PP31C-01
A Continuous IC Glaciochemical Record of the last Glacial Period from the NGRIP Ice Core
Comprehensive chemical analysis of aerosol components over the time period from 100,000 to 10,000 years before present was performed on the NGRIP ice core. In line with high-resolution chemical flow analysis, a continuous sampling of ice core melt water was performed at 55 cm resolution for ion chromatographic (IC) analysis of soluble ions. These records provide a unique potential for detailed investigations of conditions for source load and transport of aerosol during the last glacial period. A comparison with GISP2 IC record shows a good agreement between the two ice core records, but with generally higher concentrations in the NGRIP ice core. A more detailed comparison between the two ice core ion records reveals strong regional differences in source contributions and transport paths for sea salt and dust aerosol over Greenland during the glacial period, indicating a different transport pattern for aerosol over Greenland during the last glacial period.
PP31C-02 INVITED
Regional and Temporal Differences in Abrupt Climate Change Recorded in Greenland Ice
Detailed temporal and spatial examination of the processes, propagation, and variability of abrupt climate changes recorded in Greenland ice is now possible due to the availability of several deep Greenland ice cores coupled with the routine measurement of annually or better resolved proxy data. Greenland ice cores are uniquely suited for study of past abrupt climate events because relatively high snow accumulation rates allow single years to be identified well into the last glacial period. To more fully exploit the temporal and spatial resolution available in Greenland ice, here we present ice stable isotope measurements (oxygen-18 and deuterium) across the entire deglacial transition and several Dansgaard-Oeschger events in the NorthGRIP ice core with near-annual resolution, as well as several of these same climate transitions in GISP2 ice with similar detail. In both ice cores, calculated deuterium excess time series consistently show that the deuterium excess transitions occur in a single rapid step of less than a decade and are generally more abrupt than the corresponding isotopic transition. To assess regional character of abrupt climate change, we use deuterium excess transitions, in conjunction with volcanic signals, as reference points to cross date the ice cores. Once aligned, examination of the common variance of the NGRIP and GISP2 isotopic and deuterium excess time series shows remarkable coherence in one or both of the parameters in many of the finest details. This indicates that additional interpretable climate information is available with these detailed measurements and that each event may have some unique character. For example, NGRIP and GISP2 share a very similar deuterium excess transition at time around the Bolling warming, but the corresponding isotopic shifts between the two cores are markedly different. In contrast, at the end of the Younger Dryas, NGRIP and GISP2 isotopic shifts are very similar while the deuterium excess series share a sharp shift but are otherwise less similar. Meanwhile, the shape of isotopic transitions, accumulation rate patterns, and timing of climate changes relative to a suite of chemical species changes in the ice can vary from one abrupt event to the next. The emerging picture illustrates that not all abrupt climate transitions may be regionally uniform across Greenland, nor can any single sequence of events describe all abrupt climate changes, thus reducing their predictability. In the absence of a typical abrupt transition, we discuss possible forcing mechanisms in the context of different environmental boundary conditions related to ice volume, insolation, and Heinrich Events.
PP31C-03
The 1452 A.D. Kuwae Eruption Signal Derived from Multiple Ice Core Records: Greatest Eruption over the Past 700 Years
We combined 20 ice core records, 10 from the Northern Hemisphere and 10 from the Southern Hemisphere to determine the timing and magnitude of the Great Kuwae Eruption in the mid-15th century. The volcanic deposition signals were extracted by applying a high pass loess filter to each time series and examining peaks that exceed twice the 30-yr running median absolute deviation. By accounting for the dating uncertainties associated with each record, these ice core records together reveal a large volcanogenic acid deposition event during 1453-1457 A.D. The results suggest only one major stratospheric injection from the Kuwae eruption and confirm previous findings that the Kuwae eruption took place in 1452, which may serve as a reference to evaluate and improve the dating of ice core records. The average total sulfate deposition from the Kuwae eruption was 103 kg SO$_{4}$/km2 in Antarctica and 44 kg SO$_{4}$/km2 in Greenland. By applying the same technique to the other major eruptions of the past 700 yr, our result suggests that the Kuwae eruption was the largest stratospheric sulfate event of that period, probably surpassing the total sulfate deposition of the Tambora eruption of 1815, which produced 65 kg SO$_{4}$/km2 in Antarctica and 49 SO$_{4}$/km2 in Greenland.
