Vulnerability of Coastal Wetlands in the Southeastern Ljmted

The ability of coastal wetlands to keep pace with sea-level rise through vertical accretion and transgression onto adjacent upslope habitats was examined for the following coastal habitats in the southeastern United States: low salt marsh, high salt marsh, brackish marsh, and mangrove forests (fringe, basin, and overwash islands). The relationship between vertical accretion and soil elevation change was determined for these wetland types along gradients of marsh type, tidal range, and subsidence. Simultaneous measures of vertical accretion and soil elevation change were used to calculate each wetland's rate of shallow subsidence (accretion minus elevation change) and to evaluate the potential for submergence of each wetland given the local rate of sea-level rise. Significant rates of shallow subsidence were measured at 7 of 12 marsh-mangrove wetlands. Four of these seven sites experienced a significant elevation deficit (elevation minus relative sea-level rise), but none of the fow sites experienced a significant accretion deficit (accretion minus relative sea-level rise), indicating how misleading accretion data can be when evaluating the potential for submergence of coastal wetlands. These findings also indicated that subsurface processes occurring in the top few meters of the soil were as or more important in determining marsh elevation than surface accretionary processes for some of the marshes. Subsurface processes that * Likely influenced elevation included compaction, plant growth, plant decomposition, and shrink-swell from water storage. Forces driving these processes apparently included seasonal changes in water levels and aperiodic occurrences of mjor storms. Hence, the potential for submergence of some coastal marshes is best determined by calculating elevation deficits rather than accretion deficits. Rates of marsh transgression onto adjacent upland forests were determined at two high salt marshes located on Pamlico Sound, North Carolina, by radiocarbon dating of basal marsh peats along transects running from the marsh into the forest. The rate of movement of the marsh edge onto adjacent forest habitat was neither gradual nor constant, but was instead punctuated. The ages of the basal marsh sediments were grouped. indicating that transgression occurred as a series of events separated by longer periods of relatively little marshedge movement. It i s unlikely that the transgression events were initiated directly by rapid, local relative sea-level rise because the events were not synchronous between the two sites. Rather, the transgression events were likely generated by disturbance of the upland vegetation. allowing the marsh to leap forward. A likely vector for disturbance would be nlajor storms (e.g., hurricanes) or fires, although sea-level rise is still the driving force behind the hnsgression. Hence, short-term rates of marsh msgression may be meaningless and may not be useful tools to predict wetland habitat change, at least for some marshes. The findings from this study indicate that predicting the vertical buildup and migration of coastal wellands in response to sea-level rise requires site-specific information, and also that we need to understand more about the interactions among vegetation, soil, and hydrologic processes as they relate to soil elevation in marshes and mangrove forests if we are to properly manage these resources during future increases in sea level. Specifically. we need information on shallow subsidence from additional environmed sefhngs, the critical processes controlling elevation in each enviro~lental setting and site, the natural forces driving the processes controlling elevation, and the influence of management processes on shallow subsidence, marsh transgression, and the potential for marsh submergence. ' Pre.wnr Address: University of New Odeans DcpaRmeni of Geology and Geophysics. New Otlcans. LA 70148 F'rescnt Addrass: Western Cm)inn Univessity. Department of Geosciencw. Stillwell 207, Cullowhte, NC 27823 20 BIOLOGICAL SCIENCE REPORT USGSIBRDIBSR-1998-0002 Introduction Hence, the potential for coastal marsh submergence is In this chapter we present a summary of the technical results of a multiyear investigation into the relationships among accretionary processes, sea-level rise, and the potential for submergence of coastal wetlands during future rises in sea level. Technical explanations of the methodologies and experimental design used in this study are provided in Boumans and Day (1993), Cahoon et al. (1995a; 1995b; 1996a; 1996b), Reed and Cahoon (1 992), and Reed et al. (1995). Global mean sea level has risen approximately 1-2 mm/ year during the past 100 years (Gornitz 1995). The Intergovernmental Panel on Climate Change predicts a 50-cm rise in average global eustatic sea level by 2100, between two and five times the rise in the past century (Watson et al. 1996). The low and high estimates of change range from 15 to 95 cm (Watson et al. 1996). Latest estimates by the U. S. Environmental Protection Agency (USEPA) similarly indicate that global warming will likely raise sea level 15 cm by the year 2050 and that the present rate of eustatic sealevel rise will increase by 4.2 &year (twoto fourfold) by the year 2100 (Titus and Narayanan 1995). These estimates do not include sea-level rise caused by factors other than global warming (e.g., land subsidence). Areas with high local rates of subsidence, such as the Mississippi River delta, are currently experiencing relative sea-level rise rates (i.e., eustatic sea-level rise plus land subsidence) up to 10 times the global mean sea-level rise rate (Penland and Rarnsey 1990; Gornitz 1995). The potential for submergence of coastal wetlands, particularly deltas, will increase under a scenario of rising relative sea level (Gornitz 1991). In order for marshes not to become submerged as sea level rises, vertical buildup of the marsh surface will have to equal or exceed the rate of relative sea-level rise. The question arises as to how an increase in sea level will affect marsh sediment deposition, vertical accretion processes, and ultimately elevation. An increase in sea level may increase or decrease the local tide range and may also result in a phase shift of the estuary from floodto ebb-dominated (Dyer 1995). Such changes would directly affect patterns of water circulation, suspended sediment concentrations, and marsh flooding, as well as rates of vegetation growth and, ultimately, marsh accretionary processes (Dyer 1995; Reed 1995). Also, increased global warming may increase the frequency of hurricanes (Emanuel 1987), which could influence local