Stress-resistant trees are more common in urban than rural forests: A case study of Cleveland,Ohio’s natural parks

Natural parks are comprised of preserved forested natural areas that are undergoing natural ecological processes. These areas can offer a refuge for local biodiversity and contribute substantially to ecosystem services in both rural areas with relatively low population densities, as well as high-density urban areas. Forested natural parks located in urban areas should experience more stressful environmental conditions than nearby rural areas, yet we know relatively little about how urbanization impacts tree communities within these important natural habitats. To better understand the impact of urbanization on forests, we investigated the species composition, abundance, and diversity of midstory and canopy trees as well as tree seedlings in urban and rural natural parks in and around Cleveland, Ohio. We found that both urban and rural natural parks have similar tree abundance, but midstory and canopy trees as well as tree seedling communities in the urban natural parks included higher abundances of stress-tolerant species compared to rural parks. In addition, this pattern was driven by changes in native tree species, as we observed low abundance of invasive species. More stress-resistant native species in urban areas include Quercus rubra and Prunus serotina, in contrast to rural natural parks which are dominated by Acer spp. and Fagus grandifolia. Lastly, we show that urban and rural natural parks have similar species diversity within plots, but we found higher variation in community composition among urban natural parks compared to among rural parks. Furthermore, Q. rubra and P. serotina were significantly larger in rural natural parks, indicating that both environmental stress and successional stage could drive compositional differences. Thus, we show that urbanization can have unexpected effects on plant community composition and diversity. Our study refutes the idea that these are degraded habitats, highlights the need to conserve them, and suggests that characterization of local variation in self-assembled urban tree communities will provide the most accurate picture of their management needs and potential ecosystem services.     

Contributions of landscape trees in public housing estates to urban biodiversity in Hong Kong

Public housing estates (PHEs) in Hong Kong, accommodating 3.3 million of the 7-million population, have generous landscape planting in their grounds. The independent tree management regime generates a tree stock that deviates from the general urban-tree population. This study evaluated species composition, floristic diversity, importance value, and spatial distribution of trees in 102 PHEs (half of the total) occupying 8.31 km² (territory land area of 1104 km²), assessed their contribution to urban biodiversity, and developed a species selection strategy to enrich urban biodiversity. 48,823 trees belonged to 232 species, 151 genera and 59 families. Natives contributed 69 species and 10,837 trees. Species profile was skewed toward exotic species and trees. The species were divided into six frequency groups. The signature group had 45 species each with >200 trees. The dominant, common and occasional groups had 20, 26 and 48 species, respectively. 59 species in the rare group and 34 in the solitary group denoted changing and diverse species selections, respectively. Nonmetric Multidimensional Scaling (NMDS) found that species distribution in PHEs was strongly associated with species diversity, estate area and estate age. However, district and region were not correlated with NMDS. Some 98 species were significantly correlated (Spearman) with one or both NMDS axes. Species groups were analyzed to inform a species selection strategy to improve future planting program and enrich urban biodiversity. The methods and findings could be applied to south-China and other cities to rationalize urban-forest programs with the help of objective research data.     

Assessing the species diversity and vulnerability of urban tree populations in the London borough of Westminster

The level of tree species diversity in urban tree populations can have serious implications for urban forest resilience and has a direct impact on ecosystem functioning at the local level. Few studies have measured the relationship between tree species diversity and vulnerability in UK urban forests. This study analysed the species composition, species diversity and the vulnerability to pests and diseases of 10,149 public trees in the London borough of Westminster across three land use types: housing, parks, and highways. Tree species diversity was significantly different across these land use types (Shannon’s diversity index (H) was 2.47 for housing sites, 1.63 for parks and 0.83 for highways) and we found that higher diversity appears to result in reduced vulnerability, evidenced by negative correlations between tree species diversity and susceptibility to pests and diseases. A stronger negative correlation was found between vulnerability and species richness than between vulnerability and Gini-Simpson’s diversity index. Our study reinforces the role of biodiversity indices in establishing and monitoring baseline levels of UK urban tree diversity. Our findings may inform future tree planting projects, help to ensure that development does not negatively affect urban tree diversity and inform proactive strategies for urban trees to contribute to wider biodiversity conservation.     

