HR: 14:10h
AN: V23D-03    [Abstracts]
TI: Chemical and Isotopic Variability of MORB and Implications for Upper Mantle Mixing Times
AU: * Goldstein, S L
EM: steveg@ldeo.columbia.edu
AF: Lamont-Doherty Earth Observatory of Columbia University, 61 Rt. 9W, Palisades, NY 10964 United States
AU: Langmuir, C
EM: langmuir@eps.harvard.edu
AF: Harvard University, 20 Oxford St., Cambridge, MA 02138 United States
AU: Class, C
EM: class@ldeo.columbia.edu
AF: Lamont-Doherty Earth Observatory of Columbia University, 61 Rt. 9W, Palisades, NY 10964 United States
AU: Donnelly, K
EM: donnelly@ldeo.columbia.edu
AF: Lamont-Doherty Earth Observatory of Columbia University, 61 Rt. 9W, Palisades, NY 10964 United States
AU: Jones, K
EM: kjones@ldeo.columbia.edu
AF: Lamont-Doherty Earth Observatory of Columbia University, 61 Rt. 9W, Palisades, NY 10964 United States
AB: The concept of "mantle isochrons" was developed in the 1970s in classic papers from the Carnegie and Stony Brook groups, focusing mainly on ocean island basalts. They suggested that positive correlations of isotope and parent-daughter element ratios of oceanic basalts can be used to directly estimate the ages of the mantle sources, which has generated controversy ever since. Important contributions on relationships between isotopic and chemical variability, and mantle source ages, continued to be made through the 1980's and 1990's, especially from the Paris group. However, geochemists have mainly interpreted the isotopic variability in oceanic basalts to reflect mixing of unrelated mantle sources, often without direct age implications. Stan Hart has been strongly associated with both ideas. Recently, Albarede EPSL 2001 and Donnelly et al. EPSL 2004 discussed relationships between isochron ages and reservoir residence times for simple systems undergoing melting and convective mixing. The latter paper concluded, based on Rb-Sr, Sm-Nd, and U-Pb "mantle isochron" ages from a few ridge segments from the Atlantic, Pacific, and Indian oceans, that the average upper mantle mixing time is ~300 Ma, and that this reflects the average lifetime of E-MORB sources. Here we test how well this conclusion holds for global MORB using data from the PetDB database. The isotopic and chemical heterogeneity of MORB result in large scatter on isochron diagrams for Nd, Sr, Pb, and Hf isotopes for the global data and for individual ocean basins, which make interpretation difficult. We use a different approach to extract age information from the isotopic variability. A sample's Nd isotope ratio reflects the average Sm/Nd ratio seen by that Nd over geological time. Thus a large isotopic range reflects large parent-daughter chemical variability, and vice versa. We compare the dispersion of time-averaged parent-daughter ratios (that is, as implied by isotope ratios), to the actual chemical dispersion of the parent-daughter ratios in MORB. In global MORB, the "time-averaged" parent-daughter ratios show a very narrow range compared to the chemical variability (in Pacific MORB the time-averaged 147Sm/144Nd shows a small range of 0.20-0.22 while the chemical range is 0.14-0.29). This factor of 7 difference in variability reflects the competing effects of melting (which generates chemical heterogeneities) and convective mixing (which destroys them), and survival time (which allows development of isotopic heterogeneities). Because partial melting to form ocean crust has been occurring in more or less the same way through time, the large chemical dispersion that it generates should remain similar through time, and thus this huge difference between the time-averaged and the chemical parent-daughter dispersion reflects the efficiency of mantle convective mixing. A shorter average mixing time requires a greater range of "time-averaged" parent-daughter ratios to produce the variability of isotope ratios in MORB. We find that for global MORB, and for individual ocean basins, the time-averaged parent-daughter ratios and the chemical dispersion show the best agreement if the average mixing time is between 200-400 Ma, strongly indicating that the average survival time for enriched heterogeneities in the upper mantle is a few hundred million years.
DE: 1000 GEOCHEMISTRY
DE: 1025 Composition of the mantle
DE: 1032 Mid-oceanic ridge processes (3614, 8416)
DE: 1038 Mantle processes (3621)
DE: 1040 Radiogenic isotope geochemistry
SC: Volcanology, Geochemistry, Petrology [V]
MN: Fall Meeting 2005