Food in a row: urban trees offer valuable floral resources to pollinating insects

Urbanization affects the availability and diversity of floral resources (pollen and/or nectar) for wild pollinating insects. For example, urban green areas are characterized by an abundance of ornamental plant species. Increasingly, trees are planted to improve the aesthetics of urban streets and parks. These urban trees might offer important floral resources to pollinating insects. To examine the suitability of urban trees as resources for pollinating insects, we investigated the chemical composition of pollen and nectar as well as the amount of nectar produced by the nine major insect-pollinated tree species planted in cities of Western Europe, namely Acer pseudoplatanus, Aesculus carnea, A. hippocastanum, Robinia pseudoacacia, Tilia cordata, T. x euchlora, T. x europaea, T. platyphyllos and T. tomentosa. The analyses revealed that globally the Tilia trees provide pollen with lower contents of polypeptides, amino acids and phytosterols compared with the other species. Urban tree flowers offer abundant nectar with relatively high sugar contents (0.16–1.28 mg/flower); sucrose was the predominant sugar in all nectars. The investigated tree species could therefore be considered in future city plantings.     

Street trees and rural conspecifics: Will long-lived trees reach full size in urban conditions?

Urban conditions are known to affect tree growth, but not all trees respond similarly to presumed stress. I test a hypothesis that successional status of hardwood tree species, rather than taxon, will differentially affect tree size relative to age, in forest versus street plantings. In central Ohio, USA, samples (N = 230) representing 15 native tree species were matched for size between rural woodlot and city street-side conditions. Their girth was measured and their age determined by a count of annual growth rings. Age and size data were analyzed by a general linear model. Most urban trees had smaller trunk diameters than rural conspecifics of the same age. However, trees of early and mid-successional ecologies, despite smaller girth, showed no reduction in growth rate over time. Late successional species appeared to be affected by proximity to impervious areas, showing reduced growth rates, and by inference,reduced ultimate size.     

Franklin Park: 150 years of changing design,disturbance, and impact on tree growth

Urban stresses are known to affect street trees growing in spatially restricted conditions, but even in relatively large parks the cumulative effects by construction, pedestrian uses, and impacts from temporary events may have deleterious long-term effects on plant health. In Columbus, Ohio, the 93 ac (42.3 ha) Franklin Park has been subjected to major construction events and site re-configuration about once per generation—at least six times in 150 years. The first recorded shade- and ornamental tree installations occurred between 1895 and 1910. Those trees, as well as those native to the park, were mapped in 1939, 1989, and again in 2000. Testing the hypothesis that cumulative disturbance reduces tree vigor and growth rate even in open-grown conditions, tree diameters were compared over the sixty year period between 1939 and 2000. Soil compaction and percolation were measured on a 50 m grid, and maps were overlain to determine where site disturbance,buildings, utility lines, paths and pavements had occurred since 1860. Only 8% of the park's surface escaped disturbance; undisturbed areas occurred only in isolated patches. For all trees, trunk diameter increase averaged 0.5 cm/year for 61 years. While late successional species (e.g., Quercus, Acer) had growth rates less than expected for open-grown specimens, many early successional species (Celtis, Populus) were less affected. Despite availability of soil nutrients and lack of canopy or understory competition, tree health was generally poor in all size classes and the majority of the trees present in 1939 were in decline by 2000.     

Exotic trees can sustain native birds in urban woodlands

Native landscaping has been proposed as a means of increasing native bird diversity and abundance in urban landscapes. However residents’ preferences for vegetation are such that exotic plants are often preferred over natives. We investigated the extent to which native birds foraged in three common native and three exotic tree species in mixed urban woodland during four seasons. We predicted that native birds would spend more time foraging in native trees, and that food resources provided by deciduous exotic trees would be more seasonal than those provided by non-deciduous natives. Native birds spent a lot of time foraging in two of the native tree species, but very little time in native red beech (Nothofagus fusca). They used exotic oak (Quercus robur) throughout the year, and sycamore (Acer pseudoplatanus) seasonally. Oak and European beech (Fagus sylvatica) were used by the largest number of species overall, because they attracted both native and exotic birds. With the exception of tree fuchsia (Fuschia excorticata), which produces large volumes of nectar followed by fruits, all tree species were sources of invertebrates for insectivorous feeding. Seasonality of use was high only in sycamore, indicating limited support for our second prediction. We show that being native doesn’t necessarily entail being a good food source for native birds, and popular landscaping exotic species, such as oak, provide foraging opportunities across all seasons.     

