Multi-specimen and multi-site calibration of Aleutian coralline algal Mg/Ca to sea surface temperature

Higher latitude oceanic and climatic reconstructions are needed to distinguish natural climate variability from anthropogenic warming in regions projected to experience significant increases in temperature during this century. Clathromorphum nereostratum is a long-lived coralline alga abundant along...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Geochimica et cosmochimica acta 2014-08, Vol.139, p.190-204
Hauptverfasser: Williams, B., Halfar, J., DeLong, K.L., Hetzinger, S., Steneck, R.S., Jacob, D.E.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Higher latitude oceanic and climatic reconstructions are needed to distinguish natural climate variability from anthropogenic warming in regions projected to experience significant increases in temperature during this century. Clathromorphum nereostratum is a long-lived coralline alga abundant along the Aleutian archipelago that records seasonal to centennial fluctuations in seawater temperatures in its high-Mg calcite skeleton. Thus, C. nereostratum is an important proxy archive to reconstruct past seawater temperature variability in this data-poor subarctic region. Here, we measured magnesium to calcium ratios (Mg/Ca) by laser ablation-inductively coupled plasma-mass spectrometry (LA-ICP-MS) along the growth axis in six live-collected specimens from three islands in the Aleutian archipelago to assess Mg/Ca reproducibility and to calibrate algal Mg/Ca against modern gridded sea surface temperature (SST) data products. The master Mg/Ca–SST transfer function, determined by averaging the algal Mg/Ca–SST from each island (n=6), resulted in a reconstruction error of ±0.45°C, a 31–46% reduction in error compared to the reconstruction error for a single alga. The master algal-SST record interpolated to monthly and annual resolution significantly varied with gridded SST data products (r2=0.98, p
ISSN:0016-7037
1872-9533
DOI:10.1016/j.gca.2014.04.006