Defect-disorder and risk assessment of heritage trees in urban Hong Kong

Heritage trees in a city, echoing factors conducive to outstanding performance, deserve special care and conservation. To understand their structural and health conditions in urban Hong Kong, 30 defect-disorder (DD) symptoms (physical and physiological) subsumed under four tree-position groups (soil-root, trunk, branching, and crown-foliage) and tree hazard rating were evaluated. The surveyed 352 trees included 70 species; 14 species with 233 trees were native. More trees had medium height (10–15 m), medium DBH (1–1.5 m) and large crown (>15 m). In ten habitats, public park and garden (PPG) accommodated the most trees, and roadside traffic island (RTI) and public housing estate (PH) had the least. Tree dimensions and tree habitats were significantly associated. The associations between the 2831 DD and tree-position groups, tree habitats and tree hazard rating were analyzed. Fourteen trees from Ficus microcarpa, Ficus virens and Gleditsia fera had high hazard rating, 179 trees from 22 species moderate rating, and 159 trees from 55 species low rating. RTI, roadside tree strip (RTS), roadside tree pit (RTP), roadside planter (RP) and stone wall (SW) had more moderate hazard rating, and PPG, roadside slope (RS) and government, institutional and community land (GIC) more low rating. Redundancy analysis showed that DD were positively correlated with RTS, RTP, RP and SW, but negatively correlated with PPG, RS and GIC (p < 0.05). The DD significantly increased tree hazard rating and failure potential. Future management implications for heritage-tree conservation and enhancement focusing squarely on critical tree defect-disorder in urban Hong Kong were explored, with application to other compact cities.     

Park and neighbourhood environmental characteristics associated with park-based physical activity among children in a high-density city

Constructing and renovating urban parks is an effective strategy for high-density cities to satisfy residents’ needs for recreation in nature. Using Hong Kong as an example of a high-density city, the purpose of this study was to examine the associations of park and neighbourhood environments with park-based physical activity (PA) among children and examine the moderating effect of neighbourhood income in their relationships. A cross-sectional study was conducted in 32 urban parks in Hong Kong. Park-based PA and the park and neighbourhood environments were measured using systematic observation tools and the geographic information system. Neighbourhood income was extracted from the 2011 Hong Kong Population Census data on median household income. A total of 7,753 children were observed using urban parks for PA during the study. Children’s park-based PA was positively associated with the quality of supporting amenities, park safety, neighbourhood walkability, and neighbourhood quality, while the associations of park-based PA with the diversity of active facilities and park size were not significant. In addition, neighbourhood income moderated the associations of park-based PA with the diversity of active facilities, the quality of supporting amenities, neighbourhood walkability, and neighbourhood quality. These findings provide empirical evidence on the moderating role of neighbourhood socioeconomic status on the associations between park and neighbourhood environments and children’s park-based PA in a high-density city. With children’s needs for using parks for PA, our findings are beneficial for urban planners to renovate existing urban parks to enhance their utility for children living in high-density cities.     

