Wetlands DOI 10.1007/s13157-016-0769-0
ORIGINAL RESEARCH
Brackish Marsh Plant Community Responses to Regional Precipitation and Relative sea-Level Rise Elizabeth R. Jarrell 1 & Alexander S. Kolker 2,3 & Cassandra Campbell 1 & Michael J. Blum 1
Received: 25 May 2015 / Accepted: 6 April 2016 # Society of Wetland Scientists 2016
Abstract Climate-driven shifts in environmental conditions can transform the structure and function of coastal ecosystems. Here we examine how two back-barrier brackish marshes in Pamlico Sound (North Carolina, USA) responded to changes in precipitation, temperature, and relative sea level and whether local rates of accretion have kept pace with relative sea-level rise. We used the distribution of seeds in sediment cores, coupled with 210Pb-sediment geochronology, to determine patterns of community and ecosystem change over the past century. The chronologies demonstrate that both marshes recently transitioned from communities dominated by Cladium jamaicense, which prefers fresh and brackish settings, to communities dominated by Schoenoplectus americanus, which prefers brackish and saline environments. Multiple regression analysis indicates that community shifts are best explained by relative sea-level rise and regional trends in precipitation. Results also indicate that the marshes are developing an elevation deficit with respect to rising sea level, which likely influenced the conversion from C. jamaicense dominated to S. americanus dominated communities. These findings substantiate a growing body of evidence indicating Electronic supplementary material The online version of this article (doi:10.1007/s13157-016-0769-0) contains supplementary material, which is available to authorized users. * Alexander S. Kolker
[email protected]
1
Department of Ecology and Evolutionary Biology, Tulane University, New Orleans, LA 70118, USA
2
Louisiana Universities Marine Consortium, 8124 Highway 56, Chauvin, LA 70344, USA
3
Department of Earth and Environmental Sciences, Tulane University, New Orleans, LA 70118, USA
that climate-driven shifts in environmental conditions are transforming coastal ecosystems and suggest that brackish intertidal marshes may become increasingly threatened by accelerated sea-level rise and associated environmental changes expected to unfold this century. Keywords Climate change . Coastal wetlands . Schoenoplectus americanus . Cladium jamaicense . Accretion rate
Introduction Coastal ecosystems are subject to a wide range of stressors arising from climate change as well as land use conversion and other anthropogenic activity. Increased rates of relative sea-level rise (RSLR)- particularly when coupled with altered nutrient, sediment, or hydrological budgets– are promoting rapid ecogeomorphic shifts in coastal ecosystems worldwide (Scavia et al. 2002; IPCC 2014). Brackish marshes, typically dominated by plants with narrow tolerance ranges for salinity and hydroperiod, can serve as an effective prism for studying relationships between plant community composition, ecosystem attributes, regional manifestations of climate change, and human alterations of the coastal environment (Brewer and Grace 1990; Craft et al. 2009; Langley and Megonigal 2010; Wieski et al. 2010; Weston et al. 2011). Accretion rates in coastal marshes have historically increased with RSLR, resulting in elevations at or near sea level (Redfield 1972; Wilson et al. 2014), but evidence is mounting that coastal marshes in many regions do not accrete at rates necessary to keep up with accelerating increases in sea level (Donnelly and Bertness 2001; Hartig et al. 2002; Kolker et al. 2010; Kirwan and Megonigal 2013). Marsh elevation deficits have been linked to climate change, subsidence, insufficient
Wetlands
sediment loads, reduced rates of autochthonous sediment production, and increased salinity stress (Orson et al. 1985; Hartig et al. 2002; Reed 2002; Kirwan and Megonigal 2013). These and other factors, like hurricanes and storms (e.g., van de Plassche et al. 2006), can act independently and synergistically to alter the time and depth of flooding, the distribution and productivity of marsh plants, and ultimately, the persistence of coastal wetlands. We examined environmental records preserved in coastal wetland sediments because the sedimentary record is an excellent vantage point to understand how coastal systems will respond to future environmental changes (Donnelly and Bertness 2001; Horton et al. 2009; Kolker et al. 2009; Kemp et al. 2011; IPCC 2013). In particular, we assessed how two North Carolina coastal brackish marshes responded to regional climate and sea-level variability in the previous century by constructing chronologies of plant community composition and sediment accretion using seed bank records of two ecologically contrasting sedges, Cladium jamaicense and Schoenoplectus americanus, and naturally occurring radioisotopes. Both of the plant species studied produce seeds in proportion to abundance (Sherfy and Kirkpatrick 1999; Saunders 2003; Saunders et al. 2006), and their seeds can persist in sediments for centuries to millennia (e.g., Törnqvist et al. 2004). Historic records for three environmental variables that could promote shifts in brackish marsh plant communitiesRSL (relative sea level), precipitation, and temperature- were obtained and compared against seed bank profiles and geochronologies. These comparisons allowed us to test the hypothesis that compositional shifts in plant communities of brackish marshes on the mid-Atlantic coast are driven by regional manifestations of climate change and to infer whether responses occurred at rates comparable to the pace of environmental change.
