Symmetric assembly and disassembly processes in an ecological network.

The processes whereby ecological networks emerge, persist and decay throughout ecosystem development are largely unknown. Here we study networks of plant and arbuscular mycorrhizal fungal (AMF) communities along a 120 000 year soil chronosequence, as they undergo assembly (progression) and then disassembly (retrogression). We found that network assembly and disassembly were symmetrical, self-reinforcing processes that together were capable of generating key attributes of network architecture. Plant and AMF species that had short indirect paths to others in the community (i.e. high centrality), rather than many direct interaction partners (i.e. high degree), were best able to attract new interaction partners and, in the case of AMF species, also to retain existing interactions with plants during retrogression. We then show using simulations that these non-random patterns of attachment and detachment promote nestedness of the network. These results have implications for predicting extinction sequences, identifying focal points for invasions and suggesting trajectories for restoration.

Symbiotic soil fungi enhance ecosystem resilience to climate change.

Substantial amounts of nutrients are lost from soils through leaching. These losses can be environmentally damaging, causing groundwater eutrophication and also comprise an economic burden in terms of lost agricultural production. More intense precipitation events caused by climate change will likely aggravate this problem. So far it is unresolved to which extent soil biota can make ecosystems more resilient to climate change and reduce nutrient leaching losses when rainfall intensity increases. In this study, we focused on arbuscular mycorrhizal (AM) fungi, common soil fungi that form symbiotic associations with most land plants and which increase plant nutrient uptake. We hypothesized that AM fungi mitigate nutrient losses following intensive precipitation events (higher amount of precipitation and rain events frequency). To test this, we manipulated the presence of AM fungi in model grassland communities subjected to two rainfall scenarios: moderate and high rainfall intensity. The total amount of nutrients lost through leaching increased substantially with higher rainfall intensity. The presence of AM fungi reduced phosphorus losses by 50% under both rainfall scenarios and nitrogen losses by 40% under high rainfall intensity. Thus, the presence of AM fungi enhanced the nutrient interception ability of soils, and AM fungi reduced the nutrient leaching risk when rainfall intensity increases. These findings are especially relevant in areas with high rainfall intensity (e.g., such as the tropics) and for ecosystems that will experience increased rainfall due to climate change. Overall, this work demonstrates that soil biota such as AM fungi can enhance ecosystem resilience and reduce the negative impact of increased precipitation on nutrient losses.

Composition of fungal soil communities varies with plant abundance and geographic origin.

Interactions of belowground fungal communities with exotic and native plant species may be important drivers of plant community structure in invaded grasslands. However, field surveys linking plant community structure with belowground fungal communities are missing. We investigated whether a selected number of abundant and relatively rare plants, either native or exotic, from an old-field site associate with different fungal communities. We also assessed whether these plants showed different symbiotic relationships with soil biota through their roots. We characterized the plant community and collected roots to investigate fungal communities using 454 pyrosequencing and assessed arbuscular mycorrhizal colonization and enemy-induced lesions. Differences in fungal communities were considered based on the assessment of α- and ß diversity depending on plant 'abundance' and 'origin'. Plant abundance and origin determined the fungal community. Fungal richness was higher for native abundant as opposed to relatively rare native plant species. However, this was not observed for exotics of contrasting abundance. Regardless of their origin, ß diversity was higher for rare than for abundant species. Abundant exotics in the community, which happen to be grasses, were the least mycorrhizal whereas rare natives were most susceptible to enemy attack. Our results suggest that compared with exotics, the relative abundance of remnant native plant species in our old-field site is still linked to the structure of belowground fungal communities. In contrast, exotic species may act as a disturbing agent contributing towards the homogenization of soil fungal communities, potentially changing feedback interactions.

Host identity is a dominant driver of mycorrhizal fungal community composition during ecosystem development.

Little is known about the response of arbuscular mycorrhizal fungal communities to ecosystem development. We use a long-term soil chronosequence that includes ecosystem progression and retrogression to quantify the importance of host plant identity as a factor driving fungal community composition during ecosystem development. We identified arbuscular mycorrhizal fungi and plant species from 50 individual roots from each of 10 sites spanning 5-120 000 yr of ecosystem age using terminal restriction fragment length polymorphism (T-RFLP), Sanger sequencing and pyrosequencing. Arbuscular mycorrhizal fungal communities were highly structured by ecosystem age. There was strong niche differentiation, with different groups of operational taxonomic units (OTUs) being characteristic of early succession, ecosystem progression and ecosystem retrogression. Fungal alpha diversity decreased with ecosystem age, whereas beta diversity was high at early stages and lower in subsequent stages. A total of 39% of the variance in fungal communities was explained by host plant and site age, 29% of which was attributed to host and the interaction between host and site (24% and 5%, respectively). The strong response of arbuscular mycorrhizal fungi to ecosystem development appears to be largely driven by plant host identity, supporting the concept that plant and fungal communities are tightly coupled rather than independently responding to habitat.