Pre-filtration significantly alters plankton size distribution

19.02.2019  by  Tina Silovic


Figure 1. Top: scheme of ecosystem division in various particle sizes; Left side: scheme of the ideal particle separation; Right side: scheme of expected separation expected to occur in practice.

The main focus of imaging flow cytometers has been for years to produce reliable counts and image all particles flowing through. An equally important issue, yet receiving less attention, is to which extent the samples entering the instrument reliably represent the original particle size distribution in the source water. First thing tampering with this reliability is the fact that samples for flow cytometry are always prefiltered over a protective screen during the sample pre-processing protocol or by directly protecting the inlet of the instrument in autonomous operation. The pore size of that filter differs from instrument to instrument, but the principle is the following: each instrument has a flow cell with limiting dimensions D, which is protected by filter pore size D; all particles < D are passing through, whereas particles > D are blocked by the mesh. In practice however, such discrete size fractionation does not exist.  In order to investigate the fractionation effect on plankton communities we performed an experiment by using CytoSense flow cytometer with limiting orifice of about 950 m diameter injector tube and inlet tubing of 800 m in internal diameter. The results showed us that selective under-sampling already starts when particles have a size of around a third of the sample inlet (Figure 1). What does this mean in practice? It means if you want to analyse the whole microplankton community (cell size range: 20-200 m) in a sample, you would need to pre-filter the sample through a filter or screen with not less than 600 m pore size. In conclusion, we want to stress that not taking this fact into account could cause severe underestimation of certain groups in the original water source.

For more information download the white paper "Sample filtration reduces representative sampling to 1/3 of filter pore size".