Particle settling from variable mixed layers an experiment

As part of an effort to improve the empirical description of the sedimentary losses of phyto-plankton from suspension, Reynolds employed several approaches to measuring the sedimentary flux in the large limnetic enclosures in Blelham Tarn, UK. These cylindrical vessels, 45 m in diameter, anchored in 11-12 m of water contained sufficient water (~18 000m3) to behave like natural water columns. Their hydraulic isolation ensured all populations husbanded therein were captive and virtually free from external contamination (Lack and Lund, 1974; Lund and Reynolds, 1982; see also Section 5.5.1). The deployment of sediment traps was to have been a part of this programme and it was decided that the choice of traps and the authenticity of their catches could be tested within the special environment of the enclosures. The plan was to add a measured quantity of alien particles to the water column (actually, Lycopodium spores, well steeped in wetting agent and preservative), then to monitor the subsequent loss from suspension in the water and to compare the calculated flux with the sediment trap catches. Three such experiments were carried out, under differing hydro-graphic conditions. The results were published (Reynolds, 1979a) but the unexpected bonus of the experiments was the contrasting rates of loss from suspension of ostensibly identical spores under the varying conditions.

In the first experiment, carried out in winter, the spores were dispersed over the enclosure surface, during windy conditions which intensified in the subsequent few days. A near-uniform distribution with depth was quickly established (see Fig. 2.20). The spores (d = 32.80 ± 3.18; pc = 1049 kg m-3; ~ 2.2) had a measured sinking rate (ws) of 15.75 ^ms-1 at 17 °C, which, adjusted for the density and viscosity of the water at the 4-5 °C obtaining in the field, predicted an in-situ intrinsic sinking rate of 0.96 m d-1. The theoretical time for spores to eliminate the enclosure (at the time, H ~ 11.8 m) was thus calculated to be (t' = ) 12.3 days. In fact, the elimination proceeded smoothly, always from a near-uniformly distributed residual population at an average exponential rate of -0.10md-1, which value corresponds to a 95% removal in (te =) 30 days. The ratio te/t' is lower than predicted in Section 2.6.2 (2.44 against 3.0). This maybe explained by probable violation of the initial assumption of full mixing of the water column throughout the experiment. Although no significant density gradient developed, continuous and complete vertical mixing of the enclosure cannot be verified. Nevertheless, the outcome is sufficiently close to the model (Fig. 2.20) solution for us not to reject the hypothesis that entrained particles are lost from suspension at an exponential rate close to

In the second experiment, commenced in June, spores were dispersed at the top of the

Figure 2.20

_ Modelled (M) and actual (A) depth-time distributions of preserved Lycopodium spores (of predetermined sinking characteristics) introduced at the water surface of one of the Blelham enclosures on each of three occasions (1,9 January; 2, 3 June; 3, 9 September) during 1976, under sharply differing conditions of thermal stability. Lycopodium concentrations plotted as cylindrical curves; density gradients plotted as dashed lines. * - indicates no field data are available. Redrawn from Reynolds (1984a).

stratified enclosure during relatively calm conditions. Sampling within 30 minutes showed a good dispersion but still restricted to the top 1 m only. However, 4 days later, spores were found at all depths but the bulk of the original addition was accounted for in a 'cloud' of spores located at a depth of 5-7 m. After a further 7 days, measurable concentrations were detected only in the bottom 2 m of the column, meaning that, effectively, the addition had cleared 10 m in 11 days, at a rate not less than 0.91 m d-1. Adjusted for the density and viscosity of the water at the top of the water column, the predicted sinking rate was 1.42 m d-1. Thus, overall, the value of t' for the first 10 m (= 7 days) was exceeded by the observed te (= 11 days) by a factor of only 1.57. Part of the explanation is that sinking spores would have sunk more slowly than 1.42 m d-1 in the colder hypolimnion. However, the model explanation envisages a daily export of the population from the upper mixed layer (varying between 0.5 and 4 m during the course of the experiment), calculated as N exp — (ws/hm), whence it continues to settle unentrained at the rate ws m d-1. To judge from Fig. 2.20, this is an oversimplification but the prediction of the elimination is reasonable.

The same model was applied to predict the distribution and settlement of Lycopodium spores in the third experiment, conducted during the autumnal period of weakening stratification and mixed-layer deepening. Variability in wind forcing was quite high and a certain degree of re-entrainment is known to have occurred but the time taken to achieve 95% elimination from the upper 9 m of the water column (te = 18.0 days) at the calculated in-situ sinking rate (ws) of 1.32 md-1 exceeded the equivalent t' value (9/1.32 = 6.82) by a factor of 2.6.

The three results are held to confirm that the depth of entrainment by mixing is the major constraint on elimination of non-motile plankters heavier than water, that the eventual elimination is however delayed rather than avoided, and that prolongation of the period of suspension is in proportion to the depth of the mixed layer, wherein u* > 15 (ws).

0 0

Post a comment