, M.L. Urbanization as a major cause of biotic homogenization. Biol. Conserv. 2006, 127, 247–260. [Google Scholar] [CrossRef]
  39. Xu, S.; Zhao, Y.; Yan, J.; Peng, Z.; Zhang, W.; Zhang, Y.; Han, Y.; Wang, J.; Chang, J.; Xu, K. Light availability and anthropogenic stress shape plant understory invasions in understory of urban forests: A case study in Shanghai. Biol. Invasions 2023, 25, 3223–3236. [Google Scholar] [CrossRef]
  40. Kaproth, M.A.; Eppinga, M.B.; Molofsky, J. Leaf litter variation influences invasion dynamics in the invasive wetland grass Phalaris arundinacea. Biol. Invasions 2013, 15, 1819–1832. [Google Scholar] [CrossRef]
  41. Lusizi, Z.; Motsi, H.; Nyambo, P.; Elephant, D.E. Black (Acacia mearnsii) and silver wattle (Acacia dealbata) invasive tree species impact on soil physicochemical properties in South Africa: A systematic literature review. Heliyon 2024, 10, e24102. [Google Scholar] [CrossRef]
  42. Le Jeune, E.; Guiller, A.; Spicher, F.; Horen, H. How does the invasion of forests by Rhododendron ponticum disrupt the transformation of carbon in soils? Biol. Invasions 2026, 28, 43. [Google Scholar] [CrossRef]
  43. Gastauer, M.; Mitre, S.K.; Carvalho, C.S.; Trevelin, L.C.; Sarmento, P.S.M.; Meira Neto, J.A.A.; Caldeira, C.F.; Ramos, S.J.; Jaffé, R. Landscape heterogeneity and habitat amount drive plant diversity in Amazonian canga ecosystems. Landsc. Ecol. 2021, 36, 393–406. [Google Scholar] [CrossRef]
  44. Helmus, M.R.; Keller, W.; Paterson, M.J.; Yan, N.D.; Cannon, C.H.; Rusak, J.A. Communities contain closely related species during ecosystem disturbance. Ecol. Lett. 2010, 13, 162–174. [Google Scholar] [CrossRef]
  45. Dinnage, R. Disturbance alters the phylogenetic composition and structure of plant communities in an old field system. PLoS ONE 2009, 4, e7071. [Google Scholar] [CrossRef]
  46. El-Barougy, R.F.; Dakhil, M.A.; Abdelaal, M.; El-Keblawy, A.; Bersier, L.F. Trait-Environment Relationships Reveal the Success of Alien Plants Invasiveness in an Urbanized Landscape. Plants 2021, 10, 1519. [Google Scholar] [CrossRef]
  47. Čeplová, N.; Lososová, Z.; Zelený, D.; Chytrý, M.; Danihelka, J.; Fajmon, K.; Láníková, D.; Preislerová, Z.; Řehořek, V.; Tichý, L. Phylogenetic diversity of central-European urban plant communities: Effects of alien species and habitat types. Preslia 2015, 87, 1–16. [Google Scholar]
  48. Spicer, M.E.; Royo, A.A.; Wenzel, J.W.; Carson, W.P. Understory plant growth forms respond independently to combined natural and anthropogenic disturbances. For. Ecol. Manag. 2023, 543, 121077. [Google Scholar] [CrossRef]
  49. Hakkila, M.; Le Tortorec, E.; Brotons, L.; Rajasarkka, A.; Tornberg, R.; Monkkonen, M. Degradation in landscape matrix has diverse impacts on diversity in protected areas. PLoS ONE 2017, 12, e0184792. [Google Scholar] [CrossRef]
  50. Biswas, S.R.; Mallik, A.U.; Braithwaite, N.T.; Biswas, P.L. Effects of disturbance type and microhabitat on species and functional diversity relationship in stream-bank plant communities. For. Ecol. Manag. 2019, 432, 812–822. [Google Scholar] [CrossRef]
  51. Lepczyk, C.A.; Aronson, M.F.J.; Evans, K.L.; Goddard, M.A.; Lerman, S.B.; MacIvor, J.S. Biodiversity in the City: Fundamental Questions for Understanding the Ecology of Urban Green Spaces for Biodiversity Conservation. Bioscience 2017, 67, 799–807. [Google Scholar] [CrossRef]
  52. Müller, I.B.; Buhk, C.; Lange, D.; Entling, M.H.; Schirmel, J. Contrasting effects of irrigation and fertilization on plant diversity in hay meadows. Basic Appl. Ecol. 2016, 17, 576–585. [Google Scholar] [CrossRef]
  53. Gao, Z.; Song, K.; Pan, Y.; Malkinson, D.; Zhang, X.; Jia, B.; Xia, T.; Guo, X.; Liang, H.; Huang, S.; et al. Drivers of spontaneous plant richness patterns in urban green space within a biodiversity hotspot. Urban For. Urban Green. 2021, 61, 127098. [Google Scholar] [CrossRef]
Diversity 18 00265 g001
Figure 1. Study site locations in Shanghai, China.
Figure 1. Study site locations in Shanghai, China.
Diversity 18 00265 g001
Diversity 18 00265 g002
Figure 2. Relationships between taxonomic and phylogenetic diversity of the understory vegetation and three selected environmental factors including non-native species richness (non_SR), distance from the city center (Distance, km) and visitor volume (persons yr−1). (a–c) native species richness (SR) vs. environmental factors; (d–f) Faith’s phylogenetic diversity (PD) vs. environmental factors; (g–i) mean pairwise phylogenetic distances (MPD) vs. environmental factors; (j–l) mean nearest taxon distances (MNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Figure 2. Relationships between taxonomic and phylogenetic diversity of the understory vegetation and three selected environmental factors including non-native species richness (non_SR), distance from the city center (Distance, km) and visitor volume (persons yr−1). (a–c) native species richness (SR) vs. environmental factors; (d–f) Faith’s phylogenetic diversity (PD) vs. environmental factors; (g–i) mean pairwise phylogenetic distances (MPD) vs. environmental factors; (j–l) mean nearest taxon distances (MNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Diversity 18 00265 g002
