How to translate aerosol mass spectrometer (AMS) data easily begins with one distinction: a number labelled non-refractory PM1 is not automatically the same measurement as total PM1, PM2.5 or black carbon. The AMS reports a particular operationally defined fraction of particles that travel through its inlet and vaporize at its heated surface. A correct interpretation follows the particle path rather than treating every mass concentration as interchangeable.
The mechanism can be understood without memorising jargon. An aerodynamic lens focuses sampled particles; a chopper and particle time-of-flight system add size information; a heated vaporizer and electron-impact ionization turn evaporated molecules into fragment ions. Software converts those signals into estimates of organic, sulfate, nitrate, ammonium and chloride mass, but that conversion requires calibration, background subtraction and assumptions about collection efficiency.
This guide explains how to read those assumptions, test the arithmetic and separate observed signal from inferred composition. All numerical examples are explicitly illustrative rather than measured observations. The aim is transferable reasoning: after reading a graph, you should be able to tell somebody which particles could be missing, which units have been used, and what independent evidence would justify a conclusion.
The first translation: what the instrument actually measures
An aerosol mass spectrometer does not weigh a scoop of air on a balance. It samples an aerosol flow, selects particles according to the characteristics of an aerodynamic lens, heats particles reaching the vaporizer, ionizes material that evaporates, and translates a spectrum of mass-to-charge peaks into a chemically resolved concentration. The phrase non-refractory PM1 is consequently a measurement definition, not a synonym for every particle smaller than one micrometre. In many applications it covers organics, sulfate, nitrate, ammonium and certain chloride-containing material, while refractory black carbon, mineral dust and many sea-salt components are not fully quantified by the usual instrument configuration.
The first question on any plot should therefore be: what entered the lens, what reached the vaporizer and what material the instrument could detect? If the answer is unclear, a comparison to a filter-based PM2.5 result is at best provisional. A PM2.5 sampler uses a different size inlet and can include coarse material between PM1 and PM2.5. Its filter may also gain or lose semivolatile compounds depending on temperature, humidity, handling and chemistry. Calling a disagreement an instrument failure before checking these distinctions confuses two definitions with a single truth.
Follow a particle rather than memorising instrument parts
Imagine a particle suspended in air approaching the AMS inlet. The inlet and lens are built to transmit a useful range of particle sizes into a lower-pressure region. The particle continues as a beam, while most gas molecules are pumped away. A mechanical chopper may periodically interrupt the beam to provide the timing reference needed for a particle time-of-flight size measurement. Further downstream, a heated surface causes suitable particle constituents to evaporate. Electron-impact ionization creates positive fragments, and the mass spectrometer sorts ions according to mass-to-charge ratio.
Each stage has a physical purpose and a potential loss. Inlet losses can exclude particles before detection; aerodynamic focusing can favour some diameters; particles can bounce instead of sticking to the vaporizer; material that does not volatilize under the operating conditions can be underrepresented; electron-impact ionization can fragment a parent molecule into several smaller ions. The interpretation is strongest when its author can state which of these processes was corrected, which was assumed and which remains unknown. A colourful stacked plot hides this chain, but it cannot make it disappear.
Non-refractory does not mean chemically harmless
The adjective non-refractory tells you about volatilization and the operational measurement, not toxicity or safety. A particle component may evaporate at the instrument’s heated surface but persist long enough in ambient air to influence visibility, climate or exposure. Equally, a constituent not detected efficiently by this setup is not absent from the atmosphere. Refractory soot and crustal minerals may be important in the same air mass. Use the precise phrase measured non-refractory aerosol mass rather than total pollution load when describing the output.
A technically accurate sentence can remain approachable: ‘The instrument observed a rise in the fraction of sampled fine particles that it can vaporize and quantify; it did not independently measure every particulate constituent.’ This translation has more explanatory value than declaring that the mass spectrometer measured all PM1. It also alerts a reader to ask for independent black-carbon, elemental, gravimetric or optical measurements if a broader composition is needed. The missing fraction is not necessarily small, constant or of the same chemical origin on every day.