Methods Study Region Pamlico Sound (Fig. 1) is a back-barrier coastal lagoon located on the Atlantic coast of North Carolina (35.3128° N, 75.9372° W) that is characterized by shallow water depths, negligible slopes, low elevation, and extensive coverage by hydric soils (Moorhead and Brinson 1995). Freshwater is delivered to the Sound from the Albermarle system, which enters from the north, and from the Pamlico and Neuse Rivers, which enter from the west. Average monthly salinity in the Pamlico Sound has ranged from 0 to 17 ppt over a recent 30 year period (1980–2009), with a mean of ~7 ppt (Carpenter and Dubbs 2012). Within the Sound, salinity generally decreases from east to west, and values are typically lowest with increased freshwater inputs during the late winter to early spring (Abbene et al. 2006; Carpenter and Dubbs
Fig. 1 Map of study area. Open triangles represent sample sites and closed stars demonstrate locations where data were collected on precipitation, temperature, and RSL (map adapted from d-maps.com, credit: Daniel Dalet)
2012). Tidal range in the Pamlico Sound varies from ~10 cm at Hatteras to ~100 cm at Beaufort and Morehead City (Fig. 1; Abbene et al. 2006; NOAA 2014). Although inorganic sediment content in the Sound appears to be increasing (Cooper et al. 2004), the small tidal range in the Sound implies relatively small tidal currents and thus limited capacity to delivery and distribute sediment onto marsh surfaces, which suggests that organic production is especially important for maintaining marsh elevation in the face of rising sea level (Moorhead and Brinson 1995). These attributes make marshes in the Pamlico Sound attractive targets for studying the effects of sea-level rise (Pruitt et al. 2010; Kemp et al. 2011). Like estuaries elsewhere, the Pamlico-Albemarle system has experienced substantial environmental changes over the past century. Between 1982 and 2006, urban development increased from ~418 to ~1660 km 2 within the system (Carpenter and Dubbs 2012). Factors associated with increased urban development, such as road and ditch construction, have been linked to nutrient and sediment loading (Cooper et al. 2004; Carpenter and Dubbs 2012). Regional precipitation, temperature, and estuarine salinity have all increased over the past century (Abbene et al. 2006; Carpenter and Dubbs 2012). In recent decades RSLR has also increased, from ~2 to ~5 mm year−1 (Horton et al. 2009; Kemp et al. 2011; NOAA 2014). Concurrently, the North Carolina coast has been experiencing shoreline erosion and net loss of coastal wetlands. The state had 352 km2 of salt and brackish marsh in 1994; by 2001, 24 km2 had converted from marsh to open
Wetlands
water and 2 km2 had transitioned to upland elevations, resulting in a total loss of 7.3 % of salt and brackish marsh in less than a decade (Carle 2011). Comparisons of historical aerial photographs to 1998 Digital Orthophoto Quarter Quadrangles indicate that shoreline erosion rates have varied from 0 to ~1.5 m year−1 in Pamlico Sound (Riggs and Ames 2003; Carpenter and Dubbs 2012). Study System The spatial variation of salinity regimes in an estuarine system like Pamlico Sound can create distinct zones of plant communities (e.g., Odum 1998). In this region, intertidal marshes in the areas of lowest elevation and highest salinity are dominated by Spartina alterniflora. Higher elevation marshes experiencing brackish conditions tend to be dominated by Juncus roemerianus and Schoenoplectus americanus. Tidal freshwater marshes are often comprised of more diverse assemblages that can include broadleaf plants such as Sagittaria lancifola and Pontedaria cordata, although these assemblages can also be dominated by sedges like Schoenoplectus pungens (Blum et al. 2005) and Cladium jamaicense (NCDEHNR 1997). Sharply delimited zonal bands occur between monodominant assemblages within salt marsh ecosystems (e.g., S. alterniflora in low salt marsh) as a consequence of exposure to stressors like inundation (Odum 1998, Donnelly and Bertness 2001). Sharp transitions can also occur among contrasting assemblages in tidal brackish and freshwater marshes, though local environmental heterogeneity often results in bounded mosaics rather than zonal bands (Odum 1998). Transitions among ecosystems also are often bounded, but shifts can sometimes be attenuated as a result of hybridization (Ayres et al. 2004, Blum et al. 2005). This study focused on two ecologically contrasting sedges, C. jamaicense and S. americanus, as indicators of marsh plant community composition in tidal brackish and freshwater marshes. Commonly known as sawgrass, C. jamaicense is a perennial sedge that grows throughout the year in coastal marshes of the southeastern United States (Steward and Ornes 1975). It is typically found in fresh to slightly brackish or oligohaline marshes with background levels of salinity of ≤ 5 ppt (Brewer and Grace 1990; Ewe et al. 2007; USFS 2014). Above-ground primary productivity of C. jamaicense is negatively correlated with salinity and time of inundation (Valentine 1977; Childers et al. 2006; Macek and Rejmankova 2007). In the Florida Everglades, where it has been well-studied, populations of C. jamaicense have shifted inland 3.3 km since the 1940’s, a change attributed to RSLR and water management practices (Ross et al. 2000). Schoenoplectus americanus is a perennial sedge species that is found in marshes from South America to Nova Scotia and along the Pacific coast of the United States (Sipple 1978). On the Atlantic coast, S. americanus tolerates salinities of 2 ppt to
