2018 ESA Annual Meeting (August 5 -- 10)

COS 96-7 - In search of lost species: Limitations to native understory occurrence beneath N-fixing trees

Thursday, August 9, 2018: 10:10 AM
333-334, New Orleans Ernest N. Morial Convention Center
Elizabeth M. August-Schmidt, Ecology, Evolution and Marine Biology, University of California Santa Barbara, Santa Barbara, CA, Carla D'Antonio, Environmental Studies, University of California Santa Barbara, Santa Barbara, CA and Stephanie G. Yelenik, Pacific Island Ecosystems Research Center, U.S. Geological Survey, Hawaii National Park, HI
Background/Question/Methods

Over the past fifty years, seasonally dry woodlands in Hawaii have been severely degraded by pasture grass invasion and subsequent fire, resulting in a shift from tree-dominance to an exotic grass-dominated savanna. The previous canopy dominant, non-N-fixing Metrosideros polymorpha, is almost non-existent in many areas. By contrast, stands of three N-fixing tree species are now common in submontane burned areas. These include native legumes Acacia koa and Sophora chrysophylla and exotic actinorhizal N-fixer, Morella faya. Previous work has demonstrated that understory communities developing beneath these N-fixing species differ from each other, from burned areas without trees and from the unburned ‘reference community’ dominated by Metrosideros. Thus, tree species identity affects understory community assembly, even within the N-fixing functional group. Additionally, many native species are missing from the understory of N-fixers, while exotic species are successful. Here we ask: (1) What lifecycle stages (seed, seedbank, seedling establishment) constrain native species recruitment? To pinpoint constraints, we surveyed the seedbank and natural seedling occurrence in all community types. (2) How does canopy species identity affect seedling success of native woody species? We tested the extent to which N-fixer canopy species suppressed native recruitment via promotion of exotic grasses by outplanting seedlings from six woody species (four native, two exotic) into three treatment types (exotic grasses intact, grasses removed, or grass cleared at planting but allowed to regrow) in replicate sites of each canopy type.

Results/Conclusions

Most native shrub species – particularly those with fleshy fruits – were absent from both the seedbank and from seedling censuses in fire-disturbed areas, suggesting that dispersal is a key limitation that must be overcome to restore this plant community. In our outplanting study, some individuals of all native species survived in each canopy species-treatment combination, but native outplant survival was lower beneath N-fixing trees compared to non-N-fixing site types. The primary mechanism through which N-fixing trees influenced native seedling survival varied by species. For example, Morella faya, a productive dense-canopied species, inhibited native understory via light reduction whereas Sophora chrysophylla, an open-canopied N-fixer with high soil N-availability, acted primarily via promoting understory grasses. Overall, native survival was negatively correlated with soil N-availability, yet plant growth in this region is N-limited, consistent with an indirect effect of N-fixer canopy on understory recruitment via increased exotic grass competition. In contrast, exotic shrub seedlings had high survival under N-fixers.