Species composition and structure of regenerated and remnant forest patches within an urban landscape

Regenerated and remnant forest patches were inventoried in Syracuse, New York, USA to determine differences in structure, species composition, human disturbances, and landscape context. Patches had similar mean stem diameter, total stem density, and total basal areas, but differed with respect to diameter distribution, disturbance regime, landscape context, and occurrence of introduced species. In regenerated patches, 23 introduced species were inventoried and they accounted for 48% of relative density. In remnant patches, only seven introduced species were inventoried and they accounted for 17% of the relative density. Cluster analyses identified two community types for remnant patches—sugar maple and black oak—and three for regenerated patches—sugar maple, Norway maple, and boxelder. For remnant patches, Rhamnus cathartica dominated the small diameter class in the black oak cluster, and Acer saccharum dominated the small diameter class in the sugar maple cluster. For regenerated patches, introduced species—A. platanoides and R. cathartica—dominated the small diameter class in the Norway cluster, and a mixture of native and introduced species—A. negundo, R. cathartica, A. saccharum, and Rhus typhina—dominated the small diameter classes in the sugar maple and boxelder clusters. Functionally, land covers containing remnant and regenerated patches, such as vacant lots and greenspaces, had the highest net rate of carbon sequestration (848.7 mt/ha/yr).     

Plant-mycorrhizal fungi associations along an urbanization gradient: implications for tree seedling survival

Mycorrhizal fungi symbioses can be critical determining if established seedlings survive or not. Currently, in remnant forests, plants and their fungal symbionts are exposed to varied anthropomorphic effects related to urbanization, however, little is known about their impact on this association. We investigated the relationship between mycorrhizal fungi and tree seedling survival along an urbanization gradient. We planted three species of temperate tree seedlings (Acer rubrum, Carya ovata, and Quercus rubra) in three landscapes: urban, suburban, and rural. We measured the percent of root length colonized by mycorrhizal fungi and monitored survival during their first growing season. We analyzed mycorrhizal colonization as a function of landscape type (urban-rural) and additional variables known to contribute to mycorrhizal colonization (phosphorus, nitrogen, initial plant height). We then analyzed seedling survival as a function of the degree of mycorrhizal fungi colonization associated with the landscape gradient and of additional environmental factors (light and soil moisture). Within a species we found no changes in the levels of mycorrhizal fungi colonization across the urbanization gradient. Each species, however, had markedly different levels of colonization. Survival of A. rubrum was independent of mycorrhizal colonization, while C. ovata and Q. rubra had a significant positive response to increased mycorrhizal fungi. These findings highlight the resilience of mycorrhizal communities across the rural-urban gradient typical of this region, but they also underscore the potential sensitivity of some tree species to lower levels of mycorrhizal fungi colonization.     

Nest selection by red-headed woodpeckers across three spatial scales in an urban environment

Forest habitat is important for a variety of woodpecker species, and is under pressure from urbanization. Red-headed Woodpeckers (Melanerpes erythrocephalus; RHWO) were once abundant across Eastern North America, and their populations have been declining since the 1960s. Their distribution encompasses urban centers, and since urban habitats differ from natural forest areas, our goal was to understand RHWO nest-site selection in an urban context. We addressed two main questions 1) what are the characteristics of RHWO nest selection across multiple spatial scales and 2) how do RHWO nest tree characteristics in city parks compare to those in forest preserves? This work was done in Cook County, IL, which includes Chicago, the third-largest city in the USA by population. We examined 34 RHWO nest trees used between 2010 and 2013, their surrounding habitat, and the landscape within a 1 km radius. Used trees and habitats were compared to paired unused trees and habitats, and landscape-scale characteristics were compared to random locations. Advanced decay of the nest tree, low canopy cover and increased presence of fungus on trees in the surrounding habitat were the best predictors of RHWO nesting in the area. Nests were most commonly found in forested areas outside of dense urban areas. However, we did not detect significant differences in the characteristics of the nest trees located in forest preserves and city parks. Our findings are consistent with nest selection studies in rural and natural areas, suggesting that forest habitats in metropolitan landscapes can support RHWO nesting.     

