A substantial literature exists on the water motion required 

 to resuspend marine sediments (reviewed in Nowell et al., 1981), 

 but there are very few published data on the fluid flow capable 

 of resuspending fluff layers. The water flow necessary to 

 resuspend sediments is usually parameterized at the critical 

 shear velocity (sometimes called friction velocity or critical 

 entrainment velocity) , and is given by the expression - 



u* =(t/p ) 1 / 2 



* 



where u is the critical shear velocity, t is the critical shear 



* 



stress, and p is the water viscosity. The u parameter has units 

 of velocity (e.g., cm/sec) but does not actually represent a 

 velocity. Depending on the topography of the bottom sediment, 

 its value is often on the order of 1/2 5 of the bottom current at 

 an elevation of a meter or more above the sediment-water 

 interface. The critical shear velocities necessary to resuspend 

 marine bottom sediments are usually >0.8 cm/sec. (Nowell et al., 

 1981) . The only published literature that allows estimation of 

 the u* necessary to resuspend the fluff layer is from the 

 combined bottom photo and current meter time series of Lampitt 

 (1985), which results in a critical u* of about 0.3 cm/sec. 



We chose a u* of 0.4 cm/sec as an operating shear velocity 

 at which to run the flume. The fluff layer therefore takes on an 

 operational definition based on its susceptibility to erosion, 

 rather than a depth definition or a non-quantifiable type of 

 resuspension such as might be done in a box core. 



The size of the flume was dictated by the sample 

 requirements on which we can conduct the protein and C/N 

 analyses - about a gram of fluff layer material. As a 

 conservative calculation, assume a fluff layer thickness of 2 mm 

 consisting of material with a water content of 98% and a solids' 

 density of about 1.0 g/cm 3 . The flume's floor (19.9 cm x 14.2 

 cm) has an area of 283 cm 2 , which is sufficient to produce the 

 required qram of material with these conservative assumptions. 



The inverted flume nozzle (Figure 1) was designed with the 

 goal of achieving a homogeneous shear velocity field over the 

 entire confined sediment area. A trip bar at the opening of the 

 flume creates a controlled stress distribution at the beginning 

 of the flow. In a square duct aligned horizontally, a boundary 

 layer flow would then develop with the u* prescribed by laminar 

 or turbulent bottom stress formulae, depending on the Reynolds 

 number. This u* would vary with downstream distance. The top 

 wall of the flume nozzle was angled to preserve the desired u* 

 distribution over the length of the flume. The optimal angle of 

 this top wall was determined in an experimental program using 

 skin friction sensors (Gust, submitted) and a pumping rate of 

 22.9 1/min. This experimental work was performed in a 

 laboratory recirculating flume, with flow through the flume 

 nozzle provided by an aquarium pump. The skin friction sensor 

 was mounted flush with the bottom of the larger recirculating 



217 



