Aquaculture Foam Fractionation: Skimmer Removal and Fine Solids

Foam at the top of a skimmer is not a complete solids balance

A recirculating aquaculture system may use a foam fractionator, sometimes called a protein skimmer, to collect certain fine particles and surface-active organic substances from water. The device makes bubbles, allows selected material to concentrate near the bubble surfaces, and removes the resulting foam into a separate waste stream. That visible foam is compelling, but it is not proof that every kind of dissolved organic carbon, ammonia, phosphorus or microscopic particle has been removed in the same proportion. Reading a performance report accurately requires tracing the water and waste streams and naming the measured substance.

This article explains foam fractionation from first principles, including interfacial chemistry, flow, contact, salinity, particle size, skimmate, reported removal percentages and uncertainty. Every worked numerical example is hypothetical teaching arithmetic, not a measured farm result or a design instruction. Safe aquaculture operation needs species-specific water-quality monitoring, equipment validation and appropriate engineering. The purpose here is to translate technical evidence into plain but exact language so that a reported improvement is neither dismissed nor overstated.

The first question is where the apparently removed mass went. A water concentration can fall because of actual capture in the foam waste, dilution, settlement, altered sampling or another stage of treatment. A foam container may look dark but contain much water; it may have high pollutant concentration yet represent only a modest total mass because its volume is small. A correct report therefore needs incoming water flow and concentration, outgoing water flow and concentration, and an appropriately characterized foam-waste stream.

1. Surface adsorption gives bubbles a selective role

Air bubbles create a gas–water interface. Some molecules and particles interact with that interface more favourably than they do with bulk water. Surface-active organic compounds can accumulate there, and particles can attach directly or through associated substances. As bubbles rise, they carry this selected fraction upward toward the foam zone. The mechanism is selective, not a universal sieve for everything dissolved in water. A compound that does not interact sufficiently with the bubble interface may pass through even when foam production looks vigorous.

This helps explain why the everyday word “protein” in protein skimmer is incomplete. The collected material can include mixtures of organic compounds, fine solids, microbes and water, depending on chemistry and operation. It is better to describe the measured removal of defined classes than to assert that all protein or all organic waste has been extracted. The distinction matters when an equipment claim moves from a seller to an engineer, or from one language into another: a colourful waste cup is not a complete chemical analysis.

2. How a fractionator fits with drum filters

A drum filter and a foam fractionator can both contribute to solids management but typically act on different physical principles and size fractions. A screen primarily retains particles larger than an effective opening or captured in its cake, while bubble separation can capture some finer material associated with the air-water interface. A treatment sequence may use both devices. If a study measures inlet water before the drum but outlet after the fractionator, it cannot attribute the entire reduction to foam fractionation alone without a controlled component comparison.

The published 2024 study by Pfeiffer and colleagues examined fine-solids removal by foam fractionation in a low-salinity marine recirculating system under several operating conditions. It reported different outcomes by configuration, illustrating why a named equipment arrangement, fine-particle definition and testing interval belong in the result. The study provides a methodological reference, not a universal percentage that applies to every freshwater pond or marine hatchery. Translators should retain the words “under tested conditions” when that limitation governs the inference.

3. The meaning of fine particles depends on the method

A water-quality laboratory may sort suspended material by nominal screen or filter size, such as a fraction smaller than a stated micrometre cutoff. That size boundary depends on the method, shape and behaviour of particles; it does not describe a perfect set of spherical objects. A fractionator may improve the concentration of one measured fine fraction while leaving larger particles mostly unaffected. A report should therefore state the size class, collection method and whether concentrations are by mass, number or another particle metric.

When translating a statement like “removal of particles under 55 micrometres”, preserve the less-than qualifier and unit. The phrase should not become “removal of all particles” or “water sterilization”. Particle counts and suspended-solids mass also answer different questions: a very large number of tiny particles may contribute modest mass, whereas fewer large fragments can dominate gravimetric solids. If one study uses a particle counter and another uses TSS, direct comparison needs methodological reconciliation.

4. Read total suspended solids correctly

Total suspended solids, or TSS, are commonly reported in milligrams of measured suspended matter per litre under a specified filtration and gravimetric method. A reported 8 mg/L means eight milligrams per litre on that method’s analytical basis, not eight grams per tank or eight particles per millilitre. To calculate an approximate mass flow one multiplies concentration by volumetric flow. Without the flow, a concentration reduction cannot determine total removal rate. This is particularly important where only a fraction of system water passes through a sidestream skimmer.

