Sea-level change during the last 2500 years in New Jersey, USA
Introduction
Late Holocene (last ∼2500 years) sea-level reconstructions with decimeter vertical and multi-decadal temporal resolution provide important information for investigating the relationships between sea-level change and temperature. Such data calibrate and test the validity of models that predict sea-level changes under scenarios of global climate evolution (e.g. Rahmstorf, 2007, Bittermann et al., 2013). Instrumental measurements of sea level (tide gauges and satellites) are too short to capture pre-anthropogenic trends and the long-term (100s–1000s of years) response of sea level to temperature variations. This time-series limitation hinders the development of predictive models and is a motivation for reconstructing late Holocene sea-level changes. Proxy-temperature data show distinct climate phases during the late Holocene such as the Medieval Climate Anomaly, Little Ice Age, and 20th century warming (e.g. Moberg et al., 2005, Mann et al., 2008, Ahmed et al., 2013, Thompson et al., 2013). In contrast, relatively little is known about sea level during this period, although there is some evidence that persistent positive and negative departures from regional, linear background rates (driven primarily by glacio-isostatic adjustment; GIA) occurred prior to the onset of modern sea-level rise in the late 19th or early 20th centuries (e.g. Gehrels, 2000, van de Plassche, 2000, Sivan et al., 2004, González and Törnqvist, 2009, Kemp et al., 2011).
Salt-marsh sediment is one of the most important archives for reconstructing relative sea level (RSL) during the late Holocene. Under regimes of RSL, rise salt marshes accumulate sediment to maintain their elevation in the tidal frame (Morris et al., 2002). The resulting sequences of salt-marsh sediment accurately preserve the elevation of past RSL, which is the net result of all driving mechanisms. The vertical precision of RSL reconstructions is maximized by employing sea-level indicators that differentiate among salt-marsh sub-environments to estimate the tidal elevation where the sediment was originally deposited (paleomarsh elevation; PME). Salt-marsh foraminifera are sea-level indicators because their distribution is controlled by the frequency and duration of inundation, which is principally a function of tidal elevation (e.g. Scott and Medioli, 1978, Horton and Edwards, 2006). Foraminifera are abundant in salt marshes where they form assemblages occupying narrow elevational ranges making them suitable for quantitative and precise PME reconstructions. Bulk sediment geochemistry can also be employed as a sea-level indicator. In regions where salt marshes are dominated by C4 plants such as the mid-Atlantic and northeastern U.S., measured δ13C values readily identify sediment of salt-marsh origin (e.g. Middleburg et al., 1997, Wilson et al., 2005, Tanner et al., 2010). RSL reconstructions also require the timing of sediment deposition to be estimated. Sediment that accumulated under low-energy conditions on salt marshes is often undisturbed and well suited to developing detailed chronologies. Radiocarbon is the principal means to date late Holocene salt-marsh sediment, but alternatives are necessary for the period since approximately 1650 AD because of a plateau on the calibration curve (e.g. Reimer et al., 2011). Age–depth models developed from composite chronologies incorporating radiocarbon dates and age markers of pollution and land-use change enable RSL to be reconstructed with the multi-decadal precision necessary to describe small (decimeter) RSL changes (e.g. Marshall et al., 2007). The resulting RSL reconstructions filter out short-lived (annual to decadal) sea-level variability because of the time-averaging effect of sedimentation and sampling. The resulting records are analyzed using numerical tools to identify and quantify the timing and magnitude of persistent (decadal to centennial) phases of sea-level evolution.
Relative sea-level changes in New Jersey over the past ∼2500 years were reconstructed to determine how and when persistent sea-level trends deviated from background rates. Reconstructions were developed from salt-marsh sediment at two sites (Leeds Point and Cape May Courthouse; Fig. 1) using foraminifera and stable carbon isotopes (δ13C) as sea-level indicators and age–depth models constrained by composite chronologies of radiocarbon, 137Cs activity, and pollen and pollution chrono-horizons. Change point analysis identified four persistent periods of sea-level behavior during the last 2500 years that mark positive and negative departures from a linear background rate. The new reconstructions demonstrate that the rate of sea-level rise since ∼1850 AD exceeds any previous persistent rate in the late Holocene.