Contrasting strategies for tree species to cope with heat and dry conditions at urban sites

Street trees planted at tree pits with impermeable pavements are exposed to a high level of both, atmospheric and edaphic stress. For many species tree physiology, optimal growth, and consequently long-term vitality may be hampered under these harsh conditions. Therefore, maintaining the vitality and promoting the ecosystem services provided by urban trees, for example aesthetic values and microclimatic benefits, it is important to establish trees with a high tolerance to stressful urban environments. Measurements of leaf gas exchange, leaf water potential, leaf surface temperature and chlorophyll-fluorescence were used for identification of specific response of Corylus corluna L., Ginkgo biloba L., Liriodendron tulipifera L., and Tilia cordata Mill. ‘Greenspire’. The summer months in 2013 with a period of heat and drought over eight weeks only interrupted by two rainy days were appropriate to access tree’s strategies to cope with low water availability. During the measurement period, the influence of high temperatures, high values of vapor pressure deficit (VPD) and low soil water availability on stomatal conductance, water-use efficiency (WUE) and leaf water potential (Ψmin, Ψmax ) was highest for Liriodendron followed by Ginkgo. These species showed high reductions in WUE in case of higher air temperatures and high values of VPD. Results indicate an avoidance strategy under increasing heat and VPD. In contrast, Corylus and Tilia followed a tolerance strategy indicated by decreasing values of Ψmin and lower reductions of WUE. In conclusion, wide species-specific variation in response to high temperatures and high VPD indicates that substantial potential exists in the selection for dry and hot urban places.     

Distribution and activity of bats at local and landscape scales within a rural–urban gradient

We examined the relationship between bat species activity and composition and the extent of forest cover and urbanization in and adjacent to 11 U.S. National Park Service, National Capital Region Parks in Maryland, Virginia, West Virginia, and Washington, D.C., from 2003–2004, using mist nets, harp traps, acoustical detectors, and visual observations in a variety of habitats. Our efforts included 363 trap nights across 74 sites along with acoustical sampling at 362 sites. We captured 383 bats and identified 6,380 echolocation passes of 6 species. Both overall and species-specific activities were affected more by forest fragmentation within parks than by urbanization adjacent to parks. With an ability to exploit anthropogenic structures for day-roosts, big brown bats (Eptesicus fuscus) were the most ubiquitous and probably the most abundant species in NCR Parks, particularly in forested, urban parks. Northern myotis (Myotis septentrionalis), and to a lesser extent, little brown myotis (Myotis lucifugus) were more prevalent in forested, rural parks of the Ridge and Valley and Blue Ridge than in eastern, less forested urban parks of the Piedmont and Coastal Plain physiographic provinces. Retention of larger, residual forest tracts and day-roosting habitat (i.e., trees and snags) would be beneficial to most species, as urban expansion continues throughout the region.     

Tree diversity,distribution, history and change in urban parks: studies in Bangalore,India

Urban parks constitute critical biodiversity hotspots in crowded, concrete-dominated city environments. Despite the importance, they remain little researched. This paper assesses the biodiversity and distribution of trees in urban parks in the southern Indian city of Bangalore. 127 plots were used to survey tree distribution in parks across the city. The distribution is largely dominated by a few common species. The proportion of exotic species was very high, with 77% of trees belonging to introduced species. Park history had an impact on distribution. Old parks had fewer but larger trees, and greater species diversity compared to recently established parks. Old parks also differed in species composition, having a greater proportion of large canopy trees compared to young parks. Examination of size distributions revealed that large canopied species were gradually being phased out, and replaced by narrow and medium sized tree species which are easier to maintain, but which may not provide the same environmental and ecological benefits. Greater attention requires to be paid to the selection of trees in cities, not just with a view to easy maintenance as is currently the case, but to select an appropriate mix of trees that supports biodiversity and maximizes environmental and ecosystem services.     

Park pruning prompts a competitive reversal of an exotic tree, <Emphasis Type="Italic">Ligustrum lucidum</Emphasis>, in urban forests of Japan

