Paleoceanography - Ice Core Evidence for Climate Change

1 PALEOCEANOGRAPHY III. Atmospheric gas record in ice cores A. Methodological issues; firn/ice transition; age of air; gravitational fractionation; bubble compression and relaxation; gas extraction; reactions with ice and/or water or solids; impurities. 1. Firn/ice transition: depth correlated with temperature due to effect of T on pressure sintering: Adapted from source: Craig and Wiens (1996). 2. "Gravitational equilibrium for isotope and perfect gas ratios is described by the Gibbs equation:" (Craig and Wiens, 1996) ⎡ gz(∆M) ⎤ R = exp⎢ ⎣ RT ⎥⎦ Ro e.g. in a 100m diffusive firn layer, 84Kr should be enriched over 36Ar by 1.28%, 14 enriched over N by ~0.4‰ 15 N is The driving processes are a balance between gravitational forcing, forcing heavier isotopes to underlay lighter isotopes, and random molecular diffusion, working against the gradient established by gravity. The result can be derived from the barometric equation P = Po exp[ Mgz ] RT describing the pressure of a gas above the surface of the earth that would be observed if molecular diffusion was the dominant mode of vertical transport [i.e., no turbulent diffusion, as seen in the atmosphere](Dalton, 1826; Gibbs, 1928). 2 3. Thermal diffusion: during a sudden warming, where the surface is warmer than the bottom 15 40 of the firn layer, the cold bottom end is enriched in heavier isotopes: δ N = αN∆T and δ Ar = αAr∆T where αN and αAr are the thermal diffusion coefficients for N2 and Ar respectively. 4 4. Methods of gas extraction: melting/freezing cycles for CO2, N2O; needle-crushing for CO2. B. CO2 1. Vostok ice core CO2: 8 C. δ18O2 1. Dole Effect: δ18O2 of atmosphere is +23.5‰ relative to SMOW. a. Photosynthesis: H2O + CO2 = O2 + CH2O δ18O2(photo) = δ18O(water) + A (kinetic isotope effect during photosynthesis) where δ18O(water) = δ18O(ocean) + W (where W is the weighted mean difference between the isotopic composition of the ocean and the water immediately used for respiration) b. Respiration: O2 + CH2O = H2O + CO2 δ18O2(resp) = δ18O2 + B (respiratory kinetic isotope fractionation) c. At steady-state, δ18O2 - δ18O (ocean) = W + A - B 2. Gross Productivity and atmospheric oxygen residence time source: Bender et al. (1985) 3. Terrestrial O2 and δ18O2 mass balance 9 source: Bender et al. (1994) 4. Marine O2 and δ18O2 mass balance source: Bender et al. (1994) 5. Byrd ice core 6. Vostok ice core 15 E. O2/N2: relation to local insolation, use for time scale development. Image removed due to copyright considerations. Please see: Bender. EPSL 204 (2002): 275. F. N2O IV. Other tracers in ice cores: dust (note Sr+Nd isotope work), 10Be, volcanic ash, chemicals, crystal size in ice cores; atmospheric pressure of bubbles A. Sea-salt, volcanic acid, and dust in Greenland ice cores 18 Image removed due to copyright considerations. Source: Hammer et al. (1985). B. Seasalt and dust in the Vostok ice core 19 10 Al ng g-1 50 100 150 δ18O -55 100 -60 50 0 0 500 1,000 1,500 2,000 Depth m Al concentration as a function of real depth (green line). The climatic reference is given by δ18O (yellow line). The lower dashed line represents the estimated background. 20 10 δ18O 100 150 -55 Nam ng g-1 -60 100 0 50 Age 103 yr BP 0 500 1000 Depth m 1,500 As previous figure, with Nam instead of Al. 2,000 21 B. 10Be Image removed due to copyright considerations. C. Calcium in the GISP2 ice core: relation to "1500 year" climate cycle and Dansgaard/Oeschger events 22 Image removed due to copyright considerations. Source: Mayewski et al. (1997). Image removed due to copyright considerations. Source: Mayewski et al. (1997). E. Methanesulfonic acid (dimethyl sulfide product) 1. Marine organisms produce DMSP (dimethylsulfoniopropionate: 23 CH3 \ S+-CH2-CH2-COO/ CH3 this is converted to DMS [(CH3)2S] when they are munched up. 2. DMS is volatile and goes into the atmosphere 3. DMS is oxidized in the atmosphere to two byproducts with a "branch ratio": sulfuric acid (H2SO4) and methanesulfonic acid (MSA: CH3SO3H). Cycle is complex with many intermediates; branch ratio appears to depend mainly on temperature (low MSA:nssSO4= at warmer temperatures) 4. The products are transported to the ice and recorded there as non-sea-salt sulfate (nss SO4=) and MSA. 24 -420 a 13 Age (kyr) 56 66 33 141 107 118 ) -460 �D ( -440 -480 -500 A -520 MSA (p.p.b.) 50 B C D E F G H b 40 30 20 10 0 n.s.s. SO42- (p.p.b.) 300 c 200 100 MSA/n.s.s. SO42- (%) 0 20 d 15 10 5 0 0 500 1,000 1,500 2,000 Depth (m) Depth profiles of a, �D, b, MSA, c, non-seasalt sulphate and d, the ratio of MSA to non-seasalt sulphate in the Vostok ice core. Upper and lower stage boundaries are marked by vertical lines. (�D is deuterium isotope profile) 25 source: Legrand et al. (1991) F. Chemicals: other major ions and trace elements d. GRIP borehole Monte-Carlo simulations Dahl-Jensen et al. (1998) e. How to resolve this discrepancy? i. Seasonality of precipitation (less snow in LGM winter)? ii. Shifted O18-T relationship due to cool tropics? H. Other indicators 1. Nd-Sr isotopes: argues that Greenland LGM dust came from Asia Image removed due to copyright considerations. Source: Grousset, Biscaye et al. V. The last 1000 years: Little Ice Age and Medieval Warm Period 26 Image removed due to copyright considerations. Please see: Figure 3 in Dansgaard W., S. J. Johnsen, H. B. Clausen, and C.C. Langway J. "Climatic record revealed by the Camp Century Ice Core." In Late Cenozoic Ice Ages. Edited by K. K. Turekian. Yale University Press, 1971, pp. 37-56. VI. Rapid climate change in ice cores: Younger Dryas, interstadials, etc. A. Younger Dryas and Bolling-Allerod B. Interstadials C. Synchroneity of rapid climate events VIII. The ice core time scale: relative and absolute stratigraphy, accuracy and precision A. The players 1. layer-counting (hoar frost layers, dust layers, O18 cycles, chemical signals 2. δ18O2 3. CH4 4. 10Be 5. correlations to other climate records B. The state of the art, 1998 1. Absolute chronology a. Ice core layer counting at GISP2 appears to be the winner for precision and accuracy for the past ~40,000 years. It is consistent with δ18O2 but offers more precision. Beyond that period, it becomes increasingly less objective and inaccurate. i. a possible competitor in this interval is the calibrated radiocarbon record linked to key climate events. The accuracy ultimately should be the same or better, but problems in sorting out atmospheric D14C variability and phase leads and lags make it more problematical. b. δ18O2 links us to the marine chronology prior to that time (the marine chronology being constrained by coral 230Th/U dates correlated to foraminiferal d18O and "orbital tuning" 2. Relative chronologies a. δ18O2 links ice cores between the two hemisphere with a relative precision of a few hundred to several thousands of years. b. detailed CH4 records can correlate ice cores between the hemisperes to several decades to several hundredds of years, given sufficient temporal resolution of the core and sampling. 27 c. 10Be provides one or two absolute spikes that can test the accuracy the chronology provided by the above methods, and perhaps allow for a quicker homing in. If, as now seems very likely, the 10Be spikes are linked to fluctuations in the earth's magnetic field, it may also allow a link to chronologies in marine sediments and continental materials. d. correlation to other highly resolved climate records (e.g. ice δ18O compared to gray scale of Cariaco Trench varved sediments, bioturbation index of Santa Barbara Nasin sediments) is useful, but carries inherent uncertainty regarding the phase relationships (some events may lead, others may lag, even though linked to the same system. For example: maximum summer warmth lags maximum incoming radiation by several months...) e. Volcanic ash: Ram and Gayley (1991) observed a volcanic ash shard layer 1950 m depth in the Dye 3 ice core which they suggested was the same as a marine ash layer observed by Ruddiman and Glover (1972) - hence linking the marine and ice core chronologies at ~57 kyrBP. f. Acidity links Indonesian marine record to central Greenland ice core? (can we be sure that this volcanic eruption was in fact Toba and not some other volcano?) 28 Image removed due to copyright considerations. Please see: Figure 1 in Zielinksi G. A., P. A. Mayewski, L. D. Meeker, W. Whitlow, and M. S. Twickler. "Potential atmospheric impact of the Toba mega-eruption ~71,000 years ago." Geophys Res Lett 23. (1996): 837-840.