Introduction
Mycorrhizal associations are found in the majority of terrestrial plants (Kivlin, Hawkes, and Treseder 2011; Wang and Qiu 2006) and are increasingly seen as fundamental drivers of plant community composition and ecosystem functioning (Klironomos et al. 2011; Neuenkamp et al. 2018; Wagg et al. 2014). Different mycorrhizal types have in general different nutrient provision traits: arbuscular mycorrhizas (AM) are usually more important in the uptake of inorganic nutrients, thereby improving nutrient and carbon cycling (Phillips, Brzostek, and Midgley 2013), whereas many ectomycorrhizas (EcM) and ericoid mycorrhizas (ErM) can directly free up nutrients locked up in organic matter (Read, Leake, and Perez-moreno 2004; Sulman et al. 2017). Therefore, different ecosystems are dominated by different mycorrhizal types. The dominance of a certain mycorrhizal strategy is determined both by environmental conditions and by mycorrhiza-mediated feedbacks, implying that even relatively minor disturbances could cause lasting changes in mycorrhizal communities (Averill et al. 2022; Carteron, Vellend, and Laliberté 2022). Existing studies in natural (Clavel et al. 2020; Gerz et al. 2019) and agricultural settings (van der Heyde et al. 2017; Schnoor et al. 2011) show a range of responses of mycorrhizal communities to different types of anthropogenic disturbances. However, these studies remain limited to local scales, and all took place in Europe or North America, and thus generalizing their results remains challenging (Öpik et al. 2010; Soudzilovskaia et al. 2017). Here, we explore the impact of anthropogenic disturbances on the dominant mycorrhizal types of mountain plant communities, using a global database from the Mountain Invasion Research Network (MIREN,www.mountaininvasions.org) (Haider et al. 2022).
The MIREN database consists of 2822 surveys of both native and non-native plant diversity and cover for 100 m² plots, both close to and away from 62 mountain roads covering 11 mountain ranges (Haider et al. 2022). Mountains are highly valuable ecosystems, being home to a disproportionately large share of global biodiversity (Rahbek et al. 2019) and providing crucial ecosystem services (Grêt-regamey, Hanna, and Gre 2021), that are under increasing pressures from anthropogenic disturbance (Dainese et al. 2017; Kohler et al. 2010). Mountain roads are ideal systems to study the impact of human-induced disturbances as they have a strong but locally limited effect on their surroundings (Müllerová, Vítková, and Vítek 2011; O’Farrell and Milton 2006; Pollnac et al. 2012) while covering large elevational, and consequently, large climatic gradients. The plant cover data from MIREN was combined with the FungalRoots database (Soudzilovskaia et al. 2020) to estimate the proportions of plant cover associated with each mycorrhizal type in disturbed roadsides and in the adjacent undisturbed vegetation.
Our expectation was that the disturbance caused by roads would cause a shift in types of mycorrhizal associations, as road disturbance is known to alter plant community composition (Lembrechts et al. 2016; McDougall et al. 2018). As roads tend to favour ruderal species, commonly AM or non-mycorrhizal (NM) plants, to the detriment of more perennial and woody plants, more likely to be associated with EcM and ErM fungi, we expected road disturbance to correlate with increased AM and NM representation and decreased EcM and ErM. As a result, the impact of roadsides on mycorrhizal dominance would be larger if the native vegetation is EcM or ErM-dominated, rather than AM-dominated.
As non-native plant species are typically AM or NM (Pringle et al. 2009) and non-native species are favored by disturbance, we hypothesized that high non-native plant species cover would be found in plots and regions with native communities dominated by AM- or NM-associated species. In turn, non-native species presence should be higher along roadsides.
More specifically, we tested the following hypotheses:
1) Road disturbance is associated with changes in the distribution of mycorrhizal types in mountain plant communities at the global scale.
2) The strength and direction of the disturbance effect on mycorrhizal type distribution depends on local environmental conditions and on the dominant mycorrhizal type of the region’s native community.
3) Non-native plant success is highest in AM- and NM-dominated communities and in plots affected by road disturbance.