Home > Overview > Key Findings > Key Outcomes > GCE III - Plant Communities

GCE III - Key Finding in 2015

    Local gradients are more important than regional processes in structuring salt marsh plant communities

    Community similarity is thought to decay with distance; however, this view may be complicated by the relative roles of different ecological processes at different geographical scales, and by the compositional perspective (e.g. species, functional group and phylogenetic lineage) used. Kunza and Pennings (2008) found that plant diversity was greater in Texas than in Georgia marshes. Guo et al. (2015) built on these results by examining similarity as a function of distance among samples. Based on studies in other ecosystems, they hypothesized that community turnover in salt marshes would be more rapid at local versus larger geographical scales; and that community turnover patterns would diverge among compositional perspectives, with a greater distance decay at the species level than at the functional or phylogenetic levels. They examined the characteristics of plant community composition at a series of salt marsh sites being studied in the GCE domain in comparison to sites located on the Gulf Coast in TX, and found that there was strong variation in community composition within individual salt marsh sites across elevation; in contrast, community similarity decayed with distance four to five orders of magnitude more slowly across sites within each region. Overall, community dissimilarity of salt marshes was lowest in the GA sites, intermediate in TX, and highest between the two regions (Fig. 1). These results indicate that local gradients are relatively more important than regional processes in structuring coastal salt marsh communities. The results from the three compositional perspectives generally showed similar patterns, suggesting that in ecosystems with low species diversity, functional and phylogenetic approaches may not provide additional insight over a species-based approach.


    Fig. 1 Relationship between plot distance and plot dissimilarity (Bray-Curtis) for each of 59 transects in GA and 49 transects in TX based on species (panels A and D), functional groups (panels B and E), and taxonomic groups (panels C and F). Lines represent linear regression fit to the data for each transect. Histograms of Mantel r values (range 0 to 1) for each transect are shown; significant values are to the right of the dashed lines. From Guo et al. 2015.


This material is based upon work supported by the National Science Foundation under grants OCE-9982133, OCE-0620959, OCE-1237140 and OCE-1832178. Any opinions, findings, conclusions, or recommendations expressed in the material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.