Passive Benzene Fenceline Tubes and EPA Method 325

Passive benzene fenceline monitoring uses sorbent tubes to collect volatile organic compounds from air over an extended period without a powered sampling pump. The laboratory measures the benzene recovered from the tube; a validated diffusive uptake rate and exposure time translate that recovered mass into an average air concentration. A passive tube does not record every minute of a release, and a single tube is not by itself a complete facility-emissions or legal-compliance assessment.

To read a result from United States EPA Methods 325A and 325B, check the tube identity, start and stop times, approved sampling configuration, uptake rate, laboratory blanks, field blanks, duplicates and site meteorology. Then distinguish the concentration measured at a monitoring location from the facility-impact calculation and the longer-term action-level procedure specified for covered petroleum refineries. Those are related but not identical quantities.

This guide explains passive diffusion from first principles, works numerical examples with nanograms, millilitres and cubic metres, diagnoses contaminated blanks and missing tubes, and teaches how to avoid mistaking one two-week measurement for a health or compliance verdict. Examples with concentrations are hypothetical unless identified as regulatory method specifications. Read the current EPA methods and governing regulation when an official determination is at stake.

What a passive sorbent tube actually samples

A passive sampling tube contains a sorbent material chosen to retain targeted volatile compounds such as benzene. During exposure, molecules move through a defined diffusion path driven by the concentration difference between ambient air and the sorbent uptake region. Air movement outside the diffusion opening affects the surroundings, but the calibrated geometry of the passive sampler creates an effective uptake rate under defined conditions.

The technique is attractive because many locations can be monitored over days without providing each sampling point with an electrical pump. Yet it is incorrect to say that every tube pulls a fixed physical airflow through itself. A passive uptake rate often has units of millilitres per minute as an equivalent air-sampling rate for the target compound. It is a calibration parameter associated with molecular transport and retention, not necessarily the motion of a metered air stream.

An exposed tube accumulates analyte mass over its deployment interval. At the laboratory, a properly prepared tube can be thermally desorbed and analysed using gas chromatography and a suitable detector. The amount attributed to benzene is corrected and qualified using the method’s calibration and quality controls. The final airborne concentration depends on the analytical mass, the compound-specific uptake rate and exposure duration.

A protective cover matters. Diffusive tubes installed outdoors must withstand weather while allowing representative contact with ambient air at their intended opening. Incorrect shelter geometry, obstruction, direct contamination or mishandling can bias the result. A field worker must not invent a cover or relocate an exposed tube halfway through the collection period without following the approved monitoring plan and documenting the alteration.

EPA Method 325A covers deployment and collection; companion Method 325B covers sampler preparation and laboratory analysis. This division is operationally important. A pristine chromatogram cannot repair an incorrectly timed field exposure, and perfect field placement cannot repair a contaminated laboratory blank. Both halves must be traceable before the reported concentration can be treated as dependable evidence.

The key equation and its units

In a simplified passive-sampling calculation, average concentration equals corrected collected mass divided by the product of uptake rate and exposure time. The denominator represents an equivalent sampled air volume. If collected mass is expressed in nanograms, uptake rate in millilitres per minute and time in minutes, the analyst must convert volume appropriately before presenting micrograms per cubic metre.

Suppose a hypothetical approved tube has an uptake rate of 0.40 millilitres per minute for benzene under its applicable conditions. Over fourteen days, exposure time is 14 times 24 times 60, or 20,160 minutes. The equivalent sampled volume is 0.40 times 20,160, or 8,064 millilitres. This is 8.064 litres and 0.008064 cubic metres. Every conversion belongs in the worksheet.

If the corrected net benzene mass is 40 nanograms, that is 0.040 micrograms. Dividing 0.040 micrograms by 0.008064 cubic metres gives approximately 4.96 micrograms per cubic metre. This is an illustrative fourteen-day mean under the assumed uptake rate; it says nothing about the highest one-minute benzene concentration during the interval. The result cannot be converted to a peak without additional time-resolved evidence.

An equivalent shortcut can be checked dimensionally: one nanogram per litre is one microgram per cubic metre. The 40 nanograms accumulated over 8.064 equivalent litres give 4.96 nanograms per litre, numerically equal to 4.96 micrograms per cubic metre. The shortcut is legitimate only because both mass and volume conversions are done consistently. A common thousand-fold error is to convert nanograms to micrograms but forget litres to cubic metres.