Diversity 18 00265 g003
Figure 3. Relationships between standardized phylogenetic diversity of the understory vegetation and three selected environmental factors including non-native species richness (non_SR), distance from the city center (Distance, km) and visitor volume (persons yr−1). (a–c) standardized mean phylogenetic distance vs. environmental factors (SES.MPD); (d–f) standardized mean nearest taxon distances (SES.MNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Figure 3. Relationships between standardized phylogenetic diversity of the understory vegetation and three selected environmental factors including non-native species richness (non_SR), distance from the city center (Distance, km) and visitor volume (persons yr−1). (a–c) standardized mean phylogenetic distance vs. environmental factors (SES.MPD); (d–f) standardized mean nearest taxon distances (SES.MNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Diversity 18 00265 g003
Diversity 18 00265 g004
Figure 4. Relationships between β-diversity of the understory vegetation and three selected pairwise environmental factors including pairwise non-native species richness (non_SRpairwise), pairwise distance from the city center (Distancepairwise, %) and pairwise forest area (Visitorpairwise, persons yr−1). (a–c) β mean pairwise phylogenetic distances (βMPD) vs. environmental factors; (d–f) β mean nearest taxon distances (βMNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Figure 4. Relationships between β-diversity of the understory vegetation and three selected pairwise environmental factors including pairwise non-native species richness (non_SRpairwise), pairwise distance from the city center (Distancepairwise, %) and pairwise forest area (Visitorpairwise, persons yr−1). (a–c) β mean pairwise phylogenetic distances (βMPD) vs. environmental factors; (d–f) β mean nearest taxon distances (βMNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Diversity 18 00265 g004
Diversity 18 00265 g005
Figure 5. Relationships between standardized β phylogenetic diversity of the understory vegetation and three selected pairwise environmental factors including pairwise non-native species richness (non_SRpairwise), pairwise overstory coverage (Coveragepairwise, %) and pairwise forest area (Areapairwise, ha). (a–c) standardized β mean pairwise phylogenetic distances (SES.βMPD) vs. environmental factors; (d–f) standardized β mean nearest taxon distances (SES.βMNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Figure 5. Relationships between standardized β phylogenetic diversity of the understory vegetation and three selected pairwise environmental factors including pairwise non-native species richness (non_SRpairwise), pairwise overstory coverage (Coveragepairwise, %) and pairwise forest area (Areapairwise, ha). (a–c) standardized β mean pairwise phylogenetic distances (SES.βMPD) vs. environmental factors; (d–f) standardized β mean nearest taxon distances (SES.βMNTD) vs. environmental factors. Dashed lines represent nonsignificant regression results (p ≥ 0.05). Solid lines represent significant regression results (p < 0.05) with shaded areas representing the 95% confidence interval.
Diversity 18 00265 g005
Table 1. Descriptive statistics of key environmental factors in study forests.
Table 1. Descriptive statistics of key environmental factors in study forests.
Environmental FactorMeanMedianSDCV (%)QR1QR3
Overstory coverage (%)68.9970.0015.182260.0080.00
Distance from city center (km)26.7025.5917.086411.1438.77
Forest area (ha)55.9632.8561.4511010.3985.25
Visitor intensity (person yr−1)978,35351,9031,310,75213410002,225,397
Non-native species richness2.632.002.05781.004.00
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Xu, K.; Chen, Y.; Lu, R.; Zou, G.; Peng, Z.; Xu, S. Multi-Scale Variations in Understory Community Diversity and Their Driving Mechanisms Under Urbanization Pressure: A Case Study of Shanghai, China. Diversity 2026, 18, 265. https://doi.org/10.3390/d18050265

AMA Style

Xu K, Chen Y, Lu R, Zou G, Peng Z, Xu S. Multi-Scale Variations in Understory Community Diversity and Their Driving Mechanisms Under Urbanization Pressure: A Case Study of Shanghai, China. Diversity. 2026; 18(5):265. https://doi.org/10.3390/d18050265

Chicago/Turabian Style

Xu, Kang, Yeqian Chen, Ruisen Lu, Guiwu Zou, Zhi Peng, and Shanshan Xu. 2026. "Multi-Scale Variations in Understory Community Diversity and Their Driving Mechanisms Under Urbanization Pressure: A Case Study of Shanghai, China" Diversity 18, no. 5: 265. https://doi.org/10.3390/d18050265

APA Style

Xu, K., Chen, Y., Lu, R., Zou, G., Peng, Z., & Xu, S. (2026). Multi-Scale Variations in Understory Community Diversity and Their Driving Mechanisms Under Urbanization Pressure: A Case Study of Shanghai, China. Diversity, 18(5), 265. https://doi.org/10.3390/d18050265

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop