Diagenetic fractionation of Ge and Si in reducing sediments: The missing Ge sink and a possible mechanism to cause glacial/interglacial variations inoceanic Ge/Si

The average Ge/Si ratio in the ocean is determined by the budgets for each of these elements. Previous budget formulations have assumed that: the only important sink for both elements is burial as opal, based on studies of th e Si cycle and the close oceanic coupling observed between inorganic Ge and Si distributions. However, these budgets implied two paradoxes: (1) hydrot hermal flow through ocean ridges is smaller than predicted by other tracers , and (2) the lower Ge/Si ratio of opal deposited during glacial times comp ared to that deposited during interglacial times required enhanced weatheri ng during cooler, drier climates. Both paradoxes could be resolved if a sig nificant sedimentary sink for Ce other than opal burial could be identified , and the objective of this study was to search for one. Two pore water pro files collected in Equatorial Pacific sediments show that Ge and Si behave similarly in the upper 10 cm of sediment, indicating no evidence for a sign ificant non-opal sink for Ge in oxic sediments. By contrast, profiles in se veral cores from the California Margin demonstrate that in reducing sedimen ts, Ge diagenesis is poorly coupled to Si diagenesis: significant Ge remova l is evident, both downcore and sometimes in the near-surface. Benthic flux chamber measurements at three continental slope stations, all with an oxic layer less than 1 cm thick and large iron gradients in near-surface pore w aters, showed that 55 +/- 9% of the Ge released by opal dissolution is sequ estered. However, at two locations with anoxic sediments but little pore wa ter Fe+2 in the upper 2 cm, flux measurements indicated little fractionatio n from the oceanic ratio during diagenesis, implicating the importance of i ron for fractionating Ge from Si during diagenesis. If the Ge sequestration observed in the iron-rich CA margin sediments is typical of all slope sedi ments (using a depth range of 200-1000 m), then the Ge sink is sufficient t o bring the hydrothermal budget based on Ge into concurrence with that base d on other tracers. The temporal variation in oceanic Ge/Si could be explai ned if Ge and Si inputs remain constant and the effective diagenetic fracti onation of Ge increases by a factor of 2-3 during glacial times. Increased fractionation would require that glacial periods are characterized by incre ased opal dissolution in iron-rich reducing sediments; this could be caused by (1) thinning of the oxygenated sediment layer in response to decreased bottom water oxygen concentrations or increased rain of organics to the sea floor, (2) increased rain of iron-rich detrital sediments in areas receivi ng high opal rain, (3) increased rain of opal to sediments in margin areas. If the oceanic Ge/Si ratio reflects increased rain of diatom opal or organ ic carbon in margin areas during glacial periods, it may indicate an increa se in the efficiency of the biological pump for CO2 during glacial times. C opyright (C) 2000 Elsevier Science Ltd.