The uptake rate is not universal. Different sorbents, tube geometries, compounds, exposure periods and conditions may require validated rates. Do not take the 0.40 millilitre-per-minute teaching value and apply it to a different model or analyte. A rate determined for benzene does not automatically describe toluene or another volatile compound on the same sorbent.

Time integration changes the question

A fourteen-day passive result represents time-integrated uptake over the exposed interval, under the method’s assumptions. It does not reveal the exact chronology of peaks and troughs. Imagine two air histories: one holds a steady average level for two weeks; another has several brief higher exposures balanced by long lower periods. Both can lead to similar accumulated masses even though their short-term emission behaviour differs greatly.

This matters when interpreting a complaint about an odour or suspected short release. A passive tube may support a statement about average benzene during the deployment, but it cannot identify the moment of a transient event. Time-resolved instruments, operational records, meteorology and a suitable investigation may be necessary. A high integrated result justifies investigation; it does not automatically supply the event timing.

Start and stop times must be recorded accurately. A tube installed late or recovered early has a shorter exposure than planned. Using the nominal fourteen-day denominator for a thirteen-day tube underestimates the mean concentration if all other assumptions remain. Conversely, using too short an exposure duration can exaggerate the calculated concentration. Date labels alone may be insufficient if exact clock times are missing.

Time-zone discipline helps on facilities operating across daylight-saving transitions. Record timestamps with a clear zone or UTC basis and calculate actual minutes of exposure rather than treating every calendar transition as twenty-four hours without checking. A two-week campaign can span a clock change in some jurisdictions. The method’s elapsed sampling time is the relevant quantity for uptake calculations.

A missing deployment or damaged tube should not be replaced with a guessed concentration merely because a plot requires a continuous line. Explain why the result is absent and follow the monitoring plan’s invalidation or substitution rules. An observation that was never collected cannot become a compliant measurement through spreadsheet interpolation.

Why field blanks are part of the measurement

A field blank is a sampling tube treated like field equipment but kept capped as the method specifies. It travels through field handling and shipping without being intentionally opened for diffusive sampling. Its purpose is to reveal contamination associated with preparation, storage, transport and handling. It does not measure background benzene at the installation site in the same way as an uncapped exposed tube.

If the blank contains measurable benzene, the analyst must assess the result against the applicable method criteria. A small nonzero blank is not automatically grounds to delete an entire campaign, but a high blank can mean that apparent sample mass partly reflects contamination rather than ambient exposure. Method 325B prescribes checks and qualifying conditions; do not invent a blank subtraction rule for every situation.

EPA’s method describes field blank procedures including tubes from comparable batches and shipping history, with at least two field blanks and a ratio tied to the number of exposed tubes in a monitoring exercise. These procedural requirements exist because contamination can be systematic across the campaign. A single perfectly clean laboratory instrument does not compensate for a field container that contaminated multiple sorbent tubes.

Suppose a hypothetical field sample has 42 nanograms of benzene and an associated blank has 8 nanograms. One might be tempted to subtract 8 and publish 34 nanograms without further thought. Whether a particular adjustment is approved depends on the analytical method, quality-control outcome and laboratory reporting protocol. A blank is first evidence about potential contamination, not permission to perform an unreviewed algebraic correction.

An unusually large blank should prompt a record review: how were tubes conditioned, stored, capped, transported and handled? Were sample and blank tubes shipped together as directed? Did marking materials or incompatible containers introduce vapours? Did the laboratory’s instrument blank also show contamination? The goal is to identify where benzene entered the sample train, not simply hide an awkward result in a final table.

Duplicates, replicate precision and uncertainty

Two colocated duplicate tubes can help evaluate the repeatability of the combined field and analytical process over the same interval. If duplicate results are 4.5 and 5.1 micrograms per cubic metre, their mean is 4.8. Their absolute difference is 0.6. A common relative percent difference definition divides 0.6 by the mean 4.8 and multiplies by one hundred, giving 12.5 per cent.

Relative percent difference is not identical to analytical bias. Two duplicates can agree tightly while both carry the same uptake-rate error or systematic contamination. Conversely, a moderate difference could arise from real variability in deployment exposure, sampler handling or laboratory analysis. Precision is one component of data quality, not an all-purpose certification.

