Soil microbes play a key role in controlling ecosystem functions and providing ecosystem services. Yet, microbial communities in urban green space soils remain poorly characterized. Here we compared soil microbial communities in 41 urban parks of (i) divergent plant functional types (evergreen trees, deciduous trees and lawn) and (ii) different ages (constructed 10, ∼50 and >100 years ago). These microbial communities were also compared to those in 5 control forests in southern Finland. Our results indicate that, despite frequent disturbances in urban parks, urban soil microbes still followed the classic patterns typical of plant-microbe associations in natural environments: both bacterial and fungal communities in urban parks responded to plant functional groups, but fungi were under tighter control of plants than bacteria. We show that park age shaped the composition of microbial communities, possibly because vegetation in old parks have had a longer time to modify soil properties and microbial communities than in young parks. Furthermore, control forests harboured distinct but less diverse soil microbial communities than urban parks that are under continuous anthropogenic disturbance. Our results highlight the importance of maintaining a diverse portfolio of urban green spaces and plant communities therein to facilitate complex microbial communities and functions in urban systems.

Urban green spaces provide ecosystem properties fundamental to the provision of ecosystem services, such as the sequestration of carbon and nutrients and serving as a reservoir for organic matter. Although, urban vegetation influences soil physico-chemical properties, it remains unknown whether ecosystem properties depend on plant species portfolios. We tested the influence of three common functional plant groups (evergreen trees, deciduous trees, grass/lawn) for their ability to modify soils in parks of various ages under cold climatic conditions in Finland. We hypothesized that (i) plant functional groups affect soils differently resulting in divergent ecosystem properties, and (ii) that these ecosystem properties also depend on park age. We included 41 urban parks of varying ages (10, 50, and >100 years) and additional control forests. Park soils were sampled for physico-chemical parameters up to 50 cm depth. Our data indicate that plant functional groups modify soils differently, especially between the evergreen and lawn treatments at 50 and >100 year old parks. Soils under evergreen trees had the lowest pH and generally the highest percentage organic matter, percentage total carbon and percentage total nitrogen. Soil pH remained the same, whereas concentrations of organic matter, total carbon and total nitrogen declined by depth. Soils in the reference forests had lower pH but higher percentages organic matter, total carbon, and total nitrogen than those in parks. We estimate that old parks with evergreen trees can store 35.5 and 2.3 kg N m −2 -considerably more than in urban soils in warmer climates. Our data suggest that plant-soil interactions in urban parks, in spite of being constructed environments, are surprisingly similar to those in natural forests.

Urban greenspaces provide ecosystem services like more natural ecosystems do. For instance, vegetation modifies soil properties, including pH and soil organic matter content, yet little is known about its effect on metals. We investigated whether the accumulation and mobility of heavy metals, nutrients and carbon is affected by plant functional types (evergreen or deciduous trees, lawns) in urban parks of varying ages in southern Finland. Plant types modified soil physico-chemical parameters differently, resulting in diverging accumulation and mobility of metals and other elements in park soils. However, the effects of plant functional type depended on park age: lawns in parks of ca. 50 y old had the highest contents of Cr, Cu, Fe, Mn, Ni, and Zn, and in these, and older parks (>100 y old), contents of most metals were lowest under evergreen trees. The mobility of metals and other elements was influenced by the amount of water leached through the soils, highlighting the importance of vegetation on hydrology. Soils under evergreen trees in young parks and lawns in intermediately-aged parks were most permeable to water, and thus had high loads of Ca, Cr, Cu, Fe, Ni, tot-P and tot-N. The loads/concentrations of elements in the leachates was not clearly reflected by their content/concentration in the soil, alluding to the storage capacity of these elements in urban park soils. Our results suggest that in urban systems with a high proportion of impermeable surfaces, park soil has the potential to store nutrients and metals and provide an important ecosystem service particularly in polluted cities.

Plant functional type influences the abundance and distribution of soil biota.With time, as root systems develop, such effects become more apparent. The relationship of plant type and time with the structure and abundance of soil microbial and invertebrate communities has been widely investigated in a variety of systems. However, much less is known about long-term soil community dynamics within the context of urban environments. In this study, we investigated how soil microbes, nematodes and earthworms respond to different plant functional types (lawns only and lawns with deciduous or evergreen trees) and park age in 41 urban parks in southern Finland. As nonurban controls we included deciduous and evergreen trees in 5 forest sites. We expected that microbial biomass and the relative abundance of fungi over bacteria would increase with time. We also expected major differences in soil microbial and nematode communities depending on vegetation: we hypothesized that i) the presence of trees, and evergreens in particular, would support a greater abundance of fungi and fungal-feeding nematodes over bacteria and bacterial-feeding nematodes and ii) the fungi to bacteria ratio would be lowest in lawns, with deciduous trees showing intermediate values.In contrast to our predictions, we showed that old deciduous trees, rather than evergreens, supported the highest fungal abundances and fungal-feeding nematodes in the soil. Consistent with our predictions, microbial biomass in urban park soils tended to increase with time, whereas -in contrast to our hypotheses -fungal-feeding nematode abundance declined. Even in the oldest parks included in the current study, microbial biomass estimates never approximated those in the minimally managed natural forests, where biomass estimates were three times higher. Anecic earthworm abundance also increased with time in urban parks, whereas abundances of fungal-feeding, plant-feeding and omnivorous nematodes, as well as those of epigeic and endogeic earthworms remained constant with time and without any distinct differences between urban parks and the control forests. Our findings highlight that although urban park soils harbor diverse soil communities and 2

Ectomycorrhizal (ECM) fungi are important mutualists for the growth and health of most boreal trees. Forest age and its host species composition can impact the composition of ECM fungal communities. Although plentiful empirical data exist for forested environments, the effects of established vegetation and its successional trajectories on ECM fungi in urban greenspaces remain poorly understood. We analyzed ECM fungi in 5 control forests and 41 urban parks of two plant functional groups (conifer and broadleaf trees) and in three age categories (10, ∼50, and >100 years old) in southern Finland. Our results show that although ECM fungal richness was marginally greater in forests than in urban parks, urban parks still hosted rich and diverse ECM fungal communities. ECM fungal community composition differed between the two habitats but was driven by taxon rank order reordering, as key ECM fungal taxa remained largely the same. In parks, the ECM communities differed between conifer and broadleaf trees. The successional trajectories of ECM fungi, as inferred in relation to the time since park construction, differed among the conifers and broadleaf trees: the ECM fungal communities changed over time under the conifers, whereas communities under broadleaf trees provided no evidence for such age-related effects. Our data show that plant-ECM fungus interactions in urban parks, in spite of being constructed environments, are surprisingly similar in richness to those in natural forests. This suggests that the presence of host trees, rather than soil characteristics or even disturbance regime of the system, determine ECM fungal community structure and diversity. In urban environments, soil and trees improve environmental quality and provide essential ecosystem services. ECM fungi enhance plant growth and performance, increasing plant nutrient acquisition and protecting plants against toxic compounds. Recent evidence indicates that soil-inhabiting fungal communities, including ECM and saprotrophic fungi, in urban parks are affected by plant functional type and park age. However, ECM fungal diversity and its responses to urban stress, plant functional type, or park age remain unknown. The significance of our study is in identifying, in greater detail, the responses of ECM fungi in the rhizospheres of conifer and broadleaf trees in urban parks. This will greatly enhance our knowledge of ECM fungal communities under urban stresses, and the findings can be utilized by urban planners to improve urban ecosystem services.