PP31C-04
Weakening of the South Asian Monsoon Since 1600 A.D. from a Mt. Everest Ice Core
A 108m high-resolution ice core drilled to bedrock from the East Rongbuk Col of Mt. Everest covers the last 1500 years. Analysis indicates a weakening of the Asian monsoon since 1600 AD. Empirical orthogonal function (EOF) analysis on the major ion (Na$^{+}$, K$^{+}$, Mg$^{2+}$, Ca$^{2+}$, Cl$^{-}$, NO3$^{-}$, SO$_{4}$$^{2-}$) and δD timeseries demonstrates that the Everest site is influenced by continental air masses associated with the Asian high pressure system and marine air masses associated with the South Asian monsoon. Everest Ca$^{2+}$ is positively correlated with NCEP surface pressure from May-September over central Asia from 1948-2001, and Cl$^{-}$ is negatively correlated. This indicates that when pressure is relatively higher over central Asia more continental air masses penetrate the Everest site, and conversely lower pressures is associated with more marine air masses. An inverse correlation of Everest δD and June-September NCEP precipitation rate from 1948-2001 confirms the results of previous studies that the amount effect is the primary control on δD in the Everest region (Kang et al., 2002, Tian et al., 2001; 2003). A trend towards increasing continental and decreasing marine air masses since 1600 suggests higher pressure over central Asia, and a reduction in monsoonal influence at the Everest site. Increasing δD and decreasing net mass balance since 1600 provides further support for a weakening of the South Asian monsoon over the Everest region. Proxy records from low elevation sites south of the Himalayas indicate a strengthening over the same time period (Anderson et al. 2002; von Rad et al., 1999; Wang et al., 2005). These regional differences are associated with higher pressure over Asia during summer and a southward shift in the mean summer position of the Intertropical Convergence Zone, as supported by a reduction in seasonal differences of insolation.
PP31C-05
Evidence from an Ice Core of a Large Impact Circa 1443 A.D.
Published data on melt water from the Siple Dome ice core show distinct anomalies at 1443.16 A.D. The Ca value is 111 ppb, over 9 times the next highest Ca value between 850-1760 A.D. The K value is 20 ppb, about 1.4 times the next highest K value. The Ca anomaly may be due to partial dissolution of CaCO3 microfossils from the 24 km Mahuika bolide impact crater on the southern New Zealand shelf. Deep-sea samples of the Mahuika ejecta layer contain >98% carbonate microfossils. The Mahuika impact may have produced tsunami runups of 130 meters in Jervis Bay, Australia. The Australian megatsunami deposits date to 1450±50 A.D. We analyzed the melt water from 8 ice-core samples from the West Antarctic Siple Dome ice core that date from 1440-1448 A.D. The 1443 A.D. level contained a peak in K of 53 ppb as compared to a background of $~$6-7 ppb. Ca was high at 26 ppb but this is not as pronounced as reported earlier. We extracted solid material from the melt water. Except for the 1443 A.D. horizon and one fractured grain at the 1442 A.D. level, most samples were barren except for typical dust. At the 1443 A.D. level, we found 5 carbonate microfossils (coccoliths?) from 5 to 20 microns across. Two were round and solid. One microfossil appeared either caught during mitosis or broken during deformation and elongation. Another carbonate microfossil was unbroken, but appeared deformed into a square. We found a Cu grain with a small amount of oxygen. It is most likely a grain of native copper with an oxidized surface. Deformed microfossils and native minerals are both characteristic of bolide impacts. We also found many microcrystalline magnetite cubes, with an average crystal size of 0.3 microns or less. The high magnetic susceptibility of impact ejacta layers is caused by microcrystalline magnetite. We found a grain of conchoidally fractured feldspar $~$15 microns long. A semi-quantitive EDAX analysis found 21% Si, 55% O, 9% Al, 5% Na, 3% K, 2% Fe, and 1% Ca (atomic %), well within the range of K-feldspar compositions. Because Fe does not fit into the feldspar structure, its occurrence implies either that the Fe-bearing feldspar is a glass, or that the Fe is in microcracks within the grain. As ice is not Fe-rich, the former is more likely. Because conchoidal fracture is characteristic of glass, this suggests that the feldspar is a glass (maskelynite) derived from an impact onto continental crust. We also found Al Fe oxide, Ti Al oxide, and amphibole. A semi-quantitative EDAX analysis of the latter found 53% O, 20% Si, 5% Na, 4% Al, Mg, and Fe, 3% Ca, and 0.5% K (atomic %) with trace Ti, S and Cl, close to the composition of the alkali amphibole richterite, which forms in contact metamorphosed limestones (skarns). The Al Fe oxide is most likely hercynite, a spinel that forms in contact metamorphic aureoles in silica-poor environments. All mineral grains had distinct edges. We also found radiating, fibrous crystals of a Ca Na silicate. An EDAX analysis of the mineral found 59% O, 13% Ca, 8% Si, 3% Na, and 1% Mg (atomic %). The Ca Na silicate is most likely pectolite (NaCa2Si3O$_{8}$), which has radiating, fibrous crystals and forms in skarns. The presence of minerals characteristic of contact metamorphism is important as we have found abundant skarn facies minerals in the Mahuika ejecta layer within deep sea sediment. Thus, our data taken together are most consistent with an impact ejecta layer within the Siple Dome ice core that comes from the Mahuika impact event about 4044 kilometers away; providing a well-constrained date for the event around 1443 A.D.