TSS is also a methodological quantity: filter medium, handling, drying and sampling variation can influence results. If a foam fractionator acts preferentially on small particles, the difference between two low TSS values may be close to method uncertainty. Replicated, appropriately paired samples become important. A report should describe the sampling interval and whether feed events, tank cleaning or filter backwash changed the solids loading. Numbers that look precise to a tenth of a milligram still need evidence of actual measurement precision.

5. An inlet/outlet concentration change is not all the mass

Suppose a constructed water stream enters at 12 mg/L TSS and leaves at 9 mg/L. The concentration-based removal is 100×(12−9)/12 = 25 percent. At an inlet flow of 20,000 L/h, the nominal input load is 240,000 mg/h, or 240 g/h. If the outlet also truly flows at 20,000 L/h, output is 180 g/h, leaving an apparent difference of 60 g/h. These numbers express a defined interval and assumption; they do not establish the actual collected solids without accounting for all flows and measurement error.

In reality the foam cup removes liquid, and there can be additional drainage or return lines. If the treated-water outlet flow is slightly smaller, a complete balance must use its measured volume. Some material can accumulate temporarily inside the unit rather than enter the foam waste immediately. The responsible wording is “apparent component mass reduction from matched water-flow measurements” unless the skimmate and internal inventories complete the accounting. Precision in the label prevents an apparently simple 25 percent result from becoming an unsupported recovery claim.

6. Skimmate concentration and volume must be combined

The waste liquid collected from foam is often called skimmate. It can be more concentrated in some constituents than the incoming water, but its total removed mass depends on both concentration and collected volume. Suppose 2 L of skimmate contain 500 mg/L of a measured solids fraction during a particular interval. The product is 1,000 mg, or 1 g, of that fraction. The dark colour of the cup does not allow an observer to infer a larger or smaller mass. Use the correct assay on an adequately mixed and representative waste sample.

A high skimmate concentration can accompany relatively small volume; a wetter foam may yield more volume but less concentration. A comparison based only on appearance or skimmate concentration can favour a setting that removes less actual contaminant mass. If the objective is waste export, report dry mass per time and water loss per time alongside the concentration. The phrase “more concentrated foam” does not necessarily mean “more effective overall removal”, and two reports can disagree simply because they use different effectiveness metrics.

7. Dry foam and wet foam are operationally distinct

Skimmer users may describe a dry, stable head of foam or a wetter flow of frothy liquid. These labels are qualitative shorthand for drainage and collection behaviour, not precise standards. The amount of liquid entering the waste cup influences apparent removal, water replacement needs and pollutant concentration. A dry collection mode might yield a small concentrated waste stream, while a wetter mode yields a greater waste volume. Which result is better depends on the desired removal mass, water balance and actual effluent quality.

A fair performance assessment names water flow through the skimmer, aeration conditions, waste volume, waste concentration, and treated-water concentration under matched periods. If a trial changes both bubble generation and cup height, it cannot attribute all differences to one setting. When converting operational notes into English, preserve which adjustment was made and when. Do not turn the casual word “dry” into a fabricated quantitative moisture percentage.

8. Salinity and foam stability are not universal constants

Surface chemistry and bubble behaviour can change with salinity and the composition of dissolved organic matter. A skimmer that generates stable foam in one marine system may behave very differently in freshwater. Species of organic material, ionic conditions, pH and water temperature all influence adsorption and stability. Therefore performance results obtained at a particular salinity should retain that context. It is misleading to present a low-salinity marine study’s particle-removal result as an expected freshwater percentage.

Even within a farm, salinity can change after water replacement, evaporation, salt addition or production-stage transitions. A comparison between two days with different salinity may confound an equipment adjustment. A good log records conductivity or salinity measurement and its instrument basis along with treatment parameters. If foam collapses unexpectedly, the operator should investigate chemistry and equipment conditions rather than assuming more air alone will solve the problem.

9. Bubble size and contact time influence capture

Smaller bubbles can provide more gas-water interfacial area per unit gas volume, but capture also depends on bubble distribution, residence time, turbulence, collision probabilities and foam drainage. Simply increasing air rate can change both contact and turbulence in competing ways. A performance curve may have an optimum within the tested range rather than monotonic improvement. An equipment advertisement that says “more bubbles means more cleaning” is therefore an incomplete causal explanation.