Read the concentration denominator before interpreting a number
Concentrations are often reported in micrograms per cubic metre of sampled air, written µg/m³. An aerosol mass concentration is a mass estimate divided by a sampled volume, not a percentage of particles or a count of individual aerosol objects. If the instrument draws a controlled volumetric flow, the flow, sample time, gas conditions and any corrections belong in the measurement record. One must determine whether the volume corresponds to ambient temperature and pressure or a stated reference condition before mixing datasets from different systems.
Consider a purely illustrative 15-minute record giving 8 µg/m³ organics. It does not mean each particle contains eight micrograms, and it does not mean eight per cent of the total particulate mass is organic. The fraction only becomes calculable after identifying an appropriate common denominator, such as the contemporaneous measured AMS non-refractory mass, with matching sampling windows. If organics equal 8 µg/m³ and the summed measured non-refractory components equal 20 µg/m³, the organic fraction of that operational sum is 8/20 = 0.40, or 40%. It is not automatically 40% of PM2.5.
Particle size: vacuum aerodynamic diameter is a qualified language
The time between chopper opening and detection can provide a particle flight-time signal related to its velocity. Conversion to size depends on calibration and the relevant aerodynamic definition. The vacuum aerodynamic diameter obtained from an AMS is not automatically equivalent to electrical mobility diameter from an SMPS or aerodynamic diameter determined under another flow regime. The relation depends on particle effective density, shape and other assumptions. Saying two peaks have the ‘same size’ when their x-axes describe different physical quantities can create an artificial conflict.
For practical communication, preserve the instrument’s axis term in the caption and translate it immediately: the size is inferred from particle motion in the instrument’s low-pressure beam. If comparing it to an SMPS peak, first discuss the conversion model, calibration particles and any expected density differences. A highly porous soot agglomerate, for instance, should not be converted with the same unexamined spherical-density assumption as a compact inorganic salt. The correct reader task is to recognize related, but not interchangeable, diameter metrics.
Why a mass spectrum is not a catalogue of whole molecules
Electron-impact ionization deliberately creates fragments. A signal at one mass-to-charge ratio can therefore be contributed by different chemical parents, and a single parent may distribute signal across several ion channels. Software fragmentation tables and high-resolution measurements help attribute and aggregate ions into component families. They do not turn every peak into a unique molecular structure. When a report identifies organic aerosol, it usually means a measured and algorithmically apportioned organic component, not that every organic molecule has been structurally identified.
This distinction matters for translation and education. The term fragment is not a failed measurement; it is a normal product of the ionization method. The term species is context-sensitive: AMS component classes such as sulfate and nitrate are chemical classifications inferred from a set of ion signals, whereas molecular speciation would name individual compounds. A responsible description explains the level of chemical specificity actually supported. It is possible to measure robust sulfate mass without being able to state which sulfate salt supplied every sampled particle.
Sulfate channels and the problem of overclaiming source
Sulfate-associated ion fragments may show a clear increase while meteorological conditions change. The instrument can support a conclusion that the measured non-refractory sulfate component increased over the recorded interval. It cannot, without other evidence, specify a particular power station, ship, industrial facility or long-range transport pathway. Sulfate can arise through atmospheric oxidation and processing under different conditions, and the same measured end product may have multiple plausible histories. The observation and source attribution are distinct logical steps.
To move from composition to a defensible source argument, add air-mass trajectories, co-pollutants, meteorology, independent sulfate filters, regional emissions knowledge and uncertainties. Consider competing hypotheses rather than using a single coincident time series as proof. If sulfate rises on a humid afternoon, the next question is whether the sampling and particle chemistry altered, whether a regional air mass arrived, or whether local oxidation and partitioning changed. A well-translated graph frames the diagnostic possibilities before proposing a source.
Nitrate signals and temperature-sensitive partitioning
Particulate nitrate can be particularly sensitive to gas-particle partitioning and sampling conditions. The AMS signal for nitrate should not be silently equated to total atmospheric nitrate, because gas-phase nitric acid and particulate nitrate are different reservoirs. Volatile ammonium nitrate may also redistribute when air is warmed, dried or diluted. A concentration decrease across an instrument inlet need not exactly reproduce an outdoor chemical loss; sample-line conditions could matter. Interpret a pattern with this operational qualification attached.