The laboratory’s calibration curve, recovery, instrument blank and method detection limit all matter. A concentration printed to three decimal places may look authoritative, but reporting precision must reflect the analytical method and uncertainty. Do not equate extra decimal digits with extra environmental certainty. A low result close to the detection limit may require a qualifier and explanation of what ‘not detected’ actually means.

‘Below detection’ does not mean zero benzene. It means the method did not establish an amount above a specified analytical threshold under the applied procedure. A nondetect can still be informative when the detection limit is sufficiently low for the study purpose. If the method detection limit is higher than a decision threshold, however, the absence of a positive identification cannot demonstrate that a lower threshold was met.

Uncertainty also enters through uptake rate and duration. A ten-per-cent error in assumed uptake rate directly affects the calculated concentration in the opposite direction when collected mass and time stay fixed. If the denominator is too large, concentration is biased low. This proportional relationship provides a useful independent check when a tube lot or uptake parameter is found to have been entered incorrectly.

Fenceline concentration is not facility impact

A tube installed at a property boundary can collect benzene from multiple contributing directions and sources. Some of what it measures may reflect regional or nearby background that is not attributable to the monitored facility. Meteorological data and the required station network provide context for evaluating potential facility influence. The concentration at one tube is therefore not automatically equivalent to the facility’s incremental contribution.

For covered US petroleum refineries, the regulatory fenceline programme uses method-specific calculations and a facility-impact quantity commonly denoted delta-c. The long-term action-level comparison is not simply one tube’s two-week benzene concentration. The regulatory method, station placement and calculation rules must be consulted in full before declaring an action-level exceedance.

EPA’s September 2025 enforcement alert discusses annual-average delta-c based on twenty-six recent fourteen-day periods and an action level of 9 micrograms per cubic metre for the relevant refinery requirements. This is a specified compliance action mechanism, not a universal population exposure standard, medical diagnosis or claim that 9 micrograms per cubic metre is safe for every individual or duration.

A facility may have a high reading at one station while upwind and downwind context changes through the period. A proper fenceline interpretation considers the measurement network and local meteorology. Wind direction by itself cannot reconstruct all individual source contributions, but ignoring it entirely can lead to unjustified attribution. EPA Method 325A explicitly incorporates local meteorological records into monitoring.

If a reporting system displays a single ‘facility impact’ number, verify how the site obtained it. Was it a location concentration, a difference between selected stations, a period-specific delta-c or an annual rolling aggregate? Do not transfer a regulatory threshold between these quantities because they share the same concentration unit. A unit match does not establish that the definitions match.

A worked rolling-period illustration

Twenty-six fourteen-day periods span 364 days in the simple arithmetic 26 times 14. In regulatory practice, actual sampling schedules and accepted periods follow applicable programme rules. A series of twenty-six hypothetical period-specific values can have an arithmetic average, but the correct annual calculation must use the official defined delta-c quantity and required treatment of valid or missing periods.

As a conceptual example rather than an EPA compliance determination, imagine twenty-five equally weighted accepted values of 4 micrograms per cubic metre and one value of 14. Their arithmetic mean is (25 times 4 plus 14)/26, or 114/26, approximately 4.38 micrograms per cubic metre. The one high episode changes the average but does not make the annual average equal 14. This example teaches averaging, not the full regulatory selection algorithm.

Replace the high observation with 144 and the mean becomes (100 plus 144)/26, approximately 9.38. That crossing of 9 is a mathematical example only. Whether a covered refinery must take specific action depends on whether the input observations are correctly computed delta-c values, how the method requires rounding and whether all relevant regulatory conditions are satisfied.

An analyst who calculates a rolling average should retain the full list of constituent periods, their accepted values and date ranges. A single dashboard number is insufficient for audit. When a new period enters the window, the oldest period leaves. Changes in the rolling mean may reflect both events, so it is misleading to attribute every increase solely to the newest sample.

The most useful communication separates ‘new two-week result’, ‘rolling average of accepted facility-impact periods’ and ‘current action status’. Each term has different meaning. People who receive a public report should not need to reverse-engineer a regulatory formula from unexplained coloured badges or a vague statement that benzene has ‘passed a limit’.

Diagnosing unexpected high values

Suppose one station’s fourteen-day concentration rises substantially while other stations remain closer to their previous levels. Begin with metadata. Was the tube placed and retrieved at the intended time? Was the cap fully removed? Did the shelter change? Were there unusually high blank masses, unresolved duplicates or analytical flags? A large number cannot be attributed confidently until basic sample validity is established.