Optimization of tree locations to reduce human heat stress in an urban park

Trees provide cooling benefits through shading and evapotranspiration; they are regarded as an important measure in heat-resilient urban planning and policies. Knowing where to plant trees for maximum cooling benefits, given practical and resource constraints, remains a challenge in both practice and research. Literature in the field of tree modeling and location optimization is limited, either by the incompleteness in accounting for tree shading, evapotranspiration, and the modifying effect of wind, or by the slow-running speed of the Computational Fluid Dynamics model, making them less applicable in practice. This paper describes a novel method to search for the optimal locations for trees to maximize their cooling benefits in an urban environment. A rapid simulation model was applied to assess on-site heat stress under the influences of trees, which was evaluated using field measurements conducted under hot, temperate, and cool weather conditions in an urban park in Hong Kong. It was then linked to a genetic algorithm in search of a near-optimal tree layout. The proposed method was tested in the same park, and it can automatically identify locations to plant new trees to minimize heat stress, subject to practical constraints such as avoiding existing buildings and utilities. It can also identify the optimal locations to rearrange the existing 55 trees, hypothetically, which can cool the park by up to 0.3 ℃ in on-site average equivalent temperature compared with the worse scenario. Trees can cool the most if they are concentrated on the leeward side of the park, rather than spread evenly. The proposed method runs significantly faster than existing approaches, and it can inform research and landscape design practices concerning park cooling as a goal.     

Urban parks can maintain minimal resilience for Neotropical bird communities

Birds may use urban parks as shelter and refuge, contributing with numerous ecosystem services upon which humans and other organisms depend on. To safeguard these services, it is important that bird communities of urban environments hold some degree of resilience, which refers to the capacity of a system to absorb disturbances and changes, while maintaining its functions and structures. Here we assessed the resilience of the bird community inhabiting an urban park in the Southeast region of Brazil. We classified birds in feeding guilds and identified discontinuities and aggregations of body masses (i.e., scales) using hierarchical cluster analysis. We then calculated five resilience indices for our urban park and for a preserved continuous forest (reference area): the average richness of functions, diversity of functions, evenness of functions, and redundancy of functions within- and cross-scale. The urban park had less species, lower feeding guild richness, and lower within-scale redundancy than the reference area. However, they had similar proportion of species in each function, diversity of functions, evenness of functions, and cross-scale redundancy. The lower species richness and, consequently, the lack of some species performing some ecological functions may be responsible for the overall lower resilience in the urban park. Our results suggest that the bird community of the urban park is in part resilient, as it maintained many biological functions, indicating some environmental quality despite the high anthropogenic impacts of this area. We believe that urban forest remnants with more complex and diverse vegetation are possibly more likely to maintain higher resilience in the landscape than open field parks or parks with suppressed or altered vegetation. We propose that raising resilience in the urban park would possibly involve increasing vegetation complexity and heterogeneity, which could increase biodiversity in a large scale.     

Twelve-year change in tree diversity and spatial segregation in the Mediterranean city of Santiago,Chile

Tree diversity is one of the most important components of urban ecosystems, because it provides multiple ecological benefits and contributes to human well-being. However, the distribution of urban trees may be spatially segregated and change over time. To provide insights for a better distribution of tree diversity in a socially segregated city, we evaluated spatial segregation in the abundance and diversity of trees by socioeconomic group and their change over a 12-year period in Santiago, Chile. Two hundred vegetation plots were sampled across Santiago in 2002 and 2014. We found that overall abundance and diversity of urban trees for the entire city were stable over 12 years, whereas species richness and abundance of native tree species increased. There was segregation in tree species richness and abundance by socioeconomic group, with wealthier areas having more species and greater abundance of trees (for all tree species and native species) than poorer ones. Tree community composition and structure varied with socioeconomic group, but we found no evidence of increased homogenization of the urban forest in that 12 years. Our findings revealed that although tree diversity and abundance for the entire city did not change in our 12-year period, there were important inequities in abundance and diversity of urban trees by socioeconomic group. Given that 43% of homes in Santiago are in the lower socioeconomic areas, our study highlights the importance of targeting tree planting, maintenance and educational programs in these areas to reduce inequalities in the distribution of trees.     

