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 3 Oxygen-isotope paleothermometry I. 1940's: Kullenberg's invention of the piston core. Previously, deep-sea coring was done by gravity coring, where a pipe with an in-line stopper at the top and "core catcher" at the bottom (interleaved flexible metal fingers) was dropped into the seafloor. Friction against the walls of the pipe limited the length of core that could be obtained. Modern gravity cores can get cores ~ 5 m in length; older ones obtained only a meter or two. Kullenberg introduced a piston into the pipe; the piston was rigged with a cable that prevented it from moving down with the pipe, creating "suction" which helped force the sediment into the pipe as it moved down into the sediment. Modern piston cores can obtain as much as ~50 m of sediment (although more typically, ~20 m). II. 1950's: Emiliani A. Using Urey's mass spectrometer, which required about 5 mg of calcium carbonate (about 100-200 individuals of foraminifera), Emiliani analyzed a variety of species of planktonic foraminifera from the Caribbean and found an apparent depth stratification, with some species (G. sacculifer, G. ruber) recording isotopic temperatures close to that of surface seawater; others recording colder temperatures. B. Emiliani analyzed downcore records of apparent surface dwellers throughout the Atlantic; making a correction for changes in the isotopic composition of seawater (more about this later), he calculated a 6-8° decrease in tropical ocean surface temperatures during glacial periods (i.e. about 1.5‰ δ18O increase). He found evidence for many glacial/interglacial cycles over the last half million years; he coined the isotope stage stratigraphy system (now commonly referred to as "MIS" (Marine Isotope Stage); and he argued that the data supported the Milankovitch mechanism of climate change.2 1. Emiliani developed an isotope stage numbering scheme, based on periods of warmer (odd) and colder (even) that he could recognize reliably in his records (smaller changes were ignored). 2. This work created quite a stir, and was quickly criticized on several grounds: a. It violated the prevailing 4-ice-age theory from continental stratigraphy. b. Meteorologists though that the tropical temperature change seemed excessive. c. Micropaleotologists found discrepancies between their initial micropaleontological work (G. menardii stratigraphy) and the down-core O-18 record. d. Biologists (e.g. Bé) argued that foraminiferal ecological shifts may have altered the depth habitat of organisms (and hence temperatures). e. The time scale (based on 230Th/231Pa) was criticized. f. Various statistical errors were pointed out. 3. Despite all of this criticism, with 1996 hindsight we can say that Emiliani was right about most of these points, with the major exception being the time scale and the amplitude of tropical cooling. • the problem of the time scale: Emiliani derived his time scale from a core which had been dated by the "231Pa/230Th" method. This method assumes that the initial Image removed due to copyright considerations.3value of the ratio at zero age is the production rate from the 235U and 234U in seawater. We now know that this assumption is incorrect: Th and Pa are fractionated by their differential particle reactivity in the ocean. III. Oxygen isotope hydrology A. As it turns out, the major problem with Emiliani's interpretation is his correction for the change in the isotopic composition of seawater. B. Oxygen (and hydrogen) isotopes in the atmosphere and precipitation. 1. Water vapor in equilibrium with water has δ18O ~ -9-11‰, varying slightly with temperature: α =1.01113 - 1.06x10-3 T + 2.88x10-7 T2. 2. If an air mass initially equilibrated with δ18O = 0 ‰ water is then isolated, cooled, and the resulting condensation continuously drawn off, the vapor becomes progressively more depleted in 18O: exponential dependence of vapor pressure on temperature and Rayleigh (1896) fractionation: RRi= fαT()−1() where T = temperature α (Τ) = Rliq/Rvap isotope fractionation (slightly temperature dependent) f = the fraction of the initial water vapor remaining (set by the equilibrium vapor and R/Ri = 18O/16O ratio of the remaining water vapor compared to its initial value4 Fraction Remaining Vaporδ 18O in cloud vapor and condensate plotted as a function of the remaining vapor in the cloud for a Rayleighprocess. The temperature of the cloud in degrees Celsius is shown on the lower axis. δ 18O values are relativeto SMOW. The increase in fractionation with decreasing temperature is taken into account. After Dansgaard (1964).Vapor20 1015 0 -20-30-25-20-15-10-51.0 0.750.500.25 0Cloud Temperature OCCondensateδ 18O (SMOW)11%9%3. Similar fractionation occurs for hydrogen isotopes, except that since the m ass difference is proportionately larger, the isotope fractionation for D/H is larger (by about a factor of 8) 4. Since most evaporation occurs in the tropical and subtropical ocean (highest vapor pressures) and since most precipitation results from the transport of this tropical water vapor towards cooler polar regions, there is a strong correlation between average annual temperature and the average annual δ18O. Mean annual δ 18O of precipitation as a function of the mean annual air temperature atthe earth's surface. Note that δ 18O values are progressively lighter as the mean annualtemperature becomes lower. After Dansgaard (1964).Mean Annual Air Temperature OCδ 18O (SMOW)-50 -40 -30 -20 -10 0 10 20 30-50-40-30-20-100South Pole85o S Horlick MTNS.N. GreenlandS. Greenland75o N Upepnavik71o N UmanakGoose Bay, LabradorCopenhagen61o N GrennedalDublinValentiaGough Is.Barbados Is. 3 Figure by MIT OpenCourseWare.Adapted from source: Broecker (1974)Chemical Oceanography.Figure by MIT OpenCourseWare.Adapted from source: Broecker (1974)Chemical Oceanography.5 5. Although Emiliani knew that modern ice was relatively light and had an isotopic composition of about -25‰, he argued that the bulk of snowfall during the glacial maximum (where ice is accumulating at lower latitudes) must have an isotopic composition of about -15 ‰, because this is the composition of precipitation