A scientific comparison should specify how gas is introduced, such as a venturi, diffuser or aspirating impeller, and how air and water flows were measured. If ozone is added to an injector, chemical oxidation may alter measured outcomes as well as physical fractionation. The two mechanisms should be distinguished before a removal result is attributed entirely to bubble adsorption. Reproducing a study means reproducing the operating context, not merely buying a device with the same name.

10. What hydraulic residence time can and cannot tell us

Suppose the water volume effectively occupying a foam-contact chamber is 300 L and measured throughflow is 150 L/min. The nominal volume-over-flow residence time is 2 minutes. This calculation does not prove that every water parcel experiences precisely two minutes with the same bubbles. Mixing, short-circuiting, recirculation and internal water hold-up affect the actual distribution. More importantly, residence time alone cannot predict adsorption of an unknown organic mixture without appropriate rate and equilibrium information.

A test description should distinguish the volume of the foam-contact zone from a sump, waste cup or downstream buffer. Including all vessel volumes in one number may give an attractive but misleading residence time. If an internal recirculation pump moves water several times through the contact chamber, state whether the reported flow is fresh process throughput or internal recycle. These are different denominators, and a poorly labelled flow can multiply or divide the apparent contact time by a large factor.

11. Dissolved organic carbon is not the same as suspended solids

Dissolved organic carbon, DOC, is an analytical category defined by a sample-preparation and measurement method. Total organic carbon, TOC, and chemical oxygen demand, COD, are related but different quantities. A foam fractionator may affect some surface-active dissolved compounds, but it does not necessarily remove all dissolved carbon or achieve a fixed proportion of COD reduction. A study demonstrating fine-solids removal is not automatically evidence of equal DOC, TOC, pathogen or nitrogen removal.

The words carbon, oxygen demand and organic matter carry different analytical definitions. A translated results table should keep them separate, with units and methods. If DOC is reported as mg C/L and COD as mg O₂/L, the identical number 10 describes different chemical accounting. Never subtract them to determine “carbon cleaned”. Instead make a separate claim for each measurement and indicate any relationship that was actually tested.

12. Filter, skimmer and biological treatment have separate jobs

A recirculating system may depend on a train of mechanical filtration, foam fractionation, nitrification, gas exchange and disinfection. These components can interact. Better solids removal may lessen the load on a biological filter or improve downstream light transmission, but a decrease in suspended material does not establish that a biofilter has gained a quantified ammonia-removal capacity. The operator must measure the relevant endpoint at each stage. This separation of mechanisms helps readers understand why one treatment unit cannot be described simply as “cleaning the whole system”.

If an experiment claims that the skimmer lowered ammonia, inspect whether ammonia was directly measured and whether biological activity, dilution or an additional reactor was involved. A foam waste stream may export nitrogen bound in organic matter, but that is not the same phenomenon as immediate removal of dissolved total ammonia nitrogen. A well-translated methods section states what changed physically, what changed chemically, and which causal claims the control conditions support.

13. The correct solids balance includes every outlet

A foam separator is a separation process, so material can leave through treated water, a concentrated waste stream or another drain. A comparison of incoming and treated-water concentration is a useful start, but not always a complete accounting of exported solids. Suppose 20,000 litres an hour enter at 10 mg/L TSS: input is 200 g/h. If 19,950 litres an hour leave in treated water at 8 mg/L, that stream carries 159.6 g/h. The missing volume and other material need to be accounted for before writing that 40.4 g/h were captured.

If a waste stream of 50 L/h holds 600 mg/L of the same solids measurement, it exports 30 g/h. The two measured outlet streams sum to 189.6 g/h, leaving an apparent 10.4 g/h discrepancy against the input. The discrepancy may reflect storage within the apparatus, sample variability, another outlet or timing. It does not establish that the remaining solids chemically disappeared. A transparent balance reports the discrepancy and its possible causes rather than forcing the numbers to add up through unsupported adjustments.

14. Wet foam cannot be reported as dry protein

A collector containing 8 kilograms of liquid foam waste does not contain eight kilograms of protein. Suppose a valid gravimetric test finds a 1.5 percent dry-solids mass fraction. The contained dry mass is 8×0.015=0.12 kg or 120 g, which may include several inorganic and organic components. Without a protein-specific assay, even the 120 g cannot be labelled pure protein. Both the moisture basis and chemical identity matter whenever removal is reported to management.

If that container accumulated over six hours, the mean captured dry-solids rate was 120/6=20 g/h. The wet collection rate was 8/6≈1.33 kg/h. These are separate operational metrics with different meanings. A salesman might advertise whichever value sounds more impressive, but a technical translation should show both and explain their relationship. That clarity helps prevent false comparisons between a wet-foam configuration and a drier one.

15. Concentrated skimmate and total mass can point in different directions

Consider two settings tested for the same duration. Setting A collects 2 L at 900 mg/L of a target analyte, exporting 1.8 g. Setting B collects 6 L at 400 mg/L, exporting 2.4 g. Setting A produces the more concentrated waste, but setting B exports more measured target mass. Neither outcome proves B is operationally preferable because it also exports more water and may require greater make-up supply. The useful comparison contains volume, concentration, mass and treatment objective.

This exercise repairs the phrase “higher waste concentration equals better skimming”. It does not. A concentration describes what is in each litre; a removal mass describes how much left over the stated interval. The distinction becomes even more important when evaluating nutrient export, chemical analyses or wastewater disposal cost. A water-intensive setting can appear highly effective by waste-cup volume alone while delivering relatively poor removal per litre of discarded water.

16. Sidestream performance is not a tank-wide percentage

A protein skimmer often treats a sidestream while other water moves through separate routes. A 25 percent reduction between that sidestream’s inlet and outlet does not mean the fish-rearing tank concentration drops by 25 percent every hour. Tank response depends on the fraction of system water routed to treatment, how well return water mixes, continued waste production and other removal mechanisms. This is a dynamic problem, not merely a percentage multiplied across a whole facility.

Suppose a 10,000-litre tank has a skimmer throughput of 1,000 L/h. Even if each pass achieves a measured reduction for the targeted material, the device treats only a volume equivalent to a tenth of tank volume in one hour, before accounting for mixing and new solids. A genuine system model would integrate concentrations, flows and sources through time. The wording “per-pass sidestream reduction” keeps the tested claim from becoming an unsupported whole-tank guarantee.

17. Fine-particle counts and fine-particle mass are different

A particle counter may report number per millilitre within a defined size band, while TSS represents measured suspended dry mass. A reduction from 2,000 to 1,500 counted particles per millilitre is 25 percent in that counted population. It need not produce a 25 percent reduction in gravimetric suspended solids because particle sizes, shapes and densities differ. A few heavier fragments can dominate mass even when numerous fine particles dominate counts. A comparison should name whether the endpoint is count, mass, turbidity or size distribution.

If particles aggregate during treatment, the count could decrease while total mass changes little. Conversely, capture of a small number of large fragments could reduce TSS sharply without a proportionate change in fine counts. These mechanisms are plausible alternatives that need measurement. A translator should preserve particle-size thresholds, counting method, filter cutoffs and unit basis. Changing “particles smaller than 55 µm” to “all particles” erases both the scope and the useful detail of the measurement.

18. Salinity can alter the meaning of comparison

Marine, brackish and freshwater systems do not necessarily support identical foam formation. Ionic strength, salinity and the specific dissolved-organic mixture influence surface behaviour and bubble coalescence. A research result obtained in water of a stated salinity should retain that detail even if the headline concerns a familiar commercial skimmer. A device’s apparent capture efficiency can shift with water replacement, evaporation, or production-stage changes that alter chemistry.

When comparing trials, log salinity, temperature, pH and organic loading alongside air rate and cup setting. If a saltwater study reports strong fine-solids removal, it does not establish the same result in freshwater without relevant tests. The best translated literature summary identifies the water matrix and cautiously describes transferability. Scientific qualifiers can be the difference between a useful planning hypothesis and a misleading purchase promise.

19. Bubble generation has several coupled effects

A smaller bubble can provide more total interface for a given gas volume, but interfacial area alone is not the whole capture process. Particles must encounter and attach to the surface, remain associated as bubbles rise, and leave with a stable foam phase. Increasing air flow can improve contact yet also alter turbulence and foam drainage. A skimmer may have a useful operating region rather than a simple rule that ever more air must be better.

Equipment comparisons should report the bubble-generation method, such as venturi injection, impeller aspiration or diffuser, along with water and gas flow. A change to the injector could influence bubble size, mixing energy and actual gas supply simultaneously. If a trial reports improved removal after changing equipment, do not attribute the entire effect to one unmeasured feature. The language should match the controlled variables and measured outcomes.

