We investigated use of δ13C in bulk organic sediment to define the botanical origin of samples preserved in coastal sediment as a means to reconstruct relative sea level in New Jersey, USA. Modern transects at three sites demonstrated that low and high salt‐marsh floral zones dominated by C4 species (Spartina alterniflora and Spartina patens) were associated with sediment δ13C values between −18.9‰ and −15.8‰ and occurred from mean tide level (MTL) to mean higher high water (MHHW). Brackish transitional settings vegetated by Phragmites australis with Iva fructescens and Typha sp. (C3 species) and freshwater upland samples (C3 species) were characterized by bulk sediment δ13C values of −27.0‰ to −22.0‰ and existed above MHHW. Parallel transects at one site suggested that intra‐site variability was not discernible. The utility of δ13C values for reconstructing relative sea level in New Jersey is limited by an inability to differentiate between brackish sediments related to sea level and freshwater upland samples. To facilitate this distinction in a 4.4 m core, we used a multi‐proxy approach (δ13C values with presence or absence of agglutinated foraminifera) to recognize indicative meanings for four sample types. Sediment with δ13C values greater than −18.9‰ was derived from a vegetated salt‐marsh and formed between MTL and MHHW. Sediment with δ13C values less than −22.0‰ and containing agglutinated foraminifera formed in a brackish transitional zone between MHHW and highest astronomical tide (HAT). This is the narrowest elevational range of the four sample types and most precise sea‐level indicator. Sediment with δ13C values less than −22.0‰ and lacking foraminifera can only constrain the upper bound of former sea level. Samples with intermediate values (−22.0‰ to −18.9‰) formed between MTL and HAT. Using these indicative meanings and radiocarbon dates, we suggest that a transition from brackish to salt‐marsh δ13C values recorded in the core took approximately 350 years (from 1800 to 1450 cal. a BP). Copyright © 2011 John Wiley & Sons, Ltd.

No abstract

We investigated the effect of tidal‐range change and sediment compaction on reconstructions of Holocene relative sea level (RSL) in New Jersey, USA. We updated a published sea‐level database to generate 50 sea‐level index points and ten limiting dates that define continuously rising RSL in New Jersey during the Holocene. There is scatter among the index points, particularly those older than 7 ka. A numerical model estimated that paleotidal range was relatively constant during the mid and late Holocene, but rapidly increased between 9 and 8 ka, leading to an underestimation of RSL by ∼0.5 m. We adjusted the sea‐level index points using the paleotidal model prior to assessing the influence of compaction on organic samples with clastic deposits above and below (an intercalated sea‐level index point). We found a significant relationship (p = 0.01) with the thickness of the overburden (r = 0.85). We altered the altitude of intercalated index points using this simple stratigraphic relationship, which reduced vertical scatter in sea‐level reconstructions. We conclude that RSL rose at an average rate of 4 mm a−1 from 10 ka to 6 ka, 2 mm a−1 from 6 ka to 2 ka, and 1.3 mm a−1 from 2 ka to AD 1900. Copyright © 2013 John Wiley & Sons, Ltd.

We present an objective and quantitative technique to reconstruct former sea level from assemblages of salt-marsh foraminifera using partitioning around medoids (PAM) cluster analysis and linear discriminant functions. Three salt marshes, representing different physiographic environments in southern New Jersey, were selected for describing the modern distribution of foraminifera from 56 surface samples. PAM estimated the number and composition of assemblages present at each site and showed that foraminifera adhered to the concept of elevation-dependent ecological zones, making them appropriate sea-level indicators.This approach has several advantages in comparison to hierarchical clustering. Application of PAM to a regional dataset generated by combining all samples identified the presence of five distinctive biozones occupying defined elevation ranges. The characteristics of these biozones were similar to those identified elsewhere on the mid-Atlantic coast of the USA. Classification of each of the 56 samples as belonging to one of the five biozones enabled us to develop linear discriminant functions, which confirmed their distinctiveness. These functions can estimate the probability that assemblages of foraminifera preserved in samples of salt-marsh sediment represent one of the five modern biozones. Recognition of these biozones in sequences of salt-marsh sediment provides a means to reconstruct sea level. We collected a 4.0 m core from Leeds Point, New Jersey to investigate the practical application of this approach to reconstructing former sea level. The linear discriminant functions were used to indicate the faunal origin of 32 core samples and in cross validation tests were accurate in 54 of 56 cases.The approach described can be used as an independent means to reconstruct sea level or to check the ecological plausibility of results from other techniques including transfer functions.

Substantial spatio-temporal variation in foraminiferal inputs occur over short areal distances at the sedimentwater interface and downcore as a result of patchy distributions and seasonal reproduction; foraminiferal assemblages are in turn diagenetically overprinted by seasonal, inter-seasonal, and inter-annual changes in porewater chemistry. Seasonal surface and near-surface assemblages are typically unrepresentative of deeper assemblages that are more likely to be incorporated into the sedimentary record. Cluster analysis of ''artificially time-averaged'' (ATA) assemblages revealed a distinct change in assemblages at ϳ20 cm depth. Differential preservation of foraminifera in the upper 60 cm, and especially the upper 20 cm, of sediment may produce an apparent paleoenvironmental change that could potentially be misinterpreted as a rapid fall in sea-level over the last ϳ100-200 years.