skip navigation

Previous | View all abstracts | Next

TESTING OF THREE NEW COOL-WATER CALIBRATIONS FOR THE MG/CA PALEOTHERMOMETER SUGGESTS REDUCED INFLOW OF ATLANTIC WATER ONTO THE N-ICELAND SHELF IN THE PAST 2000 YRS

KRISTJáNSDóTTIR, GRéTA B  INSTAAR and Dept. of Geol. Sciences, University of Colorado, Boulder, CO 80303, USA.
Jennings, Anne E  INSTAAR and Dept. of Geol. Sciences, University of Colorado, Boulder, CO 80303, USA.
Lea, David W  Dept. of Geol. Sciences, University of California, Santa Barbara, CA 93106, USA.

Calibrations of Mg/Ca ratio against temperature are reasonably well-constrained for temperatures >10°C but are poorly developed for cooler temperatures, thus limiting the method’s applicability to high-latitude and deep sea studies. This study focuses on strengthening the calibration at low temperatures by using a robust data set from the W, N Iceland and Greenland margins. The Iceland margin, near the northern limit of Atlantic Water, has a spatial temperature gradient of 0 to 9°C and is thus an ideal location to conduct cool water calibrations for Mg/Ca studies. Samples from the Greenland margin were then added to obtain the coldest endmember.

Previous studies have shown that the temperature-Mg/Ca relationship in benthic foraminifera is species-specific so that one common calibration cannot be obtained. Therefore we have chosen to develop calibrations for three common arctic, benthic species: Melonis barleeanus, Cassidulina neoteretis, and Islandiella norcrossi/helenae. All three species show Mg/Ca ratios comparable to published values for benthic foraminifera (0.6-2.2 mmol/mol). For both the calibration and downcore samples Mg/Ca was measured on an ICP-MS at the Univ. of California, Santa Barbara and isotope analysis was done at the Leibniz Laboratory at Kiel University, Germany.

The calibration samples were collected during cruises B997 and BS1191. They were stained with Rose Bengal upon collection and sieved at >63 µm. The three species of benthic foraminifera were picked from the >250, 150-250, and 106-150 µm fractions. Both stained and pristine, unstained individuals were picked due to a limited number of stained individuals in the samples. Furthermore, in order to get enough material for both Mg/Ca and 18O analysis the three size fractions had to be combined. Before analysis each sample (containing anywhere between 9 and 140 individuals) went through a rigorous cleaning procedure established by Boyle and Keigwin (1985/1986) as modified by Boyle and Rosenthal (1996). Because of the addition of pristine, unstained foraminifera (dead foraminifera) to the samples we have chosen to calibrate our thermometer against isotopic calcification temperature (calculated from the d18O measured in the same sample) rather than to use the CTD temperature obtained during the respective cruise. M. barleeanus shows the greatest temperature sensitivity, while both C. neoteretis and I. norcrossi/helenae show a slightly lower (but quite clear) sensitivity.

The downcore samples are from core MD99-2269. The core was collected in 1999 during the IMAGES V cruise aboard R/V Marion Dufresne. Site MD99-2269 is located in 365 m water depth in Reykjafjardaráll, N Iceland shelf (66°37.53´ N, 20°51.16´ W). The core is a 25 m long calypso piston core with basal radiocarbon date of ca. 12 ka cal BP and a modern coretop. Sample resolution is ca every 50 yrs. Samples were wet sieved at >63 µm and picked for foraminifera in the following size fractions: >250, 150-250, 106-150 µm. Where available all three arctic, benthic species were analyzed. Due to small sample size the three different size fractions had to be combined. Before trace-element analysis each sample went through the same cleaning procedure as the calibration samples. However, due to the small sample sizes (< 175 µg) sonication time and reagents were scaled down (not changing the concentration, just the amount) from normal procedure to maximize sample recovery. Only results with >10% recovery were used for data analysis, although some samples with less recovery were no different from the rest.

Applying our new calibrations to the downcore samples confirms that the new calibrations function well for this temperature range. Mg/Ca ratios during the last two millennia in core MD99-2269 show a clear species-specific effect where M. barleeanus ratios are consistently lower that those of C. neoteretis or I. norcrossi/helenae supporting the view that incorporation of Mg into foraminiferal calcite is species-specific. However, each species reconstructed a temperature record that was similar to the other two species and was within the expected range for this site. Cool conditions of <1°C are observed during the LIA; a warm spike of 4°C at 500 cal yr BP is seen in all three species; a second warm spike of 5°C is observed at 1500 cal yr BP, coincident with a peak in carbonate content in the core. Carbonate has a similar trend as the reconstructed temperature during the last two millennia suggesting that it could serve as an indicator of temperature on the N Iceland shelf. An overall cooling trend (and decrease in carbonate) from 2 ka cal BP to the present is interpreted as a decrease in inflow of warm Atlantic water in the Irminger Current to the site. We hypothesis that as the inflow decreased the Atlantic water changed from being a bottom water mass at the site and became an intermediate watermass as is observed in the area today.

REFERENCES
Boyle, E. and Keigwin, L. D., 1985/1986, Comparison of Atlantic and Pacific paleochemical records for the last 215,000 years: Changes in deep ocean circulation and chemical inventories: Earth and Planetary Science Letters, v.76, p. 135-150.



Boyle, E. & Rosenthal, Y., 1996, Chemical hydrography of the South Atlantic during the last glacial maximum: Cd vs. d13C: In Wefer, G., Berger, W. H., Siedler, G., and Webb, D. J. (eds.), The South Atlantic: Present and past circulation. Berlin, Heidelberg: Springer-Verlag, p. 423-443.


Previous | View all abstracts | Next