sediment supplies and marsh accretionary processes as well as erosion caused by wind and waves. As sea level rises, marshes may migrate landward, provided there is no barrier to movement such as a fmed structure (e.g., a building or seawall). Subsidence can vary strongly among sites: for example, being high in deltas with thick Holocene deposits and low on stable geologic formations such as ancient shields. controlled more by the local environmental factors of relative sea-level rise, coastal geomorphology, sediment supply, and frequency of major storms than by the trend in eustatic mean sea level (Gornitz 1995). Therefore, predicting the potential for coastal marsh submergence caused by sea-level rise requires site-specific information and an improved understanding of the interactions among marsh vegetation, soil, and hydrological processes. In this study, we investigated marsh accretion and elevation change in coastal marshes and mangroves selected along gradients of tidal range, subsidence, and marsh type in order to evaluate the processes controlling both vertical development of the marsh surface and transgression (horizontal movement) of the marsh surface onto adjacent uplands. The goal of this study was to improve our understanding of the relationships among marsh accretionary processes, marsh elevation changes, hydroperiod, relative sea-level rise, and the potential for coastal marsh submergence so that we could better evaluate the ability of marshes to keep pace with sea-level rise. Vertical Buildup of the Marsh Surface Traditionally, the potential for marsh submergence Kds been determined by calculating accretion deficits (Reed and Cahoon 1993). Measured rates of vertical accretion are compared directly to local rates of relative sea-level rise. If accretion is not keeping pace with sea-level rise then an accretion deficit is said to exist, and the potential for submergence of the coastal marsh is high. The accretion deficit concept assumes that surface accretion measures are a good indicator of marsh surface elevation change. We also know, however, that the marsh surface subsides because of autocompaction of the Holocene marsh deposits (Kaye and Barghoorn 1964), as reflected in the sharp decline in water content as well as consolidation over the top 1 m of the marsh substrate (Kearney and Ward 1986) and as inferred from slower accretion rates for historic ('37Cs and 210Pb) and geologic (I4C) accretion methods (Reed and Cahoon 1993; Stevenson et al. 1986), and also that additional subsidence may occur through faulting and compaction of deep sediments, particularly in deltaic environments (Penland et al. 1989). Hence, the validity of the assumption that surface accretion equals elevation change should be questioned. If the increase in elevation is smaller than the vertical accretion gain because of soil subsidence, then the accretion deficit concept is underestimating the potential for coastal marsh submergence. Important questions for coastal managers include: What is the relationship between accretion and elevation change? How does this relationship vary over the range of coastal hydrogeomorphic settings? If elevation gain is slower than accretion gain, what are the important processes controlling elevation? What natural forces drive these processes? And, what are VULNERABILITY OF COASTAL WETLANDS LN THE SOUTHEASTERN UNITED STATES 2 1 the implications for managing coastal wetlands during periods of rising sea level? Sedimentation erosion table The objectives of this part of the study were to (1) mea. Marsh surface sure accretion and elevation change in marshes along gra, , ,i i. 1 .L , dients of marsh type, subsidence, and hydroperiod; (2) evaluate the relationship between accretion and elevation change both within and between sites; (3) evaluate the role of hydroperiod in determining marsh response to sea-level rise; and (4) evaluate the relationship between elevation C change and relative sea-level rise (i.e.. calculate an elevation deficit). Experimental Approach and Terminology In order to meet these objectives and answer these questions, it was necessary to develop a new investigative approach that simultaneously quantified vertical accretion and surface-elevation change with a level of accuracy sufficient lo distinquish between the influences of surface and subsurface processes on marsh elevation. Surface accretionary processes (e.g., sediment deposition and erosion) were determined from artificial marker horizon plots established on the marsh surface (Cahoon and Turner 1989). Marsh surface-elevation change was measured relative to a subsurface datum (usually 3-5 m deep) using a sedimentation-erosion table (SET) (Bournans and Day 1993). This method of measuring elevation integrates both surface processes (e.g., deposition, erosion) and subsurface processes (e.g., compaction, shrink-swell, plant growth, decomposition) occumng over the top several meters of the soil. The difference between the two simultaneous measures gives an estimate of the impact of subsurface processes on marsh surface elevation change and of the degree to which accretion measures alone underestimate the potential for coastal marsh submergence, if at all. The influence of subsurface processes on marsh elevation has been termed "shallow subsidence." These concepts and the relationship between the two methods are graphically presented in Fig. 3-1 and explained in detail in Cahoon et al. (1995a). Study sites were selected along gradients of marsh type, subsidence, and tidal range, primarily in the southeastern United States (Fig. 3-2). Most sites were located in saline coastal habitats of low salt marsh (Old Oyster Bayou, Bayou Chitigue, and Tijuana Slough National Wildlife Refuge m]), high salt marsh (St. Marks NWR and Cedar Island NWR), or mangrove (Rookery Bay) (Fig. 3-3a.b.c). Two sites were located in brackish marsh (McFaddin NWR and Three Bayous). At Rookery Bay, measurements were made in fringe, basin, and overwash island forests. The sites from the Mississippi River delta in Louisiana represent mas of high subsidence. Bayou Chitigue is a rapidly deteriorating low salt marsh, in contrast to Old Oyster Bayou, which is a stable, healthy marsh. All other sites represent low subsidence areas. A1I sites are microtidal. but the tidal range at St. Marks is five times greater than for Cedar Island. The tidal range at Tijuana Slough is three Figure 3-1. Conceptual diagram (not to scale) showing thase portions of the soil profile being measured by Ihe sedimentationerosion table (SET) and marker horizon techniques. The boundary separating shallow and deep subsidence is defhed operationally by the bottom of the SET pipe.