Ice Sheet Paleoclimatology

Ice Sheet Paleoclimatology Climate Physics and Chemistry Transformation of snow into ice Catm, global Catm, local Density (kg/m3) Age (yr) 200 0,5 400 1 550 20 800 100 840 120 Large-scale transport atmospheric circulation Wet deposition turbulence air-snow partitioning Csnow Redistribution Cfirn Smoothing and decay Cice Decay Processes 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. Observed δ18O - surface temperature relationship 10 δ18O in Precipitation (per mil) 0 -10 -20 -30 -40 -50 -60 -60 -40 -20 0 20 Surface 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. Folding near the base of the Greenland summit ice cores (I) (II) Ice Flow B B Depth A Light A Heavy Isotope Composition I) A typical shear fold in the basal part of a glacier. If the ice had not been previously folded, a stratigraphic sequence of a climate-related property (such as oxygen isotope composition) sampled by a borehole at point A might produce a simple monotonic trend as shown by line A in part II. Sampling at B, after folding, would yield the sequence shown as B in II. Multiple folding can complicate the sequence further. Abrupt climate swings during the past 100,000 years: the Bolling-Allerod, Younger Dryas, and “stadial/interstadial” “Dansgaard-Oeschger cycles • Between 10,000-65,000 years ago, there were at least 17 abrupt swings between warmer and colder climate events. • These events were first observed in the Greenland ice cores, but they have now been seen at diverse sites in the Northern Hemisphere including the tropics. • These events are not observed in the Antarctic ice cores, save possibly in dampened form. Image removed due to copyright restrictions. Citation: Grootes, P. M., Stuiver M., White J. W. C., Johnsen S., and Jouzel J. “Comparison of Oxygen Isotope Records from the GISP2 and GRIP Greenland Ice Cores.” Nature 366 (1993): 552-554. The Bolling/Allerod warming and Younger Dryas cooling Figure from GISP2 Newsletter using data from M., Grootes P., Stuiver M., White J. W. C., Johnsen S., and Jouzel J. "Comparison of Oxygen Isotope Records from the GISP2 and GRIP Greenland Ice Cores.” Nature 366 (1993): 552-554 and Alley, R. B., et al. Nature 362 (1993): 527-529. Upper figure using data from M., Grootes P., Stuiver M., White J. W. C., Johnsen S., and Jouzel J.“Comparison of Oxygen Isotope Records from the GISP2 and GRIP Greenland ice cores.” Nature 366 (1993): 552-554, and lower figure using data from J., Behl R., and Kennett J. P. “Brief Interstadial Events in the Santa Barbara Basin, NE Pacific, During the Past 60,000 Years." Nature 379 (1996): 243-246. Estimated calendar Age, kyr BP Gases in Ice Cores • Bubbles seal off at the bottom of the firn layer, ~80120 m • Hence gas is younger than the solid ice that contains it - the “gas age/ice age difference” depends on the accumulation rate • Most gases are well mixed in atmosphere; so records from Antarctic and Greenland are nearly the same; features of the records can be used to correlate chronologies between hemispheres • Gases that have been measured: – – – – CO2 O2 (18O/16O ratio) CH4 N2 O Relic paleotemperatures from borehole temperatures • Because heat diffuses through ice at a limited rate, the interior of the ice sheets is still colder than at the surface, a relic of last glacial maximum cold conditions. • Given an accurate model of advection and diffusion, one can estimate what the original temperature was from a model. • Time resolution becomes poorer further back in time (diffusional smoothing). Borehole temperature profiles in central Greenland Borehole inversions imply that Greenland summit LGM temperature was -15°C colder than at present - twice the difference predicted from δ18O. Why? • Was the slope of the δ18O-T relationship 0.45 rather than 0.65? (Why?) • Did the δ18O relationship retain the same slope but shift its intercept? (This would be expected if source water temperatures were colder.) • Did snowfall not accumulate in central Greenland in winter during the LGM? (If so, then the δ18O of the ice only reflects the summer temperatures; this suggestion, supported by a GCM model, is taken as a result that very cold temperatures limit the amount of snowfall.)