Stable isotope values in coastal sediment estimate subsidence near Girdwood during the 1964 great Alaska earthquake

Subsidence during the Mw 9.2, 1964 great Alaska earthquake lowered Turnagain Arm near Girdwood, Alaska by ~1.5m and caused rapid relative sea-level (RSL) rise that shifted estuary mud flats inland over peat-forming wetlands. Sharp mud-over-peat contacts record these environment shifts at sites along Turnagain Arm including Bird Point, 11km west of Girdwood. Transfer functions based on changes in intertidal microfossil populations across these contacts accurately estimate earthquake subsidence at Girdwood, but poor preservation of microfossils hampers this method at other sites in Alaska. We test a new method that employs compositions of stable carbon and nitrogen isotopes in intertidal sediments as proxies for elevation. Because marine sediment sources are expected to have higher δ13C and δ15N than terrestrial sources, we hypothesize that these values should decrease with elevation in modern intertidal sediment, and should also be more positive in estuarine mud above sharp contacts that record RSL rise than in peaty sediment below. We relate δ13C and δ15N values above and below the 1964 mud/peat contact to values in modern sediment of known elevation, and use these values qualitatively to indicate sediment source, and quantitatively to estimate the amount of RSL rise across the contact. To establish a site-specific sea level datum, we deployed a pressure transducer and compensatory barometer to record a 2-month tide series at Bird Point. We regressed the high tides from this series against corresponding NOAA verified high tides at Anchorage (~50km west of Bird Point) to calculate a high water datum within ×0.14m standard error (SE). To test whether or not modern sediment isotope values decrease with elevation, we surveyed a 60-m-long modern transect, sampling surface sediment at ~0.10m vertical intervals. Results from this transect show a decrease of 4.64‰ in δ13C and 3.97‰ in δ15N between tide flat and upland sediment. To evaluate if δ13C and δ15N values shift positively from peat into mud across subsidence contacts, we analyzed samples above and below the 1964 contact in 11 sediment cores along the same transect. Across the 1964 contact, δ13C and δ15N shifts are consistently positive (up to +4‰) from peat to mud. To assess these shifts in terms of RSL change across the 1964 contact, we regressed modern sediment isotope values against elevation (δ13C R=0.60, avg. SE=×1.01‰, δ15N R=0.77, avg. SE=×0.50‰). These regressions predict elevations, referenced to a site-specific high water datum, for sediment above and below the 1964 contact. By subtracting the average δ13C and δ15N- predicted elevations of mud from peat we estimate an average 1.5×0.7m of RSL rise across the contact on the transect. Our method estimates elevation changes across the 1964 contact at Bird Point that are consistent with the ~1.5m subsidence observed near this location. We conclude that δ13C and δ15N show considerable promise as proxies for paleoelevation in coastal sediment, and may be useful to estimate vertical displacements of the coast during earthquakes.