Aims: It is important to understand the effect of landscape context on biological communities to predict how biodiversity will be affected on human-dominated landscapes. While many studies have tested the effects of landscape context on the species richness and composition of vascular plants, few have compared the responses of vascular plants and bryophytes on the same landscape. We sampled non-epiphytic bryophytes and vascular plants in 184 plots to test whether three landscape context factors measured four years or four decades previously could predict bryophyte or vascular plant species richness and composition after accounting for local factors.

Location: Temperate forests and oak savannahs, Vancouver Island, British Columbia, Canada.

Methods: We used model selection and comparisons to test the effects of surrounding road density, total amount of forest, and distance to the nearest forest edge on species richness, species richness of non-disturbance-associated species, and community composition after controlling for important local predictors including substrate availability and topography.

Results: The species richness of non-disturbance-associated vascular plants was lower in plots with greater surrounding historical road density, and perennial stayer bryophyte richness declined with increasing historical road density and lower historical forest amount, suggesting a potential extinction debt. Landscape context significantly affected total species richness and community composition of vascular plants, but not bryophytes.

Conclusion: While bryophytes appear to be less sensitive overall to landscape context than vascular plants, disturbance-intolerant perennial stayer bryophytes may decline in the future in response to the increased road density and loss of forest cover that has occurred over the past four decades.

We sampled 184 20m x 20m relevé plots on the Saanich Peninsula, a region of approximately 330 km² in 2009. In each plot, we identified all vascular plants and estimated the percent cover of each. We also identified all bryophyte species occurring on soil, rock, or decomposing wood, and estimated the percent cover of each. If a species was estimated to cover less than 1% of the plot, we entered “0.5” in the data, and if it covered much less than 1%, we entered “0.1”. We did not survey epiphytic bryophytes on living trees. We also measured the slope and aspect of each plot, and estimated the percent cover of soil, exposed rock, and decomposing wood substrate. We then used a Geographical Information System (GIS) to measure the total amount of natural vegetated area within a 500m buffer of each plot, the total length of roads within the same distance of each plot, and the distance in metres from the plot to the nearest edge between natural and non-natural vegetation.