Next look at environmental context. Which winds prevailed, were there maintenance activities nearby, and is the elevated monitor located close to another potential source? These are hypotheses to investigate, not explanations to declare without evidence. A passive tube integrates across many wind regimes, so a single wind arrow on a map does not represent the full exposure period.

A comparison across tube locations should preserve each station’s actual valid period. If an analyst pairs a fourteen-day result with a neighbouring sample exposed only twelve days, both may be reported as average concentrations when corrected properly, but missing time coverage can affect representativeness. A location-to-location difference is not automatically a pure spatial gradient if the temporal coverage differs.

A repeated high result across consecutive periods may carry greater weight than a single anomalous result, especially when blanks, duplicates and meteorology are sound. But repetition does not excuse poor chain-of-custody documentation. Conversely, a single verified elevated concentration can still be important. Data quality and environmental importance are independent axes, and a proper report addresses both.

If an independent laboratory provides a different concentration for the same tube, check whether the comparison is analytically legitimate. Different laboratories may use distinct calibration details, reporting limits or blank handling. Duplicate analysis can help, but differences need to be traced to mass quantification, sorbent recovery, sample handling and uptake assumptions rather than settled by choosing the lower value.

Reader transfer problems and answer checks

A tube is exposed for exactly seven days at a validated equivalent uptake rate of 0.5 millilitres per minute. The equivalent volume is 0.5 times 7 times 24 times 60 millilitres, or 5,040 millilitres, equal to 5.04 litres. If the corrected collected mass is 25.2 nanograms, the concentration is 25.2/5.04, or 5 nanograms per litre and therefore 5 micrograms per cubic metre. The arithmetic requires the real method’s validated rate.

Another analyst accidentally uses 20,160 seconds instead of 20,160 minutes for a fourteen-day sampling period. This is not merely a formatting error: a rate expressed per minute must be multiplied by elapsed minutes. The mismatch can create a factor-of-sixty mistake. Write the unit next to each intermediate term and check that uptake-rate units cancel time, leaving an equivalent sampled volume.

A technician reports that one tube collected twice the corrected mass of a neighbouring tube. Does that prove twice the air concentration? Only if their validated uptake rates, durations and relevant conditions lead to equal equivalent sampled volumes. If one exposure lasted twice as long, doubled mass could reflect equal concentration. The numerator alone cannot establish relative airborne concentration.

A campaign has all sample results below the laboratory reporting limit but field blanks reveal significant benzene contamination. It is not defensible to state that the environment contained no benzene. The analytical evidence may be constrained or qualified by contamination and reporting thresholds. The correct next action is to review method validity and report the limits transparently.

A reader sees a refinery fenceline value above 9 micrograms per cubic metre for a single fourteen-day tube and announces that the annual facility-impact action level has been exceeded. The number and unit happen to resemble the action threshold, but they do not define the same statistic. The appropriate comparison requires the specified delta-c procedure and rolling period under the governing rules.

Design a deployment that answers the right question

Before installing the tubes, decide whether the purpose is a regulatory refinery fenceline programme, exploratory source screening, industrial hygiene or a neighbourhood exposure survey. Those uses may all involve passive sorbent technology, but they do not share every placement rule, duration, reporting obligation or decision threshold. A method validated for a particular refinery programme should not be portrayed as automatically approving every possible off-site sampling design.

The map of monitoring locations is not a decoration. It should document property boundaries, relevant equipment, surrounding roads, potential background influences and practical access. Shelters must be placed so that they can sample air as intended without being obstructed by walls, exhaust vents or nearby objects. The correct number and location of stations come from the governing method and monitoring plan, not a convenient count chosen after the readings are known.

Sampling personnel need a clear plan for bad weather and inaccessible locations. Replacing a missing tube with one collected from another location changes the spatial question. Deploying an extra tube may be useful, but the distinction between planned primary, duplicate and supplementary samples must be preserved. If a station is moved, record the reason and exact change so subsequent campaigns are not falsely presented as a stable long-term series.

Unique sample identifiers protect the chain of evidence. A sound record connects tube, station, exposure period, sampler type, collection and recovery personnel, laboratory receipt and analytical result. The same identifier should appear in field notes and lab documentation. A manually transcribed result without that connection may look plausible but is much harder to audit when unusual values appear.