20. Nominal contact time is not guaranteed capture time

A chamber with 300 litres of process-water hold-up and 150 L/min of throughput has nominal hydraulic residence time of two minutes. This is volume divided by flow, not proof that all water receives an identical two-minute encounter with bubbles. The actual residence-time distribution depends on chamber mixing, short circuits, baffles and recycle. More contact time may help some adsorption processes, but other phenomena such as foam collapse or accumulation can complicate a simple prediction.

A diagram may include a separate sump and internal recirculation loop. Adding a sump volume to the contact-chamber volume changes the reported number, but may no longer answer the question about bubble-water contact. Similarly, a pump’s internal recirculation flow should not be used as though it were new fresh water treated each minute. Name the volume and flow boundary before interpreting the ratio.

21. Water loss is itself a measurable consequence

Collecting liquid foam removes water from a recirculating system. If a skimmer exports 50 litres per day, the facility may need to replace an approximately corresponding amount after considering other inputs and losses. Replacement can change salinity, alkalinity, temperature and energy requirements. A higher wet skimmate volume is therefore not a free improvement. Good treatment comparisons describe how much water was lost per unit of target dry solids captured and how the replacement was managed.

A useful intensity metric is litres of skimmate per gram of measured dry solids. If 20 L of waste contain 100 g of dry solids, the ratio is 0.20 L/g. For another configuration, 10 L carrying only 30 g gives about 0.33 L/g. The first exports more dry mass per litre of waste under these simplified conditions. However, a wider comparison still needs inlet concentrations and treatment flow, so this ratio is not a standalone definition of overall efficiency.

22. Electricity consumption must be tied to the right output

A system operating at 0.5 kW for ten hours uses 5 kWh. If a properly measured trial recovers 0.25 kg of target dry solids, the electricity intensity is 5/0.25 = 20 kWh/kg dry solids captured. Another machine might process more water per hour yet capture a smaller fraction of the target. An energy-per-volume comparison and an energy-per-captured-mass comparison answer different questions, and neither can be replaced by a generic “efficient” label.

Maintenance and pumping losses also affect practical efficiency. A skimmer might rely on an external pump already counted in facility power, while another includes a dedicated aspirator. Define the energy boundary consistently; otherwise a favourable result may simply omit part of the consumption. Scientific translation preserves that system boundary and labels any excluded equipment.

23. Organic carbon and TSS should not be collapsed

Total organic carbon, dissolved organic carbon and chemical oxygen demand are different analytical quantities. DOC is often reported in mg C/L; COD can be reported in mg O₂/L. TSS is mass of suspended matter under a gravimetric convention. A foam fractionator may affect some of these measures differently. The fact that fine suspended material is captured does not prove all dissolved organic carbon or oxidizable compounds are captured at the same rate.

Suppose TSS falls 25 percent while DOC falls only 5 percent in matched tests. That is not a logical contradiction. It can reflect selectivity of capture and the distinct nature of the assays. Never average those percentages to create an invented overall removal score. The translated claim should state each constituent, measurement method and uncertainty separately. A reader may then decide which endpoint matters for a given water-quality objective.

24. Treatment component attribution needs controls

A recirculating aquaculture system usually has multiple treatment stages, including screening, foam separation, biological nitrification, gas exchange and sometimes disinfection. If samples are collected before all treatment and after all treatment, the overall water-quality change cannot automatically be credited to the fractionator alone. An isolated stage assessment needs appropriate inlet and outlet sampling or a controlled comparison of component operation, while considering changing feed and biomass conditions.

A fall in dissolved ammonia after a process adjustment especially needs care. Bubble capture may export organic nitrogen, but chemical ammonia conversion or biological filtration operates through distinct mechanisms. If ozone is used alongside the fractionator, oxidation can further confound attribution. The translation should identify the mechanism that was actually measured rather than treating every favourable downstream trend as proof of one device’s effect.

25. Ozone-assisted skimming adds chemical questions

Some units use ozone in the gas feed or contact treatment, potentially changing organic molecules, particle behaviour and oxidant chemistry. A study combining ozone and foam fractionation may report an improvement but cannot necessarily assign all of it to physical bubble capture. Residual oxidants, by-products and safe return-water conditions then become additional endpoints. In bromide-containing water, the possibility of bromine-derived products deserves attention as a separate chemical question.

A comparison of plain-air operation with ozone-assisted operation should state both treatment arms and which variables changed, including gas flow, reactor configuration and analytical method. If the authors did not measure pathogen reduction or fish exposure, the result should not be translated as proof of disinfection or welfare. A precise summary can acknowledge improved measured capture while declining to infer untested outcomes.

