Periodicity in Stem Growth and Litterfall in Tidal Freshwater Forested Wetlands: Influence of Salinity and Drought on Nitrogen Recycling

Many tidally influenced freshwater forested wetlands (tidal swamps) along the south Atlantic coast of the USA are currently undergoing dieback and decline. Salinity often drives conversion of tidal swamps to marsh, especially under conditions of regional drought. During this change, alterations in nitrogen (N) uptake from dominant vegetation or timing of N recycling from the canopy during annual litter senescence may help to facilitate marsh encroachment by providing for greater bioavailable N with small increases in salinity. To monitor these changes along with shifts in stand productivity, we established sites along two tidal swamp landscape transects on the lower reaches of the Waccamaw River (South Carolina) and Savannah River (Georgia) representing freshwater (a parts per thousand currency sign0.1 psu), low oligohaline (1.1-1.6 psu), and high oligohaline (2.6-4.1 psu) stands; the latter stands have active marsh encroachment. Aboveground tree productivity was monitored on all sites through monthly litterfall collection and dendrometer band measurements from 2005 to 2009. Litterfall samples were pooled by season and analyzed for total N and carbon (C). On average between the two rivers, freshwater, low oligohaline, and high oligohaline tidal swamps returned 8,126, 3,831, and 1,471 mg N m(-2) year(-1), respectively, to the forest floor through litterfall, with differences related to total litterfall volume rather than foliar N concentrations. High oligohaline sites were most inconsistent in patterns of foliar N concentrations and N loading from the canopy. Leaf N content generally decreased and foliar C/N generally increased with salinization (excepting one site), with all sites being fairly inefficient in resorbing N from leaves prior to senescence. Stands with higher salinity also had greater flood frequency and duration, lower basal area increments, lower tree densities, higher numbers of dead or dying trees, and much reduced leaf litter fall (103 vs. 624 g m(-2) year(-1)) over the five study years. Our data suggest that alternative processes, such as the rate of decomposition and potential for N mineralization, on tidal swamp sites undergoing salinity-induced state change may be more important for controlling N biogeochemical cycling in soils than differences among sites in N loading via litterfall.