17 ppt, with peak performance occurring in soils with salinities of 5 ppt to 10 ppt (Ross and Chabrek 1972; Hess 1975). Consequently, this species has been used as an indicator of brackish conditions and as a model organism for studies of ecosystem responses to climate-related stressors such as increased salinity due to sea-level rise and elevated CO2 (Blum et al. 2005; Cherry et al. 2008; Blum et al. 2010; Langley and Megonigal 2010; Kirwan and Guntenspergen 2012). Both C. jamaicense and S. americanus produce annual crops of resilient seeds with thick coats that contribute to viability and taphonomic preservation (Mohamed-Yasseen et al. 1994; Sherfy and Kirkpatrick 1999), resulting in longstanding seed banks that have served as environmental archives for reconstructing annual-to-millennial records of plant community shifts in coastal wetlands (Brush 2001; Saunders 2003; Saunders et al. 2006). Schoenoplectus seeds are so persistent that they are routinely used for 14C dating in sedimentary records (e.g., Törnqvist et al. 2004). Seed profiles of these species offer a basis for inferring prevalence over time since seed production is also positively correlated with aboveground biomass (Sherfy and Kirkpatrick 1999; Saunders 2003; Saunders et al. 2006). Seed profiles of both species exhibit little evidence of taphonomic bias due to production, deposition, or preservation (Saunders 2003, Saunders et al. 2006). Seed densities for the two species are high; when measured at the center of dense monospecific stands, seed density can reach 400 seeds m−2 for S. americanus (Diggory and Parker 2010) and 5000 seeds m−2 for C. jamaicense (Alexander 1971). Comparable densities were found in the marshes examined for this study (ESM 1). The seeds of both species are heavy compared to seeds of other wetland plant species- seeds of S. americanus weigh ~0.5 mg (Diggory and Parker 2010), while C. jamaicense seeds weigh ~3 mg (Miao et al. 1997)- which is likely to result in relatively low dispersion. Furthermore, seed viability and germination rates for both species tend to be low (e.g., Alexander 1971; Webb et al. 2009), and the depositional nature of coastal wetlands promotes anoxic conditions with low decomposition rates (Lee 1992) and reduced bioturbation that favor seed preservation. Site Description and Sampling Sediment cores were collected from two brackish marsh sites in Pamlico Sound, North Carolina; both locations support plant communities currently dominated by S. americanus (Fig. 1). Scranton Marsh is adjacent to the Buck Paysour Memorial Bridge, approximately 6 km northwest of the town of Scranton (N 35° 30.023′ W 76° 26.695′). This marsh consists of large stands of S. americanus with smaller interspersed patches of J. roemerianus. Cedar Island Marsh is located at the Cedar Island Bridge outside of Morehead City (N 34° 54.977′
Wetlands
W 76° 22.197′) and is dominated by S. americanus with interspersed stands of J. roemerianus. Both marshes were inundated by less than 10 cm when sampled, which was approximately during average high tides according to the nearest NOAA tide gauge in Beaufort, NC (NOAA 2014). Both marshes also appear to be located in areas that have been relatively stable over the past several decades according to aerial imagery (ESM 2). Ten 50-cm sediment cores were manually collected at each site in October of 2011 using a 2-cm diameter Eijkelkamp Russian peat sampler to prevent compaction. Cores were collected in each marsh in a radial configuration, at 3 locations within stands of S. americanus that were separated by 120° and approximately 100 m apart. This approach, similar to that taken in Grandin and Rydin (1998), helped maximize spatial coverage and capture within-site heterogeneity in each marsh (Saunders 2003). Cores were stored in PVC tubes to prevent deformation and transported on ice to Tulane University, where they were kept at 4 °C. Within a month of collection, all 20 cores were cut into 2 cm intervals that were stored at 4 °C until further processing. One core from each marsh was chosen for radionuclide dating based on a preliminary assessment of seed counts across the cores. Approximately 5 g of sediment was removed from each section of these cores for radionuclide dating and sediment bulk density analysis. The remaining sediment from these cores and all sediment from the other 18 cores was sieved for seeds.
Sediment Geochronology Chronologies of sediment accretion were developed using the naturally-occurring, particle reactive radionuclide 210Pb. We chose 210Pb as a dating technique for this study because, when measured via alpha-spectrometry, it can be analyzed using a constant rate of supply (CRS) model that provides unique ages for each interval of sediment, often at resolutions of