Consider two illustrative consecutive intervals with nitrate mass concentrations of 6 and 3 µg/m³, while sampled air temperature rises. A simple plot shows a 50% reduction in measured particulate nitrate. It does not independently establish that half of the ambient nitrate was chemically destroyed. The nitrate may have repartitioned into gas phase, the incoming air may have changed, or collection efficiency may differ. The next diagnostic should include temperature, relative humidity, gas measurements where available, total aerosol mass and the instrument’s collection-efficiency setting.
Ammonium, neutralisation and equivalent units
Ammonium is often interpreted alongside sulfate and nitrate, but the units need care. A mass ratio of ammonium to sulfate does not directly represent the electrical charge balance because ammonium, sulfate and nitrate have different molar masses and valences. To compare equivalents, convert each mass concentration to amount of substance and multiply by the absolute ionic charge. For sulfate, divide µg/m³ by approximately 96.06 µg/µmol and multiply by two. For nitrate, divide by about 62.00 µg/µmol; for ammonium, divide by about 18.04 µg/µmol.
In an illustrative sample containing sulfate 4.8 µg/m³, nitrate 3.1 µg/m³ and ammonium 2.7 µg/m³, sulfate charge demand is about 2 × 4.8/96.06 = 0.100 µeq/m³, and nitrate demand is 3.1/62.00 = 0.050 µeq/m³. Ammonium provides about 2.7/18.04 = 0.150 µeq/m³. These are approximately balanced under the restricted assumption that these three ions dominate the relevant charge inventory. The match does not prove the absence of other cations, acidity, measurement error or different phase distributions.
Organic aerosol is a component, not one chemical
Organic aerosol collects many carbon-containing compounds and transformation products under a common analytical label. Different environmental sources and atmospheric aging processes can produce overlapping fragment patterns. An AMS organic mass estimate can be highly useful for time trends and composition, but the word organic here does not automatically mean natural, benign or one identifiable substance. Organic mass also differs from organic carbon mass because organic molecules contain oxygen, hydrogen, nitrogen and other elements as well as carbon.
Suppose an illustrative report gives organic mass of 9 µg/m³ and independently inferred organic carbon of 5 µg C/m³. Dividing gives an OM/OC ratio of 1.8 on that basis. This is an illustrative ratio, not a universal correction to apply to every aerosol. The ratio can vary with oxidation state and source mixture. If a graph instead shows a signal at m/z 44 or another marker, do not call that signal a direct measurement of all oxidized organic aerosol without explaining how it was normalized and interpreted.
What collection efficiency means in plain English
Collection efficiency, commonly abbreviated CE, is the effective fraction of relevant sampled particle mass that the instrument ultimately registers through its collection and vaporization pathway, given how its calibration and response are defined. Some particles strike the hot surface and bounce, leaving less signal than their incoming mass would suggest. The correction is consequential: where a raw response must be divided by CE, a smaller assumed efficiency produces a larger concentration estimate. CE should not be described as the total efficiency of an entire urban monitoring programme.
For an illustrative corrected component concentration, start with a raw calibration-scaled signal corresponding to 4 µg/m³ before CE adjustment. Dividing by CE = 0.50 yields 8 µg/m³; dividing by CE = 0.80 yields 5 µg/m³. The difference between these two assumptions is 3 µg/m³, or 60% of the 5 µg/m³ result. It would be misleading to present both outputs as independent observations. They are alternative calculations based on one signal and two assumptions. Record which correction was used before making spatial or temporal comparisons.
When a default collection efficiency becomes inappropriate
A fixed default can be a useful operational starting point in some instruments and studies, but it is not a law of nature. Collection efficiency may vary with particle phase, chemical composition, relative humidity and behavior on the vaporizer. The NASA-indexed research by Middlebrook and colleagues evaluates how composition-dependent efficiency estimates can improve comparisons against independent fine particle measures. Its context matters: a result from one set of campaigns is not permission to transplant the same numerical correction into every new climate, altitude or aerosol regime.
A reviewer should ask whether the analysis software applied a constant CE, a composition-based CE, or a size-dependent approach; whether the investigators documented the justification; and whether results remain qualitatively robust under reasonable alternatives. If the main conclusion reverses merely by moving the allowed correction assumption within its uncertainty range, the conclusion should be described as conditional. A transparent sensitivity table may be more useful than an unjustifiably precise headline concentration. Precision of decimals is not certainty of the physical model.