Diversity and distribution of the urban tree population in ten major Nordic cities

In order to have a healthy and sustainable urban tree population, a high diversity of species and genera is needed. This study examined (1) the diversity and distribution of genera and species of urban trees in the Nordic region; (2) the diversity in different sites of the city, distinguishing between street and park environments; and (3) the presence of native versus non-native tree species in urban environments in the Nordic region. The analysis of tree diversity was based on urban tree databases comprising a total of 190 682 trees in 10 Nordic cities – Aarhus and Copenhagen in Denmark; Espoo, Helsinki, Tampere and Turku in Finland; Gothenburg, Malmo and Stockholm in Sweden; and Oslo in Norway. The tree databases for Copenhagen, Espoo, Helsinki, Stockholm and Tampere only record street trees, while the remaining databases also include park trees. Tilia was the most dominant genus in Arhus, Copenhagen, Espoo, Gothenburg, Helsinki, Oslo and Stockholm, while Sorbus was the most dominant in Malmo and Betula in Tampere and Turku. Tilia × europaea was the most common species, comprising 16.0% of the total number of tree species. There was a higher proportion of species in parks than in street environments. The number of non-native species was higher than the number of native species in both street and park environments. However, the number of individuals belonging to native species was higher than the number of non-native individuals in all cities and environments except park environments in Arhus. The concluding recommendation from this study regarding greater diversity of genera and species is to exploit local experiences of rare species from local urban tree databases. After appropriate evaluation, urban tree planners can evaluate these rare species in larger numbers for e.g. street environments, where the need is greatest.     

Taxonomic diversity,functional diversity and evolutionary uniqueness in bird communities of Beijing's urban parks: Effects of land use and vegetation structure

The importance of biodiversity conservation is well recognized, and the loss of biodiversity is particularly evident in highly urbanized areas. On the other hand, green spaces inside cities, as parks, can provide a resource for maintaining and increasing biodiversity, especially for bird species. However, only a few studies have addressed the effects of vegetation structure and land use composition on different components of biodiversity.Here, we explored the response of bird community composition to environmental differences related to land use composition and vegetation structure in green spaces in the city of Beijing, China. We compared the values of taxonomic diversity, functional diversity and community evolutionary distinctiveness in breeding bird communities, among ten urban parks of the world's third most populous city. Variation partitioning analysis and generalized linear mixed models were used to explore the unique and shared effects of land use composition and vegetation structure on each biodiversity metric.Park size was not associated with the diversity of bird communities in Beijing. Land use composition was the best predictor of change in bird community composition, followed by vegetation structure at ground level and the intersection between land use and vegetation structure at tree level. Water coverage increased bird species richness, while the presence of large trees increased both taxonomic diversity and bird functional richness in urban parks. Finally, the presence of patches of deciduous trees showed a positive effect on the average score of evolutionary distinctiveness of bird communities. In conclusion, we highlight that different elements of the environment are supporting different components of bird community diversity.     

Private trees contribute uniquely to urban forest diversity,structure and service-based traits

The urban forest provides our communities with a host of benefits through the delivery of ecosystem services. To properly quantify and sustain these benefits, we require a strong baseline understanding of forest structure and diversity. To date, fine-scale work considering urban forest diversity and ecosystem services has often been limited to trees on public land, considering only one or two green space types. However, the governance of urban green spaces means tree species composition is influenced by management decisions at various levels, including by institutions, municipalities, and individual landowners responsible for their care. Using a mixed-method approach combining a traditional field-inventory and community science project, we inventoried urban trees in the residential neighbourhood of Notre-Dame-de-Grȃce, Montreal. We assessed how tree diversity, composition and structure varies across multiple green space types in the public and private domain (parks, institutions, street rights of way and private yards) at multiple scales. We assessed how service-based traits – traits capturing aspects of plant form and functions that urban residents find beneficial – differed across green space types, with implications for the distribution of ecosystem services across the urban landscape. Green space types displayed meaningful differences in tree diversity, structure, and service-based traits. For example, the inclusion of private trees contributed an additional 52 species (>30% of total species) not found in the local public tree inventory. Trees on private land also tended to be smaller than those in the public domain. Beyond patterns of tree richness, size, and abundance we also observed differences in the composition of tree species and service-based traits at site-scales, particularly between street rights-of way and private yards. While species composition varied considerably across street blocks, blocks were very similar to one another in terms of mean service-based traits. Contrastingly, while species composition was similar from yard to yard, yards differed significantly in mean service-based trait values. Our work emphasises that public tree inventories are unlikely to be fully representative of urban forest composition, structure, and benefits, with implications for urban forest management at larger spatial scales.     