Ligustrum lucidum, a tree native to China has naturalized in Japan where it is often planted in parks and along roads. Naturalized individuals in urban woodlands have uncertain competitiveness against native species, particular in light of the frequent pruning undertaken in these woodlands. We compared fitness-related traits of Ligustrum against two co-occurring native species: a pioneer tree (Mallotus japonicus) and an understory tree (Eurya japonica) under pruned and un-pruned settings. A growth experiments where we assessed the leaf and whole plant characteristics of the 3 species under two environments simulating forest edge and forest interior, and two pruning treatments (with or without) was conducted. Under simulated forest edge without pruning, Ligustrum had lower stem growth and A_max (maximum photosynthetic rate) than Mallotus, suggesting that native species can competitively suppress invasive expansion. With pruning, however, Ligustrum resprouted vigorously in both spring and summer with significantly higher apical growth and total shoots than Mallotus and Eurya in forest edge and interior, respectively. Also, with similar A_max between resprouts of Ligustrum and Mallotus, we see a reversal of competitive advantage in favor of Ligustrum in forest edge. In the forest interior, Ligustrum exhibited shade tolerant properties (i.e., A_max, chlorophyll a/b, and leaf density) similar to Eurya but had more resprout shoots, suggesting that the invader also maintains an advantage under low light. The competitiveness of Ligustrum clearly increased with human disturbance in the form of park pruning. Such management practice can inadvertently enhance the invasiveness of exotic species in urban settings.     

The influence of small urban parks characteristics on bird diversity: A case study of Petaling Jaya,Malaysia

The capacity of small urban park to serve as urban habitats are rarely explored. This study analyses the characteristics of small urban parks and their potential to support urban biodiversity and ecological functions. Nine small urban parks were studied in Malaysia in August and September 2014 using the combined field survey method of structured observation and field measurements. The measured variables were divided into three broad categories of physical characteristics, species richness and human factors. Bird species richness and abundance were used as the indicators for assessing biodiversity. Pearson correlations and multiple regressions were conducted to analyse the relationships between variables and to identify which variables had a significant effect on bird species richness and abundance. The results demonstrated that park area and vegetation variables ( e.g. the percentage of tree canopy cover, open grass/ground, native-exotic plants) are the important predictors of bird species richness and abundance. The percentage of canopy covers (negative relation) and park area (positive relation) are the best predictors of bird species richness in small urban parks. Meanwhile, the best predictors for bird abundance are the percentage of canopy covers (negative relation) and native vegetation species (positive relation). Human activities and park surroundings have a marginal effect on the presence of bird species in small parks. Based on the findings, we provide two general recommendations that could probably increase bird diversity in small urban parks: (1) the park development and management plan should incorporate a social-ecological approach that can benefit both city-dwellers and bird species, and (2) findings from the study should be used to rethink the planting design and composition of especially newly established small urban parks.     

Non-native plants are a seasonal pollen source for native honeybees in suburban ecosystems

In urban and suburban ecosystems, biodiversity can depend on various non-native plant species, including crop plants, garden plants and weeds. Non-native plants may help to maintain biodiversity by providing a source of forage for pollinators in these ecosystems. However, the contribution of plants in urban and agricultural areas to ecosystem services has often been underestimated in biodiversity assessments. In this study, we investigated the pollen sources of native honeybees (Apis cerana) in an arboretum containing native trees and urban and agricultural plants in a suburban landscape. We surveyed the flowering tree species planted inside the arboretum, which were potential pollen sources. The number of potential pollen-source species of native trees peaked in June and July and decreased after August. We collected A. cerana pollen balls every month and identified plant species of pollen in the collected pollen balls using DNA barcoding. In total, we identified 29 plant species from A. cerana pollen balls. The probability of A. cerana using pollen from urban and agricultural plants was higher in July and August than in June. A. cerana collected pollen forages from native tree species (53%), but also gathered pollen from crop plants (13%), garden trees (19%) and native and non-native weeds (14%); the predominant pollen sources in September and October were the garden tree Ulmus parvifolia and the non-native weed Solidago altissima. We found that native honeybees used plants from a variety of habitats including non-native plants to compensate for apparent seasonal shortages of native tree sources in suburban ecosystems. Our results highlight the importance of assessments of both positive and negative roles of non-native plants in urbanized ecosystems to improve biodiversity conservation.     

Foliar nitrogen characteristics of four tree species planted in New York City forest restoration sites

