Limited understanding of the dynamics of suspended particulate matter (SPM) is an important prediction-limiting factor in shelf sea models. The problem concerns the size and properties of the aggregates which comprise SPM: these change on short time and length scales in shelf seas. The conventional wisdom is that turbulence is the primary physical control of aggregate size but there has been little field evidence to support this. While there is experimental and theoretical evidence for turbulence control of aggregation, there is contradictory evidence with respect to disaggregation: it has been proposed that sinking stresses, rather than turbulent stresses, are the dominant control of disaggregation. Progress in modelling SPM dynamics depends on resolving the role of turbulence as a determinant of SPM properties.
New observations are presented which provide compelling evidence for turbulence control of both aggregation and disaggregation. TKE dissipation and particle size were measured in situ at stratified sites in the northern North Sea in 110 m water depth during the period of weakening of the seasonal thermocline (in October/November) and in the Clyde Sea in 55 m water depth (April). There were similar vertical distributions of TKE dissipation E, SPM concentration C, and particle size D at both sites. At the base of the thermocline, there were minima in E and C, but a maximum in D, indicating that enhanced aggregation was occurring in this region of low turbulent stress. In the bottom mixed layer, E and C increased, while D decreased due to disaggregation in this region of increasing turbulent stress towards the seabed. Particles settling out of the low stress region at the base of the thermocline began to disaggregate when E increased to 3.2x10-6 watts m-2.
D did not correlate directly with E because aggregation is a function of collision frequency (and hence of both C and E): this can be accounted for using a simplified theoretical aggregation model which treats flocs as self-similar fractal entities and allows simultaneous floc formation and break up, specified as functions of C and E. It was found that in the northern North Sea the measured D represents an equilibrium size predicted by the model, while in the Clyde Sea tidal variation in both C and E prevented D from reaching equilibrium values in the near bed region. It is noteworthy that disaggregation in the bottom boundary was taking place at shear stresses less than 0.1Pa; these are low for the NW European shelf. In the Clyde Sea, where significant phytoplankton production was observed, there was a pronounced minimum in chlorophyll concentration at the base of the thermocline, which suggests that export of large flocs in this region is a mechanism for rapid transfer of organic carbon to the seabed. |
|
|