We could characterize a suspension of flocs based on the terminal velocity of the flocs and on the turbidity of the narrow ranges of particle size. Since the particle size will be measured in terminal velocity, the narrow ranges of particle size will also have dimensions of velocity. Thus the y axis will have units of NTU/(m/s). The x axis has units of m/s, and the area under the curve is thus the total turbidity of the suspension. A suspension with a small number of very small particles and a large number of large particles is shown in the figure.
If this suspension were in a column that was monitored by a turbidimeter, the fast settling particles would settle past the turbidimeter first and then much later the small particles would settle past the turbidimeter sensor. Thus the turbidity vs time plot is shown below.
The time on the x axis can be converted into a terminal velocity based on the distance between the top of the column and turbidimeter sensor.
Now we can see that this final figure is actually the integral of the particle size distribution shown in the first Figure. Thus we can move between the trace of turbidity vs time during the sedimentation phase to a particle size distribution. This data transformation could be one of the next steps in analyzing the data from the tube floc apparatus.
We also observe that the initial slope of turbidity vs time is expected to be flat for the first few seconds. I don't think the collected data shows that trend and thus it may be worth investigating why the turbidity was decreasing from the very beginning of the sedimentation phase. An instantaneous decrease in turbidity would require an infinite sedimentation velocity for at least some of the particles in the suspension.