Long‐term shifts in the patterns and underlying processes of plant associations in Wisconsin forests

Abstract

AbstractAimPlant species co‐occur within communities in response to variation in environmental conditions, limited species dispersal and biotic interactions. We used surveys and resurveys of the same sites of three temperate forest plant communities to study patterns of association between co‐occurring species pairs and to infer how these mechanisms contribute to community assembly. Our goal was to compare these forces among communities occupying more and less disturbed landscapes and examine how these have changed over the last 50 years.LocationWisconsin, USA.MethodsWe resurveyed 266 sites first surveyed in the 1950s to assess the patterns and dynamics of co‐occurrence among understorey plant species in three community types. We then used checkerboard scores, null models and a newly developed framework to infer the mechanisms likely to have driven community assembly. Finally, we compared these across the three communities and two time periods.ResultsSpecies co‐occur less often than expected in all three vegetation types and both periods. We detected high fractions of aggregated and segregated species pairs (up to 14% and 17%, respectively). In the fragmented southern upland forests and central sand plains, both biotic interactions and dispersal limitation may play important roles in community assembly with inferred dispersal limitation becoming more important since the 1950s. In the more continuous and intact northern upland forests, environmental filtering and biotic interactions appear to dominate community dynamics with little change over time. Aggregated and segregated species pairs made a similar contribution to our ability to infer these mechanisms.Main conclusionsEnvironmental filtering, biotic interactions and dispersal limitation all appear to affect plant community structure in both time periods. However, the influence of dispersal limitation seems to be increasing in more fragmented forest landscapes, portending shifts in community composition and dynamics. Because aggregated and segregated species pairs may be shaped by similar processes both can be used to infer processes of community assembly.