Field workers should record actual exposure times and ambient observations, not merely print ‘fourteen days’ beside every sample. A start time of 10:15 and recovery at 08:45 two weeks later gives less than a full nominal interval. Using exact elapsed minutes gives the correct exposure denominator. The note should also distinguish a tube intentionally capped during part of an interval from one that remained continuously open.

Examine a chromatographic result before the final concentration

The analytical laboratory must identify benzene in a mixture of compounds retained on the sorbent. Gas chromatography separates constituents over time, while an appropriate detector supports the compound’s quantification and, depending on method, identification. A peak in a chromatogram is not automatically benzene merely because it appears near the expected retention time. Method-specific identity criteria, standards and interference checks remain important.

Calibration links detector response to analyte mass. If a laboratory standard curve is established in nanograms per tube, the output mass should be within its validated range. A sample outside the calibration interval may need approved dilution, reanalysis or qualification. Extrapolating a calibration straight line beyond its support can produce an impressive but unreliable decimal value.

Recovery is another distinct concept. The laboratory attempts to desorb analyte from the sorbent, and the efficiency of that step is evaluated under the method’s validation requirements. A tube with poor recovery can yield a deceptively low apparent airborne concentration. A sorbent can also approach its validated capacity or experience breakthrough under unsuitable conditions, making uptake-rate assumptions unreliable.

Detection limits, reporting limits and action thresholds serve different purposes. An analytical detection limit concerns a system’s ability to distinguish analyte from measurement noise. A reporting limit addresses how results are formally quantified or reported. A regulatory action threshold is a separate decision rule for a defined quantity and averaging period. None can be replaced by the others merely because all are expressed in concentration units.

When reading the laboratory certificate, identify flags that accompany values. Estimated concentrations, blanks above criteria, out-of-range calibration or unusual duplicate differences may all be represented by short codes. A chart-maker who imports only the number column discards meaning. Maintain a dictionary of the qualifier codes and display them where they affect whether a result can support the intended claim.

Test competing explanations with a mini case file

Imagine two passive tubes installed at opposite parts of a facility boundary. The first has an accepted fourteen-day benzene concentration of 6 micrograms per cubic metre and the second 2. The difference is 4, and the ratio is three. It would be premature to attribute all 4 to the facility. Both observations may contain background, and the time-integrated local wind and source context are not yet considered.

Suppose local wind records show that the higher station was frequently downwind during operating periods. This supports a source hypothesis, but does not prove the exact emission location. Other sources may contribute, and the surrounding wind field can change during two weeks. A source-specific investigation would cross-check activity logs, meteorology, repeated spatial patterns and applicable monitoring analyses.

Now suppose the high station’s field blank was unusually contaminated and the lower station’s blank was acceptable. The comparison cannot simply be used as before. The result may require qualification under Method 325B; an apparent environmental contrast is entangled with sample validity. The right sequence is to address contamination evidence before ranking stations by apparent concentrations.

A third case has two colocated duplicate tubes reading 5.0 and 5.2. They agree closely, but both were exposed for a documented thirteen days while spreadsheet formulas assumed fourteen. Correcting the duration raises the concentration by 14/13, about 7.7 per cent, under the simplified assumptions. Duplicate agreement would not reveal the shared duration mistake. This example explains why independent procedural checks are essential.

Consider instead two laboratories analysing separate tubes from the same site and time, with results 5 and 7 micrograms per cubic metre. Before alleging misconduct or bad calibration, examine duplicate variability, blank handling, sampler preparation, uptake rate, sample mass, potential storage losses and laboratory qualifiers. A difference is a signal to investigate, not by itself a diagnosis of which step failed.

Translate chemistry without conflating it with regulation

Benzene is a specific volatile aromatic compound, and its environmental importance is not conveyed merely by saying that ‘hydrocarbons were found’. Different VOCs behave differently during sampling and have different toxicological contexts. A tube method intended for benzene can detect other VOCs with suitable validation, but one compound’s uptake rate and analytical response should not be substituted for another’s.

Likewise, emission rate and fenceline air concentration are distinct quantities. A facility can emit a mass per unit time; the concentration at a tube depends on mixing, transport, meteorology and other influences. A passive sample measures concentration at its location, not the facility’s total mass emission over the period. Inferring a release rate would require an additional source or dispersion model and suitable evidence.