Urban forests provide important environmental benefits, leading many municipal governments to initiate citywide tree plantings. However, nutrient cycling in urban ecosystems is difficult to predict, and nitrogen (N) use in urban trees may be quite different from use in rural forests. To gain insight into these biogeochemical and physiological processes, we compared foliar N characteristics of several common northeastern deciduous tree species across four newly planted New York City afforestation sites as well as at the Black Rock Forest (BRF), a rural oak-dominated forest in the Hudson Highlands, New York. Foliage sampled at BRF was consistently depleted in ¹⁵N compared to urban foliage, and Amelanchier canadensis, Nyssa sylvatica, Prunus serotina, and Quercus rubra showed significant variation in foliar nitrogen isotope signatures (δ¹⁵N) among the four urban sites. A. canadensis and P. serotina showed significantly greater ability to assimilate nitrate at BRF compared to urban sites, as measured through nitrate reductase activity (NRA). There were no significant differences in NRA among tree species growing at the four urban sites. Only P. serotina and N. sylvatica showed significant variation in foliar N concentrations (%N) both among urban sites and compared to BRF. The isotopic and %N data suggest greater N availability but less available nitrate at the newly planted urban sites compared to BRF, possibly due to different anthropogenic inputs or higher rates of nitrification and nitrate leaching at the recently planted urban sites compared to likely lower rates of N cycling in the intact rural forest. In addition, the tree species varied in their response to N availability at the urban sites, with potential implications for growth and survival. Understanding N cycling in urban systems and the associated physiological changes in vegetation is critical to a comprehensive evaluation of urban forest restoration, and may have implications for carbon sequestration and water quality issues associated with nitrate export, two important areas of management concern.     

Nest design in a changing world: Great tit <Emphasis Type="Italic">Parus major</Emphasis> nests from a Mediterranean city environment as a case study

Investigations of urbanization effects on birds have focused mainly on breeding traits expressed after the nest-building stage (e.g. first-egg date, clutch size, breeding success, and offspring characteristics). Urban studies largely ignored how and why the aspects of nest building might be associated with the degree of urbanization. As urban environments are expected to present novel environmental changes relative to rural environments, it is important to evaluate how nest-building behavior is impacted by vegetation modifications associated with urbanization. To examine nest design in a Mediterranean city environment, we allowed urban great tits (Parus major) to breed in nest boxes in areas that differed in local vegetation cover. We found that different measures of nest size or mass were not associated with vegetation cover. In particular, nests located adjacent to streets with lower vegetation cover were not smaller or lighter than nests in parks with higher vegetation cover. Nests adjacent to streets contained more pine needles than nests in parks. In addition, in nests adjacent to streets, nests from boxes attached to pine trees contained more pine needles than nests from boxes attached to other trees. We suggest that urban-related alterations in vegetation cover do not directly impose physical limits on nest size in species that are opportunistic in the selection of nesting material. However, nest composition as reflected in the use of pine needles was clearly affected by habitat type and the planted tree species present, which implies that rapid habitat change impacts nest composition. We do not exclude that urbanization might impact other aspects of nest building behaviour not covered in our study (e.g. costs of searching for nest material), and that the strengths of the associations between urbanization and nest structures might differ among study populations or species.     

Effects of ground surface permeability on the growth of urban linden trees

Street trees are an important part of urban vegetation due to their provisioning of different types of ecosystem services such as local climate regulation and contribution to aesthetical and recreational values. In order to provide these services, urban trees need to endure many stress factors not present in natural environments, such as the widespread use of impervious surfaces in the vicinity of street trees. However, few studies have evaluated the effect of this potential stress factor on urban tree growth. The aim of this study was therefore to investigate how ground surface permeability affects stem and current-year shoot growth of linden (Tilia europaea) street trees in Gothenburg, Sweden. We found that a small fraction of permeable ground surface in the vertically projected tree crown area caused lower stem growth and strongly suppressed current-year shoot growth. This finding can guide future city planning, demonstrating that the vitality of street trees is compromised when the permeable surface area in the vicinity of the tree is small.     

Habitat selection by an arboreal lizard in an urban parkland: not just any tree will do

The nocturnal, arboreal Marbled Gecko, Christinus marmoratus, is a common reptile in urban areas in southern Australia. This study compared the demographics and the distribution of geckos occupying different tree species in a modified urban park, the Adelaide Parklands, with those of a population occurring in Ferguson Conservation Park, a nearby native vegetation remnant surrounded by suburbia. The geckos in the modified urban and native remnant parks demonstrated different demographic characteristics. These differences were explained by tree circumference and the amount and type of bark cover on the tree trunk. In the Parkland, geckos were limited by the availability of thick exfoliating bark, its clearly preferred microhabitat as demonstrated by a laboratory choice experiment. The preference for trees with larger circumference, more cover, and thick bark were probably linked to thermoregulation, predator avoidance, and food resources. This study supports the premise that to support populations of urban animals, tree species selection needs careful consideration with regard to its microhabitat capabilities as well as food provision and other resource supports it can provide.     