New morphological features for urban tree species identification using LiDAR point clouds

Urban tree species identification is the basis for studying the urban-environment coordination mechanism at the species level. Although the gradual maturity of remote sensing data and related research including light detection and ranging (LiDAR) provides a good foundation for the realization of this technology, multiple reasons such as cost, data openness, study scope limitations, and weakness of traditional morphological features make such data still challenging to apply to subtropical urban trees with heterogeneous canopy structures and high biodiversity. To address the problem, we developed two large-scale LiDAR morphological features in this research by, 1) modifying the rotate image method based on the axisymmetric structure to make it easier to use, and 2) developing an innovative adaptive ellipsoid method to extract the canopy features of the non-axisymmetric structure effectively. We evaluated the ability of these two morphological features to describe 12 common subtropical urban tree (SUT) species in Hong Kong growing in urban parks and streets, obtaining an accuracy of 88%. And the advantages of the proposed method are demonstrated by comparison with existing LiDAR morphological features and mean decrease accuracy (MDA) analysis. Our results illustrated that the rotate image feature based on the axisymmetric structure did not perform as well as the adaptive ellipsoid feature based on the non-axisymmetric structure in SUT, and the combined application of these two new morphological features got further accuracy improvement. The method proposed in this study had significant advantages in terms of accuracy, the number of species included, and generalisation capability compared to existing studies on the identification of subtropical urban trees.     

Species diversity of urban forests in China

A good knowledge of species diversity is essential for urban forest planning and management. In this study, we analyzed species diversity of urban forests in China using data synthesized through a systematic review. Our analysis showed that 3740 taxa of woody plants at species level and below have been reported in urban forests in 257 cities. Merging to the species level, there were 2640 species, including 1671 trees, 743 shrubs, and 226 lianas. Salix babylonica L. was the most widely distributed urban tree species in China. Overall, native species accounted for 76.02% of the observed species while the rest were exotic species. Inside cities, parks contained more species than other types of land use. Among cities, composition similarity of urban forests decreased as spatial distances among them increased. Besides, there was a latitudinal pattern in compositional similarity of urban forests in China. The relatively low ratio of the number of woody plant species in urban forests to these naturally distributed in China indicates that there is plenty of room for increasing species diversity of urban forests in China. However, cautions must be taken to avoid increasing compositional similarity of urban forests in China at the same time.     

Changes over time in tree cavity availability across urban habitats

In urban ecosystems, tree cavities provide critical habitat for a variety of wildlife, and their occurrence is influenced by tree health, management, and cavity excavators. Changes over time in vegetative structure, human use patterns, and built environment affect the formation and persistence of tree cavities, and these changes may differ in various urban habitats. Trees with some decay are often associated with tree cavities, however, parks and residential habitats which are highly managed often lack highly-decayed trees, and large trees which are dead and damaged are likely to be removed and replaced with saplings. We surveyed changes over seven years (in 2013 and 2020) in the abundance of both excavated woodpecker cavities and decay cavities, in three urban habitats (forest, park, and residential) in the Chicago region, IL, USA. We observed greater stability of cavity abundance in managed park and residential habitats over time. Low numbers of highly-decayed trees in park and residential habitats were associated with reduced excavated cavity presence compared to forests. As expected, in both 2013 and 2020, the probability of cavity presence for both excavated and decay cavities was increased with greater tree size and higher levels of tree decay, though the patterns of this association varied between habitat types and years. The continued replacement and maintenance of existing trees means that managed park and residential habitats were more stable than unmanaged forest remnants, which are vulnerable to large changes in tree characteristics which could foster unpredictable booms or busts in cavity supply. A stable inventory of tree-cavities depends on preserving large trees, and decay of urban trees benefits habitat quality for cavity-nesters. Pruning of branches or removal of dead trees curtails the life-cycle of tree cavities in decayed branches, so that more highly managed habitats contain fewer cavities than the number of trees could potentially support. Cavity abundance could be improved in stable habitats through reduced intervention where safe, allowing cavity development to occur in situ.     

