Spatial and temporal patterns of plant community assembly in urban vacant lots inform a collaborative restoration experiment
While urban ecosystems can host surprisingly high levels of biodiversity, there are still many shifts in the characteristics of the biotic communities found in cities. In particular, urban species communities tend to become more functionally and phylogenetically homogeneous than non-urban communities. This pattern of regional-scale urban biotic homogenization is well established in the literature but very little is known about which urban environmental pressures most strongly drive patterns of community assembly from patch-to-patch within cities, at a scale which is more valuable for informing urban environmental management plans. Our study describes spatial and temporal variation in plant community assembly within a single urban neighborhood in Baltimore, MD. We conducted surveys of unmanaged herbaceous plant communities in vacant lots in Baltimore, MD during the summers of 2012 and 2013. Within each vacant lot, we established plots within areas that were previously part of the building foundation and within areas that were previously a backyard or garden, to explore how spatial variation in land-use legacies impacts contemporary plant communities. We also collected archival documents describing land use history for each vacant lot, and constructed a chronosequence of time since building demolition, ranging from 3-22 years. Additionally, we developed a partnership with the Maryland Department of Corrections, the Maryland Green Prisons Initiative, and through this collaborative project, established herbaceous native plant restorations of 25 vacant lots in the spring of 2014.
We found low beta diversity in communities established within the building foundation, compared to plant communities in remnant backyard or garden areas. We also observed divergence in functional traits between land-use legacy groups. We contend that priority effects, as a result of differential human legacies, maintained compositional divergence between local communities, regardless of similar contemporary structuring processes and lack of dispersal barriers between the legacy groups. We also found that community diversity did not increase over time in the vacant lots. There were, however, functional shifts in plant community composition, primarily related to variation in dispersal vectors. These findings suggest that dispersal limitation, particularly of biotically-vectored species in building foundation areas, plays a strong role in the assembly of urban plant communities. Finally, we discuss the early success of our experimental vacant lot restorations aimed at increasing vacant lot biodiversity and the importance of incorporating ecological theory into urban restoration design and the management of urban green spaces.