A review of tree root conflicts with sidewalks,curbs, and roads

Literature relevant to tree root and urban infrastructure conflicts is reviewed. Although tree roots can conflict with many infrastructure elements, sidewalk and curb conflicts are the focus of this review. Construction protocols, urban soils, root growth, and causal factors (soil conditions, limited planting space, tree size, variation in root architecture, management practices, and construction materials) are discussed. Because costs related to sidewalk and curb damage are substantial, a review of research addressing repair, mitigation, prevention, and litigation costs is included. Finally, future research needs are discussed.

Potential for conflicts between trees and sidewalks/curbs is high when one or more of these factors are present: tree species that are large at maturity, fast growing trees, trees planted in restricted soil volumes, shallow top soil (hard-pan underneath top-soil), shallow foundations underneath the sidewalk (limited or no base materials), shallow irrigation, distances between the tree and sidewalk of less than 2.0–3.0 m., trees greater than 15 to 20 years old.

The results of this survey indicate that cities are spending substantial sums of money to address conflicts between street tree roots and infrastructure. It can be inferred that most of these expenditures are spent dealing with problems that already exist. However, this raises the question: How much is being spent now to ensure that conflicts are minimized in the future?

Future research should concentrate on plant factors, site design, and construction of sidewalks and curbs. Also, more knowledge is needed about interactions between root growth and management techniques, such as pruning and irrigation. Finally, there is need for studies that will assist policy-makers to efficiently allocate funds among repair, mitigation, prevention, and legal remedies.

     

Effects of area size,heterogeneity, isolation,and disturbances on urban park avifauna in a highly populated tropical city

We assessed bird diversity and nesting activity in 54 urban parks in the core of a highly populated East-Asian tropical city to investigate the effects of area, isolation, habitat heterogeneity, and human disturbances on avifauna. Tree density and heterogeneity were lower while isolation was greater in small than in large or medium-sized parks. Yet, park features were often interrelated, with isolation variables negatively correlated to each other, heterogeneity positively correlated to area, and habitat variables mostly positively correlated with one another but variously to disturbances; whereas pedestrian densities wee positively correlated with green area proportions (ISO3) but negatively with the distances to large green areas (ISO2). Park size played a primary role by positively affecting species richness and numbers of nesting species, thus contributing to higher total and nesting species heterogeneity, and determining the occurrence of a greatest number of species. Pedestrian density, canopy cover, and ground heterogeneity played secondary roles, whereas isolation appeared less influential at the community level. Omnivores, granivores, and aerial insectivores appeared more abundant but were dominated by several common species. Nesting was restricted to even fewer species. While species varied in the occupancy responses to these factors, tree heterogeneity positively affected the nesting of Zosterops japonicus and cavity-nesting birds, and predator density negatively affected those of Lonchura punctulata and Streptopelia doves. Improved urban land use planning and park management incorporating these effects should be implemented to enhance breeding and avoid compositional homogenization for long-term sustainability of urban avian diversity.     

Rainfall interception of three trees in Oakland,California

A rainfall interception study was conducted in Oakland, California to determine the partitioning of rainfall and the chemical composition of precipitation, throughfall, and stemflow. Rainfall interception measurements were conducted on a gingko (Ginkgo biloba) (13.5 m tall deciduous tree), sweet gum (Liquidambar styraciflua) (8.8 m tall deciduous), and lemon tree (Citrus limon) (2.9 m tall broadleaf evergreen). The lemon, ginkgo, and sweet gum intercepted 27.0%, 25.2% and 14.3% of gross precipitation, respectively. The lemon tree was most effective because it retained its foliage year-round, storing more winter rainfall than the leafless ginkgo and sweet gum trees. Stemflow was more important for the leafless sweet gum. Because of its excurrent growth habit and smooth bark, 4.1% of annual rainfall flowed to the ground as stemflow, compared to less than 2.1% for the lemon and 1.0% for the ginkgo. Water samples were collected from throughfall, stemflow, and a nearby control site to measure concentrations of nutrients and heavy metals. Compared to the control, samples from under trees had higher concentrations of nutrients and metals (e.g., N, P, K, Zn and Cr), indicating that atmospheric deposition to tree crowns was a major source of pollutants. Nutrient and metal concentrations were highest in gingko tree’s throughfall. Its rough stem surfaces and dense branching pattern appeared to trap more pollutions than the other two trees.