长沙城市公园绿地植物群落基本类型及物种多样性研究

公园是城市园林绿地系统的主体,也是植物群落及其物种多样性最丰富的地方。研究公园植物多样性可为城市生物多样性保护提供依据。根据对长沙市3个公园绿地植物群落的调查,分析和总结了长沙城市公园绿地植物群落组成和基本类型,论述了公园绿地物种多样性特征。     

Effects of urban parks on the local urban thermal environment

Air temperatures in and outside of 60 parks in Taipei city were surveyed to study the effect of different urban parks on their surrounding thermal gradients. Results suggest that the factors governing the temperature of park surroundings are not identical to those of park interiors. Air-temperature gradients surrounding urban parks are influenced by both the horizontal transport of cool or warm air mass above parks and the evapotranspirative air-parcels from trees, creating a cool island larger than the boundaries of cool-island parks, a heat-island larger than the boundaries of strong heat-island parks, and a cool-ring outside weak heat-island parks. Such horizontal air movement is not easily detected using remotely sensed data. During daytime, the thermal environment within a park is dominated by the amount of solar input absorbed by unshaded paved area, which, when strong, can overflow to increase the temperatures of park surroundings; at night, despite park trees causing a warming effect inside parks, park surroundings are cooled by horizontal flow of evapotranspirative air-parcels from park trees. In business and other districts used mostly during daytime, it is recommended that parks and other open spaces be designed with less than 50% paved area and at least 30% trees, shrubs, and other shadings. In residential districts that are used mostly during nighttime, parks and other open spaces are recommended to be designed with more trees. Night irrigation, a measure commonly recommended for the conservation of water, is also recommended to further enhance this nighttime cooling.     

Modelling the spread of tree pests and pathogens in urban forests

Understanding the potential dynamics of tree pests and pathogens is a vital component for creating resilient urban treescapes. Epidemiologically relevant features include differences in environmental stress and tree management between street and garden trees, and variation in the potential for human-mediated spread due to intensity of human activity, traffic flow and buildings. We extend a standard spatially explicit raster-based model for pest and pathogen spread by dividing the urban tree population into roadside trees and park/garden trees. We also distinguish between naturally-driven radial spread of pests and pathogens and human-mediated linear spread along roads. The model behaviour is explored using landscape data for tree locations in an exemplar UK town. Two main sources of landscape data were available: commercially collated aerial data, which have high coverage but no information on species; and, an urban tree inventory, with low, non-random, coverage but with some species data. The data were insufficient to impute a species-specific host landscape accurately; however, by combining the two data sources, and applying either random or Matérn cluster point process driven selection of a subset of all trees, we create two sets of potential host landscapes. We find that combining the two mechanisms of dispersal has a non-additive effect, with the enhanced linear dispersal enabling new foci of infection to be established more rapidly than with radial dispersal alone; and clustering of trees by species slows down the expansion of epidemics when compared with random distribution of tree species within known host locations.     

