Uncovering the structure of foliar parasite metacommunities in California grasslands

Background/Question/Methods

While parasitic microbes have historically comprised an understudied aspect of biodiversity, they are becoming increasingly recognized as important components of ecological communities. Yet, ecology lacks a general framework for understanding patterns of parasite diversity. To test metacommunity theory as a framework for understanding parasite diversity, we developed a field study in highly diverse California grasslands. At its simplest, metacommunity theory posits that local and regional processes jointly influence the diversity of organisms in a given location. Because of the intimate association between parasites and hosts, local processes (i.e., environmental filtering / biotic interactions) are likely to operate within host individuals while regional processes (i.e., dispersal) should operate within host communities.

Using a spatially structured field survey, we investigated patterns of foliar fungal diversity to identify the processes that generate them.  We monitored foliar infections in individual grasses at three spatial scales: within one-meter-squared plots, between three plots at a site, and between 18 sites across three kilometers within the Pepperwood Preserve, CA. We then performed a series of multivariate tests to ask whether patterns of parasite diversity are consistent with species sorting within hosts (local processes) or dispersal limitation across the landscape (regional processes).

Results/Conclusions

In 2013, we performed three visual surveys of foliar fungi infecting all grass species with greater than 5% cover in each plot. We identified fungi based on symptom morphology, including fruiting bodies when present (i.e., to species or morpho-species). Common fungal species observed at the site included rusts such as Puccinia coronata and leaf-spot fungi from Colletotrichum and Stagonospora genera. Across the preserve, parasite infections were common, with up to 70% of hosts infected during a given survey.  Within host individuals, parasite density was low, with mean leaf area damaged per plant between 1% and 3% on average. Among host individuals, there was strong evidence that parasite communities sorted according to host species identity (R²=0.248, p=0.01), indicating that host individuals act as strong filters for parasite recruitment at the host-species level. There was also no evidence of a distance-decay of similarity at any of the scales measured (Mantel’s R =-0.0003, p=0.50). This indicates that dispersal limitation across the landscape may not be important for determining the structure of parasite communities in this system. Together, these patterns of foliar diversity highlight the importance of recruitment limitation within host individuals and indicate a diminished role of dispersal limitation for structuring parasite communities.