An exposure assessment introduces another layer. A passive tube at the property edge cannot automatically estimate what a particular resident inhaled indoors, at work or during travel. Individual exposure depends on location and time. Even when a measured fenceline concentration is important, it should be communicated as a monitoring observation rather than an individual dose assessment.

Readers should be cautious about describing an action level as a universal safety line. EPA’s 9-microgram-per-cubic-metre refinery fenceline action level belongs to a particular regulatory calculation and set of obligations. A figure below that line is not a blanket declaration of no health concern, and a single sample above it is not automatically the annual programme result. Context is part of numerical accuracy.

A public summary can explain both facts simultaneously: the sampling method detected benzene at a reported concentration, and the official regulatory response depends on the full defined facility-impact method and rolling assessment. Separating those claims is not evasive. It is the only way to avoid assigning a stronger meaning to a number than the measurement supports.

Why the uncertainty can change between campaigns

Passive uptake relies on laboratory validation, exposure duration and external environmental conditions. High or low temperature, strong local concentration gradients, tube geometry and sorbent behaviour can alter performance under conditions that need assessment. The method specifies requirements so that readings remain interpretable. Monitoring teams should record departures, not silently presume every environment behaves like a calibration chamber.

Background contamination can change across shipments even if the sorbent batch is consistent. Compare the correct field blanks from each exercise, as well as laboratory blanks, instead of applying last month’s blank result to this month’s samples. A campaign-level contamination problem can bias multiple monitoring locations together and thereby distort comparisons without causing an obvious single outlier.

Uncertainty in a reported annual aggregate draws on repeated sampling periods. Random errors may partially average across periods, but a persistent mistake in the uptake rate or sample-time bookkeeping affects many results in the same direction. Increasing the number of monitoring periods does not repair a shared systematic error. An audit should test both day-to-day variations and repeated procedure assumptions.

The apparent number of valid periods matters as much as the number scheduled. If one of twenty-six expected results is missing, do not act as if the rolling programme necessarily contains twenty-six interchangeable measurements. The governing rule determines how missing observations are treated. Any summary should disclose what was obtained, accepted, rejected or estimated rather than converting an incomplete collection into a seamless chart.

An uncertainty statement might identify the approximate method detection limit, qualified samples, duplicate results, blank behaviour, validated rate and exposure timing. It should avoid promising a numerical total uncertainty without the necessary validation inputs. Sometimes the honest conclusion is that the observed contrast is larger than known routine variation; at other times the evidence cannot discriminate between competing explanations.

Independent exercises and correction keys

Exercise one asks for the equivalent volume when uptake rate is 0.30 millilitres per minute and exposure lasts 10,080 minutes. Multiplying gives 3,024 millilitres, or 3.024 litres. A net mass of 15.12 nanograms then corresponds to 15.12/3.024, or 5 micrograms per cubic metre. Notice that the equivalent air volume here is smaller than in a fourteen-day example; retaining the old denominator would be a serious error.

Exercise two asks for the effect of a mistaken exposure time. A sample is truly exposed for ten days, but the worksheet assumes twelve. At fixed corrected mass and uptake rate, the calculated concentration using twelve days is ten-twelfths of the correct concentration, an underestimation of about 16.7 per cent. Conversely, correcting the original twelve-day-based result to ten days multiplies it by 12/10, an increase of 20 per cent. Different denominators explain the two percentages.

Exercise three asks whether two tubes reading 3 and 6 micrograms per cubic metre demonstrate a doubling of emissions. They do not. The readings are average concentrations at their monitoring positions and times. Different background and dispersion conditions, along with sample validity, matter. The only safe immediate comparison is that the second recorded concentration is twice the first, with qualifications if methods or periods differ.

Exercise four asks why a reported benzene result close to a laboratory reporting limit should be treated cautiously. Small analytical backgrounds can matter proportionally more at low recovered masses. The laboratory may qualify the result as estimated or not quantifiable under its reporting convention. The right answer is to state the limit and qualifier, rather than interpreting the extra decimal digits as proof of exact environmental concentration.

Exercise five asks for a final sentence suitable for a community report. An acceptable example is: ‘During the specified fourteen-day interval, the validated passive tube at Station A reported an average benzene concentration of 4.96 micrograms per cubic metre; field and laboratory quality checks and the facility-impact calculation should be reviewed separately.’ That sentence is specific without asserting an unproven source or compliance verdict.