Field and Numerical Studies of Near-Bed Aggregate Dynamics
187
floor. However, the vertical extension of these processes is unknown and may be
smaller or larger than the range of sampling heights in the BBL displayed in Fig.
la-d; but it is most likely linked to the hydrodynamic regime. The dynamics of
these aggregation-disaggregation cycles are determined by the composition and
the properties of the involved aggregates and may be different for different aggregate types. The crude field measurements of bulk parameters, like the concentration of total suspended matter or particulate organic carbon, do not allow
for such discrimination of aggregate properties, which are required to evaluate
the processes of interest. In contrast, microscopic investigation of the aggregate
images and the resulting size class distribution for different heights above the
seafloor support the idea of hydro dynamical sorting in the BBL. It has to be kept
in mind, how-ever, that the observed preserved particles only potentially represent fragments of aggregates in situ. Sample handling most likely destroys their
original configuration.
Moreover, since the hydrodynamic regime is very dynamic close to the seafloor, bulk measurements of particle composition do not allow for discrimination of the various aggregate sizes and properties. This issue has been approached by applying optical methods in the field. Aggregate size distribution,
abundance and settling velocity were studied in various environments by analysing photographic images of particles taken in situ (Milligan 1995; Asper 1986,
1987; Chen et al. 1994; Rathmeyer and Wefer 1996). In a similar attempt, using
an in situ video imaging system combined with a bottom water sampler, particles in the BBL were observed at the East Greenland continental margin (Thomsen and Ritzrau 1996). This device videotaped particles at defined distances
close to the seabed. The recorded video sequences allowed not only the size but
also the behaviour and speed of aggregates in the flow to be estimated.
Based on this experimental evidence and clues, and in contrast to the approaches described so far, a numerical multilayer 1-D diffusion -advection model was developed to investigate particle or aggregate dynamics in the BBL, under
inclusion of effects of particle interactions on the particle distribution in the
BBL. Initially, the basic setup of the presented model used distinct particle size
classes. With inclusion of particle interactions, aggregates are formed or destroyed. However, for simplicity, we prefer to use the term aggregate to describe
the model results. A few case studies are included to display the strength and
sensitivity of this model.
The model presented includes the particular hydrodynamic regime of the
benthic boundary layer and provides a continuous distribution of various distinct aggregate size classes from the seafloor into the water column, as suggested
by Newberger and Caldwell (1981). The features of the different aggregate classes and the settings of the hydrodynamic regime can be changed easily. The advantage of the numerical simulation over the analytical solution is that the model provides not only the final distribution pattern for selected aggregate classes,
but also the temporal development of their distribution pattern. The model allows this approach either for a specific height above the seafloor or for the entire
benthic boundary layer.
187
floor. However, the vertical extension of these processes is unknown and may be
smaller or larger than the range of sampling heights in the BBL displayed in Fig.
la-d; but it is most likely linked to the hydrodynamic regime. The dynamics of
these aggregation-disaggregation cycles are determined by the composition and
the properties of the involved aggregates and may be different for different aggregate types. The crude field measurements of bulk parameters, like the concentration of total suspended matter or particulate organic carbon, do not allow
for such discrimination of aggregate properties, which are required to evaluate
the processes of interest. In contrast, microscopic investigation of the aggregate
images and the resulting size class distribution for different heights above the
seafloor support the idea of hydro dynamical sorting in the BBL. It has to be kept
in mind, how-ever, that the observed preserved particles only potentially represent fragments of aggregates in situ. Sample handling most likely destroys their
original configuration.
Moreover, since the hydrodynamic regime is very dynamic close to the seafloor, bulk measurements of particle composition do not allow for discrimination of the various aggregate sizes and properties. This issue has been approached by applying optical methods in the field. Aggregate size distribution,
abundance and settling velocity were studied in various environments by analysing photographic images of particles taken in situ (Milligan 1995; Asper 1986,
1987; Chen et al. 1994; Rathmeyer and Wefer 1996). In a similar attempt, using
an in situ video imaging system combined with a bottom water sampler, particles in the BBL were observed at the East Greenland continental margin (Thomsen and Ritzrau 1996). This device videotaped particles at defined distances
close to the seabed. The recorded video sequences allowed not only the size but
also the behaviour and speed of aggregates in the flow to be estimated.
Based on this experimental evidence and clues, and in contrast to the approaches described so far, a numerical multilayer 1-D diffusion -advection model was developed to investigate particle or aggregate dynamics in the BBL, under
inclusion of effects of particle interactions on the particle distribution in the
BBL. Initially, the basic setup of the presented model used distinct particle size
classes. With inclusion of particle interactions, aggregates are formed or destroyed. However, for simplicity, we prefer to use the term aggregate to describe
the model results. A few case studies are included to display the strength and
sensitivity of this model.
The model presented includes the particular hydrodynamic regime of the
benthic boundary layer and provides a continuous distribution of various distinct aggregate size classes from the seafloor into the water column, as suggested
by Newberger and Caldwell (1981). The features of the different aggregate classes and the settings of the hydrodynamic regime can be changed easily. The advantage of the numerical simulation over the analytical solution is that the model provides not only the final distribution pattern for selected aggregate classes,
but also the temporal development of their distribution pattern. The model allows this approach either for a specific height above the seafloor or for the entire
benthic boundary layer.