Detecting,quantifying, and mapping urban trees' structural defects using infrared thermography: Implications for tree risk assessment and management

Urban trees are a fundamental and key component of urban green areas, however, they are subject to several stresses which can compromise their mechanical integrity through the development of defects such as wood decay. In this study, we evaluated the structural health state of trees in four urban parks in the city of Mytilene, Greece, using structural traits of the trees, their trunk surface temperature distribution as recorded using infrared thermography, and spatial statistics both at single tree and at park level. We developed thermal indices by analyzing data from 334 trees belonging to three main species (Robinia pseudoacacia, Morus alba and Melia azedarach). We estimated temperature spatial dependence across each tree trunk using Moran's I index, while statistically significant spatial clusters were assessed using local spatial autocorrelation statistics. Relationships between tree traits, thermal, and spatial indices were established using linear and logistic regression models. Finally, we used the Getis-Ord Gi* statistic for the identification of risky tree hotspots and we applied the kriging geostatistical procedure for mapping of such hotspots. Our results have shown that the thermal and spatial indices can sufficiently predict different types of structural defect, and to identify hotspots of risky trees and their spatial extent. This approach can successfully contribute to tree risk assessment for a more effective urban park management.     

Urban density does not impact tree growth and canopy cover in native species in Melbourne,Australia

Trees provide multiple ecosystem services in urban centers and increases in tree canopy cover is a key strategy for many municipalities. However, urban trees also experience multiple stresses and tree growth can be impacted by urban density and impervious surfaces. We investigated the impact of differences in urban form on tree growth in the City of Merri-bek, a local government area in metropolitan Melbourne, which is the temperate climate zone. Merri-bek has a gradient in population density and urban greenness from north to south, and we hypothesized that tree growth in the southern areas would be lower because trees were more likely to have less access to water with high levels of impervious surfaces. We selected three common native evergreen species, Eucalyptus leucoxylon, Melaleuca linariifolia, and Lophostemon confertus that exhibit differences in climate vulnerability and assessed the tree canopy expansion in four urban density zones in Merri-bek between 2009 and 2020 using aerial image analysis. The differences in urban form did not significantly influence tree canopy growth and all species showed similar canopy expansion rates. However, smaller trees showed a greater relative canopy increase in the ten years, whereas larger trees had a greater absolute canopy growth. Thus, older and larger trees should be protected and maintained to achieve the canopy expansion. Our study indicated that differences in urban form are unlikely to have major impacts on the growth and canopy expansion of well adapted native tree species in open, suburban centers.     

Comparison of avian diversity between urban parks adjacent to mountain and river in the largest megacity Guangzhou,South China

Urban parks can provide a resource for maintaining and increasing bird biodiversity. However, little research has been conducted to show the differences in bird community diversity between parks adjacent to mountains and rivers in the city center of highly urbanized areas. This study analysed whether the bird composition and species diversity differed between parks adjacent to Baiyun Mountain (hillside parks) and Pearl River (riverside parks) in Guangzhou, South China. From January 2018 to December 2020, 137927 birds representing 206 species were recorded. The results indicated that the number of species and individuals of birds were higher in hillside parks than in riverside parks in each season. The number of species and individuals was also higher in hillside parks than in riverside parks in different functional groups according to bird food type, residential type, and ecological type. The number of species shared by the two types of parks was 120. The unique birds' species number in hillside and riverside parks were 67 and 19, respectively. Bird diversity in terms of richness and Shannon index was higher significantly in hillside parks than in riverside parks in summer. Principal component analysis of environmental traits of 18 parks showed that plant traits in parks have relatively high and positive contribution. Correlation analysis indicated that bird diversity was positively affected by water area, the highest altitude, habitats diversity and tree evenness in park, while negatively regulated by building coverage around each park. The top 20 most abundant bird species accounted for 87.5% of the total birds in the study period. Nycticorax nycticorax, Urocissa erythroryncha and Podiceps ruficollis were observed more frequently in hillside parks, while Pycnonotus aurigaster, Passer montanus, Acridotheres cristatellus were observed more frequently in riverside parks. Principal coordinates analysis also indicated that bird communities significantly differed between hillside parks and riverside parks. In conclusion, our findings highlight the importance of conserving mountains and rivers as buffer zones for serious anthropogenic disturbance in the city center. We propose that the conservation of hillside parks is a priority strategy for enhancing urban birds biodiversity.