Relative utility of foraminifera, diatoms and macrophytes as high resolution indicators of paleo-sea level in coastal British Columbia, Canada

Abstract

A multiproxy analysis was carried out on diatom, foraminiferal and macrophyte assemblages across the saltmarsh at Zeballos, Vancouver Island, British Columbia. To determine which group, or combination of groups provided the most accurate elevational zonations, 36 stepwise multiple linear regressions (SMLR) were carried out using a variety of data transformations on an elevational training set. Adjusted R 2 values yielded statistically significant results in all analyses as follows: foraminifera (0.658–0.870); diatoms (0.888–0.974); macrophytes (0.671–0.844); foraminifera/diatoms (0.941–0.981); foraminifera/diatoms/macrophytes (0.958–0.993). The most realistic SMLR results were obtained when data transformations comprised of (ln) normalized fractional abundance data was carried out on species present in statistically significant numbers (NrfaEQ). Of the individual proxies assessed, diatoms yielded the most significant adjusted R 2 results, with the low marsh diatom Achnanthes hauckiana being one of the most important predictor variables (pv's). Amongst the foraminifera, the low marsh species Miliammina fusca and high marsh Balticammina pseudomacrescens were determined to be the most significant pv's. For macrophytes, the low marsh species Carex lyngbyei, the high marsh species Juncus balticus, Shannon–Wiener Diversity Index (SDI) and absence of plant cover on the tidal flat were the most important pv's. As SMLR analysis of all individual groups and combinations of groups yielded statistically significant results, the choice of proxies, or combinations of proxies that are suitable for paleo-sea level research is at the discretion of the researcher.