Section snippets
Study area
The New Jersey coast from Great Bay to Cape May consists of a barrier island and lagoon system separating the mainland from the Atlantic Ocean (Fig. 1). Inlets allow exchange of water between the lagoons and ocean. Great diurnal tidal ranges are smaller in the lagoons (typically 1.1 m–1.2 m, but as small as 0.17 m in the upper reaches of Barnegat Bay) than on the ocean side of the barrier islands (e.g. 1.4 m at Atlantic City). Tidal influence extends up to 25 km from the coast into bays and
Estimating paleomarsh elevation
At each site, multiple transects of hand cores were recovered to describe the underlying stratigraphy. The cores chosen for detailed analysis are Leeds Point core 10 (LP-10) and Cape May Courthouse core 8 (CMC-8) because they included some of the thickest sequences of high salt-marsh peat. Cores for laboratory analysis were collected using a Russian-type core to prevent compaction during sampling, sealed in plastic wrap and kept refrigerated. Samples of core material (1 cm thick) were sieved
Foraminifera and δ13C values in cores from Leeds Point (LP-10) and Cape May Courthouse (CMC-8)
To establish the environment and elevation of sediment deposition, foraminifera were counted in core samples positioned at regular intervals, dated levels, and to capture transitions between assemblage types and stratigraphic units. The lowest occurrence of foraminifera in LP-10 was at 3.95 m (Fig. 2). Between 3.95 m and 2.85 m the most common foraminifera was J. macrescens that occurred with T. comprimata and T. inflata. The interval between 3.13 m and 3.00 m was characterized by an unusually
Sea level change in New Jersey
The New Jersey RSL reconstruction is represented by boxes that incorporate sea-level and age uncertainty (Fig. 7). RSL in New Jersey was −4.20 m at approximately 500 BC and rose to −0.70 m at around 1850 AD. Agreement between the RSL reconstructions from Cape May Courthouse and Leeds Point between 970 AD and 1460 AD indicates that local-scale processes were not the dominant drivers of RSL in New Jersey, at least for that shared interval. The RSL reconstruction lies within the uncertainties of
Conclusions
Relative sea level was reconstructed at two sites in New Jersey from sequences of salt-marsh sediment. A multi-proxy approach combining a transfer function trained on the modern distribution of foraminifera and bulk sediment δ13C values as sea-level indicators estimated the elevation at which samples formed in one of three ways. (i) For samples with a δ13C value more depleted than −22‰, the transfer function estimate was trimmed to retain only the range above MHHW and reflect the dominance of C3
Acknowledgments
This work was supported by National Oceanic and Atmospheric Administration award NA11OAR4310101 and NSF grants EAR-1052848 and EAR-0951686. Kemp thanks a graduate student internship at NOSAMS that provided radiocarbon dating under the supervision of Dr. Mark Roberts. Vane publishes with permission of the director of the British Geological Survey who partially supported this work through the BGS Climate and Landscape research program. We thank the Edwin Forsythe National Wildlife Refuge (U.S.
References (92)
- et al.
Salt marshes as late Holocene tide gauges
Global Planet. Change
(2013) - et al.
Isotopic source signatures for atmospheric lead: the Northern Hemisphere
Geochim. Cosmochim. Acta
(2001) - et al.
Modelling the effects of sediment compaction on salt marsh reconstructions of recent sea-level rise
Earth Planet. Sci. Lett.
(2012) - et al.
Carbon-isotope, diatom, and pollen evidence for Late Holocene salinity change in a brackish marsh in the San Francisco Estuary
Quater. Res.
(2001) Rapid sea-level rise
Quat. Sci. Rev.
(2012)- et al.
Holocene sea level database for the Atlantic coast of the United States
Quat. Sci. Rev.
(2012) - et al.
Diagenesis of organic matter in Georgia salt marshes
Estuar. Coast. Shelf Sci.
(1989) - et al.
Nineteenth and twentieth century sea-level changes in Tasmania and New Zealand
Earth Planet. Sci. Lett.
(2012) - et al.
Onset of recent rapid sea-level rise in the western Atlantic Ocean
Quat. Sci. Rev.
(2005) - et al.
Upper Mississippi Pb as a mid-1800s chronostratigraphic marker in sediments from seasonally anoxic lakes in Eastern Canada
Geochim. Cosmochim. Acta
(2013)