PP31C-06
High-Resolution South Pole Dust Log and Evidence of Low Sulfur Volcanic Eruptions
As part of the commissioning of the ICECUBE Neutrino Observatory, an improved dust logger was deployed to make sub-centimeter scale measurements of ice optical properties most of the way down a 2400m South Pole borehole. These measurements, covering approximately 70,000 years, provide the highest resolution record of dust accumulation now available and also record a number of distinct volcanic ash horizons. During the last glacial period, subtle variations in the dust accumulation at South Pole appear well correlated to rapid climate changes in Greenland (i.e. Dansgaard-Oeschger events). With improved timing information, high-resolution records such as these should be able to answer the question of whether climate changes in the South typically precede or follow climate changes in the North. The most likely source of high precision timing information will be the identification of synchronous volcanic markers at both North and South poles. Existing and forthcoming work relying on sulfate concentrations has already allowed some matching volcanic horizons to be found at both poles. However, the dust logger, with its ability to detect ash layers so faint as to escape unaided recognition, opens up a complementary method in the search for correlated volcanic events. One of the most remarkable findings from this new log is the presence of volcanic events which appear to have simultaneously deposited ash at South Pole and GISP2 (prior work by us), but for which there are no reports of a global volcanic sulfate spike. This suggests that it is possible for a volcanic eruption to be sufficiently explosive as to globally distribute ash without generating the type of sulfate pulse relied upon in most previous studies of volcanism in ice cores. If such apparently contemporaneous ash layers truly come from a single eruption, then we would have no choice but to conclude that studies of volcanism relying on sulfate alone are an inherently incomplete record of major eruptive events.
PP31C-07
Interplanetary Dust Particles in the EPICA Dronning Maud Land Ice Core
Polar ice cores and marine sediments provide an archive of the influx of extraterrestrial material in the form of interplanetary dust particles. Interplanetary dust particles are marked by extremely high helium concentrations, and this signal can be used to reconstruct a record of the extraterrestrial Helium-3 flux. Determination of accurate extraterrestrial Helium-3 fluxes from marine sediment cores is limited by the uncertainties in the accumulation rates of marine cores. Ice cores, with their very well known age models and accumulation rates, provide a unique opportunity for accurate and independent estimate of the IDP-related Helium-3 flux. Previous measurements of helium isotopes in samples from the Vostok ice core (Brook et al., GRL, 2000) demonstrated the suitability and potential of ice cores and found Helium-3 fluxes similar to results from marine sediments (e.g. Marcantonio et al., Paleoceanography, 2001). A limiting factor for ice core studies so far has been the access to ice core material and, related to this, the size of the available samples. In order to overcome this limitation, we developed a new extraction method based on filtering dust particles from the "waste water" stream from the melt head of a continuous flow analysis for aerosol chemistry (Roethlisberger et al., 2000). This gives access to bulk samples allowing in principle to analyze a continuous record of Helium-3 accretion and to integrate over longer timescales (500a) thereby improving the particle statistics and thus the over-all precision. We report results of helium isotope measurements of a high-resolution record from the EPICA (European Project for Ice Coring in Antarctica) Dronning Maud Land ice core using 4-5kg samples from 400 to 1100 m depth corresponding to the time period from the LGM to the middle Holocene. We also present a pilot study comparing results from the new sampling technique to measurements on real ice samples.
PP31C-08
Sub-annual Ice-Core Record of Major Ion and Heavy Metal Variability and Sources in the North Pacific Free Troposphere, Mt. Logan, Yukon, Canada
The Mt. Logan, Yukon, Canada summit plateau (PR Col; 5300 m.a.s.l.) ice core has been continuously sampled at high resolution (2-3 cm/sample) by a novel ice core melting system with discrete sampling, and analyzed for 8 major ions and 35 trace elements. Co-registered, sub-annual timeseries covering the past 500 years reveal seasonal aerosol fluctuations dominated by dust, with sea-salt contributing less than 5% of sulfate and calcium concentrations. Dating of the top 500 years of the record is by annual layer counting. Concentration spikes of sulfate greater than three times the standard deviation (60 ppb) above the mean (75 ppb) correspond in time with historical explosive volcanic eruptions. Sulfate spikes corresponding in time with large (VEI>4) historical Alaskan eruptions, including Katmai (1912) and St. Augustine (1986), are commonly associated with concentration and crustal enrichment factor spikes in lead, cadmium, antimony, copper, zinc, bismuth and thallium an order of magnitude above background (non-volcanic event) values. Sulfate spikes corresponding in time to large eruptions from distant volcanoes, including Agung (1963), do not show a corresponding rise in heavy metal concentrations. Apart from the periodic spikes in concentration, heavy metal timeseries largely mirror those of the major dust species (Al, Fe), but maintain significantly elevated crustal enrichment factors, probably due primarily to quiescent degassing of volcanoes. Such datasets are necessary to understand the cycling of heavy metals in the free troposphere, including the relative source strength of explosive eruptions vs. quiescent volcanic degassing, and the relative strength of natural vs. anthropogenic sources.