Plankton counting is a core component of aquatic ecosystem research, and its accuracy is influenced by multidimensional factors. In the entire process from sampling to analysis, any slight deviation may cause the data to deviate from the true value. The following will discuss the key steps:
1、 Spatiotemporal heterogeneity during the sampling phase
The phenomenon of water stratification leads to significant differences in vertical distribution, and the biological density in areas with sufficient surface light and areas with oxygen deficiency at the bottom can differ by several times. In terms of horizontal space, coastal zones often form dense areas due to nutrient input, while open waters are relatively sparse. The circadian rhythm triggers significant diurnal variations - diatoms and other phytoplankton float up to attract light during the day and sink to deeper layers at night, making it difficult to capture dynamic peaks in a single sampling. Seasonal changes lead to the succession of dominant species, and the difference in cell density between the outbreak period of dinoflagellates in spring and the flourishing period of diatoms in autumn can reach more than ten times.
The physical disturbance of the water collector produces a dual effect: appropriate turbulence helps to break the aggregates, but excessive impact can damage the structural integrity of fragile populations such as chain cyanobacteria. Although standard mesh sizes (such as 20 μ m sieve silk) can trap most microorganisms, smaller individuals of the genus Synechococcus may be lost.
2、 The Art of Sample Preservation and Preprocessing
The choice of fixative directly affects the preservation of cell morphology. Although Lugol's iodine can effectively terminate metabolic activity, iodine crystals may envelop ciliates and interfere with observation; Long term storage of formaldehyde solution can cause shrinkage and deformation of tissues in juvenile crustaceans. Untimely fixed live samples, the eye movement of flagellar algae can cause cells to leave the field of view.
During the sedimentation and concentration process, gravity promotes the preferential precipitation of heavy particles, resulting in the illusion of upper and lower stratification. Although centrifugation improves the recovery rate, the shear force generated by high-speed rotation can easily damage thin-walled dormant spores. In terms of staining treatment, neutral red staining can highlight flagellar structure, but it will produce fluorescence quenching effect on chlorophyll containing cells.
3、 The technical bottleneck of microscopic observation
The resolution limit of optical microscopes restricts the recognition of ultrafine eukaryotes (<2 μ m). When using the oil mirror, the residue of cedar oil is prone to adhere to tiny cells, causing leakage. Visual field scanning strategy affects statistical validity: Random visual field method is more likely to miss rare populations than systematic grid method. When observing live organisms, the rapid movement of ciliates requires a shorter exposure time, which conflicts with the need to obtain clear images.
The subjective error caused by operator experience is particularly prominent. The consistency of different technicians in determining similar morphological species is only about 75%, especially in distinguishing between radiating diatom fragments and intact cells. Under fatigue conditions, the missed detection rate increases exponentially after continuous observation of more than 40 fields of view per hour.
4、 The mathematical trap of data processing
Improper selection of volume conversion factors can amplify initial errors, such as mistakenly calculating the volume of a conical sedimentation tube as a cylinder. There is a lack of unified standards for handling abundance outliers, and some studies directly exclude data points that exceed the mean ± 3SD, which may result in the loss of important information. Although modern image analysis systems have improved efficiency, the accuracy of software algorithms in segmenting overlapping cells is still less than 80%.
5、 Key nodes of quality control
Blank control experiments reveal the risk of reagent contamination, and blank testing is required for each batch of newly purchased chemical drugs. Parallel sample analysis indicates that the coefficient of variation for repeated sampling at the same station should be controlled within 15%. Regularly calibrating the actual volume of the counting box, it was found that the volume deviation of the specially designed counting pool after six months of use can reach 8%.
By establishing a standardized operating procedure (SOP) to standardize the complete chain from sampling time recording to final counting review, system errors can be controlled within ± 10%. Combining molecular biology methods for cross validation is becoming a new trend in improving counting accuracy.