cascading by resuspension, advection and subsequent settling of aggregates may be a 

 significant pathway for transport of carbon across the slope. Such a mechanism may explain 

 any discrepancy between sediment oxygen demand and the export flux derived from overlying 

 productivity. 



The diel (24 hr) cycle is important in biological cycles, and many diel cycles have been 

 observed in transmissometer data, aggregate abundance, mixed layer thickness and other 

 biologically related parameters. Sampling will be planned to monitor changes on this 

 important time scale. 



In-situ particle setding velocities are needed to transform aggregate abundance data into 

 aggregate fluxes. Particles produced on the shelf with high settling velocities are retained on 

 the shelf, whereas slower sinking particles are more likely to be transported seaward. Thus 

 settling velocities of particles, coupled with the flow field characteristics of the water 

 determine where deposition occurs on the continental margin. 



^^''Th was one of the few tracers identified as being able to assist in quantifying particle 

 export from the waters overlying the shelf in the vicinity of Cape Hatteras. This is due to 

 ^^"^Th's relatively short half-life (24.1 days) and its non-specific particle reactivity. The activity 

 distribution of ^^'^ should reflect particle removal processes with time-scales on the order of 

 days to many weeks. For longer time-scales (years to decades), the ^^°Pb budget can be used 

 in a similar manner, but need not be sampled as frequently. ^^"^ and ^^^b analyses are 

 conducted conveniently on the same samples. 



The ^^'*Th tracer technique has been successfully used in the open ocean to quantify 

 export fluxes from the surface ocean. The major modification for the Oceans Margins 

 Program would be to measure the advective fluxes by looking at concentration gradients and 

 water velocities determined with moorings, as well as a sampling regime which minimized 

 resuspended fluxes (by sampling surface waters in stratified waters over the outer shelf, and/or 

 determining ^^^Th resuspension fluxes by other techniques). Export fluxes of ^^^Th would be 

 compared to the sediment inventories of excess ^^^Th to determine the amount of ^^''Th (and 

 hence particulate organic C) exiting the shelf. Once the ^"^Th flux is determined, the ratio of 

 organic C or N to ^^'^ on particles (deterpiined by analysis on filters) can be used to convert 

 from ^^"^Th export to organic C and N partiftle export. 



Suspended matter concentrations in shelf and slope water will be used as a master 

 variable to predict organic carbon exchanges and rates from the continuous records from 

 moored transmissometer. This approach is based on the following observations and model 

 predictions: 1) particle concentration effects on ^^''Th scavenging rate constants, particle 

 removal rate constants, partition coefficients between particles, colloids and solution due to the 

 coagulation of ^^''Th containing coUoids with particles ("colloidal pumping"); 2) the constant 

 functionality of the partitioning between suspended particles, colloids COOlOkD and 

 COOlkD, and in many instances DOC; and 3) the similarity of the colloidal/dissolved ^^''Th 

 partitioning and COC/DOC partitioning. It will therefore be more important to measure ^^Th 

 profiles along gradients of hydrodynamic biological and chemical conditions than to measure 

 them along the real transit of a water parcel. 



23"^ shows a deficiency with respect to its parent, ^^U, (i.e.,[2^^lJ]-p'^]»0) 

 throughout the water column in slope waters 800-1500 m deep, indicating strong lateral 



16 



