GIDEON M. HENDERSON, Lamont-Doherty Earth Observatory of Columbia University, Palisades, New York 10964
CHRISTOPH HEINZE, Max-Planck-Institut für Meteorologie, Hamburg, Germany
ROBERT F. ANDERSON, Lamont-Doherty Earth Observatory of Columbia University, Palisades, New York 10964
Many chemical species exhibit particle-reactive behavior in seawater and are removed from the oceans largely by adhering to particles as they settle through the water column. In the southern oceans, these particle-reactive species are important for a number of reasons. For instance, particle-reactive pollutants such as lead and plutonium are transported from elsewhere in the oceans to this relatively pristine environment and removed there by high particle fluxes. The biolimiting nutrient iron also exhibits particle-reactive behavior that controls its rate of advection into the iron-poor southern oceans, and several particle-reactive radioactive nuclides are proving useful in this important region as proxies for past global change. Examples of the latter include the use of thorium-230 (230Th) as a constant flux indicator to assess accumulation rates of southern-ocean sediment constituents, and the use of protactinium-231/thorium-230 (231Pa/230Th) ratios to indicate past productivity (Kumar et al. 1995) or ocean-circulation (Yu, Francois, and Bacon 1996).
Several attempts have been made to model particle-reactive species for single ocean profiles, but global models of the advection and removal of these species have not been constructed. Such models, however, could offer key constraints on the rate of advection of particle-reactive species, their flux to ocean sediments, and their use as proxies of global change. As a first step toward incorporating particle-reactive species into global ocean models, we have focused on 230Th, which has several advantages for this purpose.
We have used two previously described global ocean models. The Large-Scale Geostrophic Ocean General Circulation Model (Maier-Reimer, Mikolajewicz, and Hasselmann 1993) derives ocean circulation, and the Hamburg Oceanic Carbon Cycle Model (Maier-Reimer 1993) produces surface-ocean productivity. To these models, we have added subroutines that accurately parametize the settling of surface-derived particles through the water column and the removal of particle-reactive species onto these particles as they settle. A manuscript describing this work is in preparation and further details are available at http://www.ldeo.columbia.edu/~gideon/ , together with a collation of all existing water-column 230Th measurements.
The accuracy of the particle field produced by the model is illustrated in figure 1 by a meridianal section of the western Atlantic. Absolute values produced by the model agree well with those of Brewer et al. (1976), and the pattern of particle concentration here, and elsewhere in the Atlantic, agrees well with that measured by nephelometry (Biscaye and Eittreim 1977). The accuracy of the 230Th handling is assessed by comparison with existing water-column measurements. Average depth profiles of model 230Th concentration, both in the dissolved and particle phases, agree well with the average of observed values, increasing from close to 0 at the surface to values of approximately 1 decay per minute per 1,000 liters dissolved 230Th at 5 kilometers (km). Comparison of model results with individual observed profiles demonstrates that the model is generally advecting and removing 230Th realistically. A particular success is the replication of low 230Th concentrations in the Labrador Sea; these low concentrations result from the advection to depth of low 230Th surface waters (Moran et al. 1997). The model-to-observation fit is less good, however, in the far southern oceans. Here, observed 230Th values are high, probably due to advection of water from the Weddell Sea where particle fluxes are very low and 230Th concentrations correspondingly high (Rutgers van der Loeff and Berger 1993). Model values in the far southern oceans are not, however, significantly higher than elsewhere, probably reflecting insufficient resolution in the present model to replicate accurately the complex circulation and sea-ice dynamics of the southern oceans.
Despite its failings in the far southern oceans, the model's accurate duplication of water-column 230Th values elsewhere suggests that it is advecting and removing 230Th realistically over most of the globe. This accuracy allows us to construct a global map of the removal of 230Th to ocean sediment relative to 230Th production in the overlying water column (figure 2). This map shows that, in regions where productivity is low, less 230Th is removed to the sediment than is produced. The extra 230Th is advected from these areas to be removed in areas of high productivity. The magnitude of this effect is such that up to half of the 230Th produced is advected in the most extreme cases.
Another run of the model was performed using Last Glacial Maximum boundary conditions (Winguth et al. 1996). Major features of the Glacial 230Th flux map (figure 3) are simlar to those of the Holocene with the notable exception that additional ice cover during the Last Glacial Maximum reduces removal of 230Th in the Arctic and, therefore, increases its removal in the North Atlantic and North Pacific. The Holocene and Last Glacial Maximum models demonstrate that the use of 230Th as a constant flux indicator in ocean sediments must be treated with some care to be interpreted at high precision.
Further work will refine the model in the southern oceans. Ongoing Joint Global Ocean Flux Studies (JGOFS) will help in this regard by improving understanding of the particle fluxes in the seasonal ice zone. The model will also be used to investigate other particle-reactive species. In particular, 231Pa has been introduced to the model to test the interpretation of 231Pa/230Th results from southern ocean sediments as proxies for past productivity (Kumar et al. 1995) and/or ocean circulation (Yu et al. 1996). Further work will also introduce beryllium-10 (10Be) and possibly particle-reactive pollutants to investigate their advection into, and removal from, the southern oceans.
Biscaye, P.E., and S.L. Eittreim. 1977. Suspended particulate loads and transports in the nephaloid layer of the abyssal Atlantic Ocean. Marine Geology , 23, 155-172.
Brewer, P.G., D.W. Spencer, P.E. Biscaye, A. Hanley, P.L. Sachs, C.L. Smith, S. Kadar, and J. Fredericks. 1976. The distribution of particulate matter in the Atlantic Ocean. Earth and Planetary Science Letters , 32, 393-402.
Kumar, N., R.F. Anderson, R.A. Mortlock, P.N. Froelich, P. Kubik, B. Dittrich-Hannen, and M. Suter. 1995. Increased biological productivity and export production in the glacial southern ocean. Nature , 378, 675-680.
Maier-Reimer, E. 1993. Geochemical cycles in an ocean general circulation model: Preindustrial tracer distributions. Global Biogeochemical Cycles , 7(3), 645-677.
Maier-Reimer, E., U. Mikolajewicz, and K. Hasselmann. 1993. Mean circulation of the Hamburg LSG OGCM and its sensitivity to the thermohaline surface forcing, Journal of Physical Oceanography, 23, 731-757.
Moran, S.B., M.A. Charett, J.A. Hoff, R.L. Edwards, and W.M. Landing. 1997. Distribution of 230Th in the Labrador Sea and its relation to ventilation. Earth and Planetary Science Letters , 150, 151-160.
Rutgers van der Loeff, M.M., and G.W. Berger. 1993. Scavenging of 230Th and 231Pa near the Antarctic Polar Front in the South Atlantic. Deep-Sea Research , 40(2), 339-357.
Winguth, A.M.E., E. Maier-Reimer, U. Mikolajewicz, and J.-C. Duplessy. 1996. On the sensitivity of an ocean general circulation model to glacial boundary conditions. Max-Planck-Institut für Meteorologie Internal Report . Hamburg, Germany: Max-Planck-Institut.
Yu, E.F., R. Francois, and M. Bacon. 1996. Similar rates of modern and last-glacial ocean thermohaline circulation inferred from radiochemical data. Nature , 379, 689-694.