Fungal communities decline with urbanization-more in air than in soil

Abstract

Increasing evidence suggests that degradation of biodiversity in human populated areas is a threat for the ecosystem processes that are relevant for human well-being. Fungi are a megadiverse kingdom that plays a key role in ecosystem processes and affects human well-being. How urbanization influences fungi has remained poorly understood, partially due to the methodological difficulties in comprehensively surveying fungi. Here we show that both aerial and soil fungal communities are greatly poorer in urban than in natural areas. Strikingly, a fivefold reduction in fungal DNA abundance took place in both air and soil samples already at 1 km scale when crossing the edge from natural to urban habitats. Furthermore, in the air, fungal diversity decreased with urbanization even more than in the soil. This result is counterintuitive as fungal spores are known to disperse over large distances. A large proportion of the fungi detectable in the air are specialized to natural habitats, whereas soil fungal communities comprise a large proportion of habitat generalists. The sensitivity of the aerial fungal community to anthropogenic disturbance makes this method a reliable and efficient bioindicator of ecosystem health in urban areas.

Fig. 1. Study design.

The location of the five Finnish cities are shown in the left-hand panel and the sampling sceme within each of the cities is shown in the right-hand panel, fuksia (respectively green) color illustrating the sampling carried out in urban areas (respectively natural areas). Within each site, three plots were located in natural and other three plots in urban areas, representing the core and edges of both area types. Fungal communities were sampled from the air (24 h sample taken by the cyclone sampler) and from the soil (a mixture of three soil samples). From each plot, three replicate air and three replicate soil samples were taken.

Fig. 2. Fungal communities in air and soil of urban and natural sites.

The upper row of panels: Euler diagrams for numbers of species shared or not shared between (a) air and soil, (b) urban and natural soil, (c) urban and natural air. The lower row of panels: fungal community composition for (d) all data, (e) soil data and (f) air data in the ordination space. The ordinations plots are based on Euclidian distance applied to log-transformed DNA abundance data. Only OTUs present in at least five samples were included in these analyses.

Fig. 3. Fungal taxonomic composition in air and soil of urban and natural sites.

The Krona wheel showing the taxonomic levels to which each of the 79,155 identified fungal species (OTUs) belongs. The coloring shows the taxa that are predominantly (with >10 times higher DNA abundance) found from air or from soil, from natural or from urban habitats, and the combination of these two classifications (see “Methods” for criteria used when performing the classifications). Sector areas are proportional to DNA abundance, the percentage numbers showing the percentage of total DNA abundance belonging to the taxa contained in the sector. This figure shows pooled data from all sample types, whereas sample-type specific versions are shown in Figure. For an interactive version of the Krona wheel that allows detailed examination of each taxonomic level, as well as for the same information as numerical table format, see Supplementary Information.

Fig. 4. Taxonomic composition and DNA abundance of fungi in air and soil along the gradient from urban core areas to natural core areas.

The Krona wheels show the taxonomic composition of the OTUs found from each sample type. The predicted species richness per sample is shown by S, and the predicted amount of DNA per sample is represented by the relative sizes of the Krona wheels. The coloring follows the classification in Fig. 3. As DNA abundance was measured in different units for the soil and the air, the sizes for the Krona wheels are not comparable between these two sample types. For interactive versions of the eight Krona wheels shown in this figure that allow detailed examination of each taxonomic level, see Supplementary Information.