supply oxygen and pump water out to expel waste products, we regarded the 

 biopumping model as more appropriate. In the model of McCaffrey at al (8), it 

 was assumed that organisms pumped water across the sediment- water interface 

 at a certain "biopumping rate" (with units of volume of water per unit surface 

 area per unit time). The biopumping rate was determined experimentally by 

 bringing box cores of Jamestown North sediment into the laboratory, spiking 

 the supernatant solution with '^'^Na, and measuring the decrease in the 

 supernatant ■^■^Na concentration with time. The experiments gave a 

 biopumping rate of 0.7±0.3 cm cm day . The biopumping flux is then 

 taken as the product of the biopumping rate and the difference between pore 

 water and bottom water concentrations. 



Model diffusive and advective fluxes for summer Jamestown North 

 sediments are given in Table 2-3. Surprisingly, both fluxes are of the same 

 magnitude. 



RESULTS OF BENTHIC FLUX MEASUREMENTS 



In order to test our predictions that nutrient and manganese fluxes have the 

 values calculated from the model outlined in the previous section, and to make 

 direct measuremertts of copper and nickel fluxes, we have measured benthic 

 fluxes in the field using the "bell jar" instruments developed and extensively 

 deployed by Hale (7) and Nixon et al (9). In these experiments, PVC pipe 

 halves with closed ends PVC flanges around the base are placed on the 

 sediment. At the start of the experiment, a sample is withdrawn from the 



^Assuming a biopumping rate of 0.7 cm cm day" 



17 



