MIT OpenCourseWare http://ocw.mit.edu 12.740 PaleoceanographySpring 2008For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.1PALEOCEANOGRAPHY 12.740 SPRING 2006 Lecture 7 Ice core evidence for climate change I. Polar ice caps: characteristics and flow behavior Vertical cross section of an ice sheet resting on a horizontal subsurface. Ice particles deposited on the snowsurface will follow lines that travel closer to the base the farther inland the site deposition. An ice mass formed around the divide (I-I) will be plastically deformed (thinned) with depth as suggested by the lined areas. Thedashed curve along the vertical ice core (C-C) shows the calculated horizontal velocity profile Vx (Weertman 1968b). The horizontal arrows along C-C show the adopted approximation to Vx (Dansgaard et al. 1969).yISurfaceCFlow linesxC IBed rock Figure by MIT OpenCourseWare. Adapted from source: Dansgaard et al. (1971). Note: this figure does not illustrate bedrock depression A. Total ice mass on earth today: 29 x 106 km3, equivalent to ~80m of sea level rise (but some of this would fill in space filled by sub sea-surface ice).. B. Physical Properties and Transitions 1. Depth transitions a. Snow-->firn-->ice2 b. Bubble closure and compression: appearance of air hydrate inclusions 1000-1300m; bubbles disappear completely 1500-1600m; c. Brittle ice zone (800-1200m); ice often fractures upon return to surface d. deep stable ice 2. Bubbles reform upon return of deep ice to surface 3. Exclusion of soluble salts from ice crystals and grain-boundary H2SO4 Catm, globalLarge-scale transport atmospheric circulationCatm, localWet deposition turbulenceair-snow partitioningDensity Age(kg/m3)(yr)CsnowRedistribution200 0,5400 1550 20CfirnSmoothing and decay800 100840 120CiceDecayProcesses and steps involved in transfer function, which relates concentrations in ice to those in the global atmosphere. Depth and age scales are for Greenland. Snow-to-firn transition is defined by metamorphism and grain growth; firn-to-ice transition is defined by pore closure.Figure by MIT OpenCourseWare based on Neftel, et al., 1995.3 I I. Chronology A. Annual counting 1.In the upper part of the ice, annual variations in O and H isotopes can be used to count annual layers. As the ice gets older, molecular diffusion blurs the cycles and they become ambiguous, hence limiting O18 cycle counting to the upper portion of the core (~1000 years or so, depending on accumulation rate). At low accumulation rates (e.g. South Pole), annual cycles are not at all useful; at higher accumulation rates (e.g. Dye 3), annual δ18O cycles can be discerned back as far as 3,000 years. 2. Other indicators can show seasonal cycles: a. dust b. chemical constituents (major ions) c. physical properties, such as electrical conductivity d. summer "hoar frost" formation (visually apparent on a light table) e. Since these properties do not diffuse (significantly), they can record older layers than can δ18O. 3. Any annual counting method will have some ambiguities that may lead to slight over-and under- counts. B. Flow models. Based on approximations of the physical equations driving ice flow. These may be decent, but they depend on a good knowledge of boundary conditions and their temporal evolution. These work best when used with chronological spikes deep in the record – the model helps “interpolate” between the chronological spikes.4C. Correlation with other climate records 1. Climate record correlations 2. Gas correlations a. CO2 b. CH4 c. δ18O2 D. Direct dating methods 1. In principle, it should be possible to date the CO2 in the ice bubbles by AMS 14C. In reality, no one has reported a successful 14C date. One problem is that cosmic rays striking the ice convert some of the oxygen to carbon 14 (D. Lal). E. Other methods 1. Volcanic ash layers 2. Acidity spikes from volcanic eruptions 3. U-series dating of recoil products (Fireman) I II. δ18O and δD evidence for T changes A. Stable isotope hydrology 100O in Precipitation (per mil)-10-20-30-4018-50δ-60-60 -40 -20 0 20Surface Temperature (C)Observed δ18O versus observed T (annual mean). The annual means for the IAEA [1981a] sites are computed from monthly means through precipitation weighting.Figure by MIT OpenCourseWare based on Jouzel, et al., 1987.5B. "Silty" ice near bottom - problematical δ18O. C. Camp Century Ice Core 1. Time scale is based on annual cycles of δ18O for the first millenium. 2. Below that level, time scale is based on flow model and on correlation with other climate records. Note big surprise awaiting on deep Camp Century time scale! 3. Glacial/interglacial climate signal; Younger Dryas; interstadials Figure removed due to copyright considerations. Please see: Figure 6 in Dansgaard W., S. J. Johnsen, H.B. Clausen, andC.C. Langway J. “Climatic record revealed by the Camp Century Ice Core.” In Late Cenozoic Ice Ages. Edited byK. K. Turekian. Yale University Press, 1971, pp. 37-56.6 Figure removed due to copyright considerations. Please see: Figure 9 in Dansgaard W., S. J. Johnsen, H.B. Clausen, andC.C. Langway J. “Climatic record revealed by the Camp Century Ice Core.” In Late Cenozoic Ice Ages. Edited byK. K. Turekian. Yale University Press, 1971, pp. 37-56.7D. Byrd Ice Core: high resolution Antarctic record E. Dye-3 ice core 1. Confirmation of Younger Dryas, interstadials Image removed due to copyright considerations.82. New time scale assigned to Camp Century core Image removed due to copyright considerations.Source: Dansgaard et al. (1982).9F. Vostok ice core Image removed due to copyright considerations.Source: Jouzel et al. (1987).Image removed due to copyright considerations.Source: Lorius et al. (1985).10 Figure removed due to copyright considerations. Please see: Figure 3 in Petit, et al. Nature 399 (June 3, 1999): 431. G. Renland ice core (southern Greenland11H. GRIP, GISP2 ice cores in summit, Greenland Figure removed due to copyright considerations. Grootes P. M., M. Stuiver, J. W. C. White, S. Johnsen, andJ. Jouzel. "Comparison of oxygen isotope records from the GISP2 and GRIP Greenland ice cores." Nature 366 (1993) 552-554. Please see:12 GRIP/GISP2 problems: boudinage, basal ice folding J. EPICA Dome C Figure