26. Replication and variability alter confidence

Suppose three trial removals are 12, 18 and 15 percent, with mean 15 percent. Another set has 20, 24 and 21 percent, with mean about 21.7 percent. The difference is 6.7 percentage points, not 6.7 percent relative improvement. Relative to the first mean, the increase is about 44.4 percent. These descriptive calculations say nothing by themselves about statistical significance, measurement uncertainty or whether the tests were independent.

A report needs to explain whether replicates are different production runs, dates, tanks or repeated assays on the same bottle. Repeatedly measuring one sample does not provide the same evidence as independent biological or operating replication. A larger mean may be promising, but a defensible conclusion records variation, design and confidence limits before claiming reliable superiority.

27. A useful troubleshooting sequence

When foam is sparse, consider changes in salinity, surface-active material, airflow delivery, injector function, cup height and cleaning. When foam is abundant but effluent TSS remains high, examine whether the foam is mainly water, whether the targeted particles attach to bubbles, and whether process flow bypasses the reactor. When waste cup volume increases, measure dry-solids mass before declaring improved removal. Each symptom requires its own diagnostic measurement.

An incident statement can say the waste volume increased following an adjustment while dry mass and outlet composition were not checked. That is informative and honest. The next test should compare the correct inlet and outlet on matched conditions, document operation and sample the waste stream. Appearance provides a starting hypothesis, not the endpoint of the investigation.

28. Independent exercises: concentration and flow

Exercise A: water enters at 6 mg/L TSS and leaves at 4.5 mg/L. Reduction is (6−4.5)/6×100=25 percent. At 12,000 L/h and equal inlet and outlet flows, the apparent load difference is 1.5×12,000=18,000 mg/h or 18 g/h. State the assumption about equal flows and the need to include waste discharge when creating a complete balance. A bare “18” without g/h is not a meaningful report.

Exercise B: a cup receives 4 L at 750 mg/L of a measured constituent. Its contents are 3,000 mg, or 3 g. The number 750 is concentration, not total collected mass. If one measured only cup volume, there would be insufficient information to calculate constituent mass. This exercise protects the critical denominator.

29. Independent exercises: water and power

Exercise C: a machine consumes 6 kWh during a trial and exports 0.30 kg of measured dry solids. Energy intensity is 20 kWh/kg. A second uses 5 kWh to export 0.20 kg, giving 25 kWh/kg. The first is lower on that particular ratio, but fair comparison requires similar water conditions, particle definition and time basis. It may not be superior if it exports costly replacement water or fails another important endpoint.

Exercise D: skimmate A is 900 mg/L at 2 L, and skimmate B is 400 mg/L at 6 L. Their analyte masses are 1.8 g and 2.4 g. The more concentrated skimmate is not the larger recovered mass. Explain why a separate treatment-water removal measurement and waste-volume cost still matter before recommending a setting.

30. Frequently asked questions

Does dark skimmate guarantee strong removal? No; test the analyte mass and collection interval. Does a fine-particle count reduction equal a TSS mass reduction? No. Can a foam separator replace a biofilter simply because both improve water quality? No; they have different mechanisms. Does marine capture efficiency automatically transfer to freshwater? No. Does a bigger waste cup prove more protein was removed? No. Does a concentration fall across a sidestream mean the whole tank fell by the same percent? No.

Can a skimmer be credited for changes across several treatment stages? Only with appropriate tests. Does ozone-assisted fractionation prove disinfection? Not without organism-specific evidence. Should manufacturers’ percent figures be discarded? They can be valuable when the analyte, method, conditions and uncertainty are recorded. The useful skill is to identify what they actually measure, rather than accepting or rejecting them wholesale.

31. A report another technician could reconstruct

Preserve water inlet and outlet locations, flows, waste collection route, target analyte, concentration basis, particle-size fraction, laboratory method and sampling times. Document salinity, temperature, bubble-generation approach and operating changes. Calculate concentration-based reductions and mass-rate changes separately; show the assumptions behind each. If a mass balance does not close, report the discrepancy and investigate internal hold-up, omitted outlets and uncertainty rather than inventing a perfect recovery.

For external communication, describe the device’s tested effect on a defined fraction under stated conditions. Avoid claims that it eliminates all organics, sterilizes water or necessarily cuts the biological-filter load by a particular percentage unless those endpoints were independently demonstrated. A clear result is much more useful when it can be checked and transferred cautiously to a different installation.