Sea surface temperature anomalies in the oceans at the LGM estimated from the alkenone‐U₃₇^K′ index: comparison with GCMs

Abstract

U37K′ sea surface temperature (SST) estimates for the Last Glacial Maximum (LGM) are discussed and compared to coupled general circulation model (GCM) results. We conclude that U37K′ estimates and most numerical models agree in the magnitude of the large scale SST cooling features of the glacial ocean. There was a larger cooling in the northern than in the Southern Hemisphere. The eastern boundary currents and equatorial regions were also colder. There is a consensus between models and alkenone data for a tropical cooling of about −2°C, which is in disagreement with both a smaller cooling as suggested by the CLIMAP project reconstruction, or a more pronounced cooling as suggested from coral proxy records. The comprehensive HadCM3 simulation examined reproduces a puzzling feature of the U37K′‐SST reconstruction, i.e., the warming of the glacial north Atlantic, which needs further investigation using numerical modelling and proxy data reconstruction.

Rapid succession of plant associations on the small ocean island of Mauritius at the onset of the Holocene

“…As a consequence, Mauritius may experience maximum and minimum precipitation events about every w10 kyr. However, tropical oceanic islands might be buffered during dry phases by elevated topographydriven orographic rains and the direct influence of relatively high SST (Cronk, 1997;Trend-Staid and Prell, 2002;Rosell-Melé et al, 2004). Anomalously high SST and associated rainfall can be propagated by the IOD.…”

Dynamical Reconstruction of Upper-Ocean Conditions in the Last Glacial Maximum Atlantic

“…Four MARGO compilations have significant coverage in the Atlantic Ocean, and all of them are used here: foraminiferal (Kucera et al 2005b;Hayes et al 2005) and dinoflagellate cyst (De Vernal et al 2005) assemblages, foraminiferal Mg/Ca (Barker et al 2005), and the alkenone U K 0 37 index (Rosell-Mel e et al 2004). Assumptions of seasonality in MARGO vary by proxy: annual-, winter-, and summer-mean NSST reconstructions were published for the foraminiferal and dinocyst assemblage proxies; only annual-mean reconstructions were published for the alkenone records; and annual-mean (summer mean) reconstructions were published for low-latitude (high latitude) Atlantic Mg/Ca records.…”

Atlantic Ocean circulation at the Last Glacial Maximum : inferences from data and models

“…An issue specific to assemblage-based techniques is a no-analog situation: it LGM records in our domain (see bottom row of Figure NSSTs from Alkenones. The alkenone U K ′ 37 index is a measure of the abundance of long versus short chain alkenones [Rosell-Melé et al, 2004], molecules that are produced by coccolithophores. Although the mechanism responsible for the variation of alkenone chain length with temperature is a matter of debate, laboratory and surface-sediment based empirical calibrations show a strong relationship between the index and temperature (see Rosell-Melé et al [2004] for a discussion).…”

“…The alkenone U K ′ 37 index is a measure of the abundance of long versus short chain alkenones [Rosell-Melé et al, 2004], molecules that are produced by coccolithophores. Although the mechanism responsible for the variation of alkenone chain length with temperature is a matter of debate, laboratory and surface-sediment based empirical calibrations show a strong relationship between the index and temperature (see Rosell-Melé et al [2004] for a discussion). As the index has been shown to correlate most highly with annual mean near sea surface temperature [Rosell-Melé et al, 2004] LGM temperature close to modern south of the equator, moderate LGM cooling from the equator to 45…”