Date of Award

2006

Level of Access

Campus-Only Thesis

Degree Name

Master of Science (MS)

Department

Earth Sciences

Advisor

Daniel F. Belknap

Second Committee Member

Joseph T. Kelley

Third Committee Member

Walter A. Barnhardt

Abstract

Late Quaternary relative sea-level change, driven by both isostasy and eustasy, is the primary process that drove migration of littoral environments in the northwestern Gulf of Maine over the last 15 cal kyrs BP. Earlier work suggested a lowstand at about 60 m below present sea level at approximately 11.8 cal kyrs BP. ago, followed by the on-going transgression, but the lowstand chronology is poorly constrained. To test the lowstand hypothesis, we created a base map of outer Saco Bay with a 10 km2 mosaic of multibeam bathymetry and backscatter coupled with 45 km2 of side-scan sonar. We surveyed stratigraphy and bathymetric landforms with 120 km of seismic-reflection profiles in the 40 to 80 m depth range and collected 17 vibracores of potential lowstand deposits. The area mapped by side-scan sonar is bedrock dominated with extensive gravel talus. Sand bodies are present at -60 m and give way to mud below -65 m water depth. The seismic data shows erosional shoreline features cut into glacial sediment and remnants of constructional beach environments with cores revealing up to 3 m of sand. Radiocarbon dates of cored intertidal Mya arenaria at -64 m, at -50 m and Mytilus eclulis beds at -36 m better constrain timing of the regression, lowstand and rapid transgression. The SWAN wave model run in Saco Bay at the lowstand water depths correctly estimated zones of accumulation and erosion when compared to the seismic and core data. By testing the - 60 m lowstand hypothesis and wave model with direct geophysical observations, cores and radiocarbon dates, a strong argument is made for the lowstand position and timing of sea-level changes in the northwestern Gulf of Maine. This area contains critical evidence that suggests rapid and small-scale isostatic effects not reproduced by existing numerical models.

Share