The 8.2 ka cooling event is largest climate excursion in the Holocene. This cooling event has been associated with a final, catastrophic draining of proglacial lakes Agassiz and Ojibway, which released ~163,000 km3 of fresh water into the North Atlantic (Barber et al., 1999). Greenland ice cores record a 6°C cooling, lasting ~100 years, around 8.2 ka BP. The 8.2 cooling event was widely spread (Alley et al., 1997) yet remains undetected in most high-latitude marine records. We attempt to answer if the 8.2 ka cooling event left a signal in the North Atlantic and if the 8.2 ka event manifested itself as a cooling or freshening or a combination of both.

NW Iceland was chosen as our study site for its sensitive location close to the oceanic Polar Front, the boundary between the water masses of the Irminger Current (IC) and the East Greenland Current (EGC) (Fig 1). Today the waters of the IC overlie our study site. The IC, a branch of the North Atlantic Current, carries relatively warm (~4-7°C) and saline (>35.0‰) waters northward while the EGC carries relatively cold (~0°C) and fresh (~34.4‰) waters southward. The 39-m long Calypso piston core MD99-2266 (66°13’77”N, 23°15’93”W, water depth 100 m) was retrieved at the mouth of Isafjardardjup, the largest fjord incising the Vestfirdir Peninsula, NW Iceland, as part of the IMAGES V cruise (Labeyrie et al., 2003).

We present prelimary results of Mg/Ca and d18O of the benthic foraminifer Cibicides lobatulus to detect the 8.2 ka cooling event on the Icelandic shelf. We took advantage of the high sedimentation rates of MD99-2266. A conventional sampling resolution of 100 years might not detect an event that lasted only 100 years, so we sampled at a resolution of ~15-24 years between 7700 and 8400 cal yr BP. The chronology in this interval is well constrained by 6 AMS 14C dates. We anticipated that the commonly used d18O of foraminiferal calcite, which is a function of temperature and salinity, might not have recorded the event, because a cooling would result in heavier isotopes and the freshening in lighter isotopes. We therefore also analyzed the Mg/Ca ratio of foraminiferal calcite, which appears to be a function solely of temperature. We chose Cibicides lobatulus, a benthic, epifaunal foraminifera that occurs in high percentages in the core, for both d18O and Mg/Ca analyses. We used the results of Mg/Ca to quantitatively reconstruct temperature, but first we had to calibrate the Mg/Ca of C. lobatulus against temperature. Our provisional calibration curve for the Mg/Ca of C. lobatulus against temperature (T), Mg/Ca=1.10+0.129 T, which is based on the analysis of 12 shallow (100-350 m) surface sediment samples from the SW/N Iceland and Greenland shelves with a robust temperature data set ranging from 0.6-7.2° C. During the 8.2 ka event the d18O of C. lobatulus composition became heavier by 0.2‰, which would indicate a ~1°C cooling if due to temperature alone. The temperature reconstruction based on the Mg/Ca analysis showed a ~3°C cooling at 8.2 ka, lasting ~100 years (Fig 2). We used these results to obtain the salinity contribution in the d18O by calculating the d18O of seawater using the Lynch-Stieglitz (2003) Cibicides equation: d18O(foraminifera)- d18O(seawater)+0.27=-0.21T+3.38. The results showed a lighter d18O of seawater composition by ~0.4‰. We estimate that a 0.4‰ change in the d18O of seawater corresponds to roughly a ~0.5-1.0 ‰ change in salinity. Two different scenarios could explain the mechanisms for the cooling and freshening that our proxies see on the NW Icelandic shelf. The first scenario is that the freshwater from the outburst was entrained into the waters of the North Atlantic Current and subsequently into the IC and lowered the overall salinity of the North Atlantic. The second scenario is that more water from the EGC reached our study site. The difference in salinity between the two currents is about 0.7‰ today.