Text

theses

Relative sea level (RSL) in New Jersey has exhibited varying rates of change across time and space throughout the late Holocene (last 4000 years), due to factors including glacial isostatic adjustment (GIA), ocean/atmosphere dynamics, and sediment compaction. High resolution sea-level reconstructions from salt-marsh proxies (e.g., foraminifera and geochemistry) have extended the instrumental record beyond the 20th century to link with late Holocene data. In this dissertation, I examined the accuracy of sea-level proxies, timing of the onset of modern rates of sea-level rise, and mechanisms contributing to sea-level change in New Jersey during the late Holocene.

In Chapter 1, I examined temporal and spatial variability of salt-marsh foraminifera and stable carbon isotope geochemistry (δ13C) to include seasonal/interannual and small-scale spatial changes in RSL reconstructions. I conducted a three-year monitoring experiment on four high marsh sampling stations in southern New Jersey, collecting 46 samples through time with 136 spatial replicate samples. Variations in annual standing crop that were observed did not exhibit any interannual or seasonal patterns. Foraminiferal assemblages and dominant species remained consistent on small spatial scales at each monitoring station over the study period; however, standing crop varied among replicate samples. Temporal and small-scale spatial variability in δ13C values at each station was very minimal. The foraminifera assemblages were separated into unique site-specific assemblages, and the variation across monitoring stations explained ~87% of the total variation, while ~13% of the variation in foraminiferal assemblages was explained by temporal and/or spatial variability among replicate samples. I developed a method to formally incorporate the temporal and spatial variability of modern foraminiferal distributions into a Bayesian transfer function. I applied this to a Common Era relative sea-level record in New Jersey. Because of the limited variability of foraminifera, RSL reconstructions from high marsh environments remain robust and reproducible.

In Chapter 2, I produced a new high-resolution RSL record in northern New Jersey using a Bayesian transfer function that employs salt-marsh foraminifera and a secondary proxy, stable carbon isotope geochemistry, to examine magnitudes and rates of sea-level change. I found that RSL in northern New Jersey continuously rose over the last 1000 years by 1.5 ± 0.1 m (1σ). RSL rose at a rate of 1.2 ± 0.2 mm/yr (2σ) from 1000 to 1700 CE before accelerating to a rate of rise of 2.1 ± 0.3 mm/yr from 1700 CE to present. I integrated the new northern New Jersey record into an updated global database of instrumental and proxy sea-level records of the Common Era. I used a spatiotemporal empirical hierarchical model with the global database to estimate the timing of the onset of modern elevated rates of sea-level rise. I propose ~1870 CE as the global onset of modern rates of sea-level rise. I examined the spatial variability in timing of modern elevated rates of RSL rise at nineteen sites in the North Atlantic which have the highest resolution of all of the proxy locations in the global RSL database and found asynchronous timing with a distinct spatial pattern. Elevated rates in RSL appear earliest in the mid-Atlantic region, followed by the northeastern and southeastern U.S., and latest in Canada and Europe. I suggest that the observed spatial pattern in the timing may be due to a combination of steric and ocean dynamic effects from changes in Atlantic Meridional Overturning Circulation and the Gulf Stream.

In Chapter 3, I produced a new late Holocene RSL record in southern New Jersey from basal peat units to examine mechanisms of sea-level change, including GIA, as it is the dominant cause of late Holocene RSL rise in New Jersey. The paleomarsh elevation for each basal sample was estimated through a Bayesian transfer function using salt-marsh foraminifera assemblages and bulk sediment stable carbon isotope geochemistry. Sediment compaction and tidal range change were accounted for in each sample using a geotechnical model and a paleotidal model. I found that RSL rose continuously through the late Holocene by ~6.4 m from 4.6 ka BP to 1.2 ka BP at an average rate of 1.9 ± 0.3 mm/yr. I compared RSL changes from the basal peat record with site-specific GIA models that include two components: an ice history and a viscosity model. I used one 1D model that assumes the lithosphere and mantle viscosity are laterally homogeneous and nine 3D models that allow for mantle viscosity lateral heterogeneity. I found a misfit between model predictions and RSL observations, although the 3D models were an improvement over the 1D model. The remaining misfits suggest the importance of utilizing a wide array of ice model and viscosity model parameters to find a better fit between site-specific GIA predictions and RSL observations.

ETD_10353

Ph.D.

Includes bibliographical references

doi:10.7282/t3-156z-zn37

ETD doctoral

The author owns the copyright to this work.