The inability to sample this very important material derives 

 from its very loose and biochemically labile character. Its 

 residence time as a layer distinctly on top of the sediments is 

 likely on the order of days to weeks (Lampitt, 1985) before it is 

 either metabolized or transported away by bottom currents to 

 redeposit in lower energy depositional sites. The physical 

 looseness of the material has earned for it the names "fluff" or 

 "flocculant" layer. As the latter term connotes an aggregation 

 process that is not yet demonstrated, the former term - fluff 

 layer - will be used here. 



The physical looseness of the material renders its sampling 

 quite difficult. Most benthic sampling devices deployed from a 

 ship create a severe bow wave during descent, and consequently 

 push aside this material before entering the sediment. Attempts 

 to sample from a ship generally involve, therefore, an insertion 

 of a coring device done as slowly as possible. Upon retrieval, 

 the approach is generally to either scrape as fine a layer of 

 material from the sediment-water interface as possible, or to 

 induce a gentle resuspension of the fluffy material by agitating 

 supernatant water left in the coring device. 



In this paper we describe an approach to the sampling of 

 this layer that relies on (1) an in situ visit to the sediment- 

 water interface by submersible or SCUBA diver, and (2) a flume 

 device that relies on the physical looseness of the fluff layer 

 to resuspend it without resuspending the underlying sediment. 



This paper will focus only on the design and technology of 

 deployment of a fluff layer sampler ( FLS ) . Our first submersible 

 dives were carried out in the months preceding this Symposium, 

 and the scientific results from these dives are not yet ready for 

 publication. 



DESIGN AND FABRICATION OF THE FLS 



Design rationale 



We have chosen to build a sampler that collects the fluff 

 layer by a carefully controlled water flow over the sediment- 

 water interface, such that only the fluff layer is resuspended. 

 The flow is generated in a short flume nozzle by drawing water 

 through an open end and withdrawing it to a filter assembly where 

 it is collected on a glass fiber filter. The collection is 

 designed to provide enough material to undergo analyses for 

 carbon, nitrogen, and labile protein concentrations, in order to 

 test the hypothesis that the fluff layer is a zone of 

 nutritionally enriched material relative to the underlying 

 sediment . 



216 