Ionization efficiency and why calibration must be described
Signal intensity is not equivalent to mass until the response factor is understood. AMS calibration commonly establishes ionization efficiency using a known test aerosol and instrument-specific procedures, with relative ionization efficiencies for species handled within the analysis. The instrument response can change with source settings, detector condition, flow and maintenance. Thus a long time series should include calibration history, not just final mass charts. A seemingly small systematic shift can masquerade as a seasonal trend when the dataset is combined without normalization.
Imagine a known challenge mass produces 900 arbitrary signal units during one calibration and 990 during a later calibration under otherwise equivalent conditions. That 10% response change signals a need to inspect calibration, detector or transfer conditions; it is not proof that ambient aerosol concentrations increased by 10%. The correct interpretation separates the controlled challenge from the environmental sample. Keep the calibration basis, dates and uncertainty notes with the data. If the algorithm applies different relative efficiencies across components, those settings should also travel with the graph.
Airbeam, detection background and blank-like periods
Many AMS processing procedures use gas-phase signals, including an airbeam proxy, to monitor sensitivity or normalize instrument changes. The operational details depend on the specific mass spectrometer. Data interpretation should separately identify electronic background, gas-phase residual signals, particle-related spectra and any intervals when the chopper blocks the particle beam. Subtraction can improve a chemical estimate, but subtracting the wrong background or mixing settings across runs can produce negative or implausible component estimates.
A negative small calculated value should not be treated as proof that a component physically has negative mass. It often indicates a background-corrected measurement near the detection capability or uncertainty. A reporting approach may retain signed values for statistical analysis while explicitly noting detection limits, rather than arbitrarily replacing every negative with zero and biasing an average upward. The choice depends on the measurement protocol. The reader’s task is to find the applied blank and noise conventions before calculating means, uncertainty intervals or apparent removal efficiencies.
How to read a stacked chemical-composition time series
A stacked area chart may place organics, sulfate, nitrate, ammonium and chloride in different colours. At any instant, the vertical sum equals the plotted collection of AMS components, not total atmospheric particulate mass unless the author has independent grounds to say so. If a plotted organic layer expands while sulfate stays stable, an increasing organic proportion might reflect an absolute organic rise, a fall in the other components, or both. The story depends on comparing component concentrations as well as percentages.
Always record the averaging period, missing-data treatment and time zone. A single sharp spike may represent seconds of plume passage, several minutes of averaged transport, or an interpolation artifact. If a large-data gap is joined by a line, do not infer a continuous physical transition through unmeasured time. When investigating events, compare aligned meteorological and chemical records, acknowledge instrument response and avoid attaching a specific source label to a graph solely because a peak looks recognizable. A descriptive reading should survive independent checking by someone who did not produce the plot.
Worked example: calculate a clear composition sum
Take an invented, contemporaneous AMS record: organics 7.2 µg/m³, sulfate 4.0, nitrate 3.0, ammonium 2.1 and chloride 0.2 µg/m³. The measured non-refractory sum is 7.2 + 4.0 + 3.0 + 2.1 + 0.2 = 16.5 µg/m³. Organics are 7.2/16.5 = 0.436, or about 43.6% of that sum. The denominator is explicitly the AMS non-refractory sum. It would be an error to write ‘43.6% of all air pollution’ or ‘43.6% of PM2.5’ without an independent matching measurement and definition.
Now suppose a contemporaneous independent PM1 mass estimate is 19.0 ± 2.0 µg/m³. The simple difference is 2.5 µg/m³ and the AMS sum divided by that comparison value is about 0.868. This ratio alone does not identify the missing 13.2%. Refractory constituents, inlet transmission, differing humidity, instrument uncertainties and species-dependent recovery are plausible contributors. If PM1 is derived from a different physical model rather than direct gravimetry, that model brings its own uncertainty. A good worked solution states the difference first and interprets it only after auditing the measurement boundaries.
Worked example: sensitivity to CE without double-counting
Assume the signal before a collection-efficiency correction corresponds to 5.4 µg/m³ of one component under the adopted response calibration. At CE = 0.60, the corrected estimate is 5.4/0.60 = 9.0 µg/m³. At CE = 0.75, it is 7.2 µg/m³. The first is 1.8 µg/m³ greater than the second. Relative to the 7.2 result, the sensitivity is 25%. Both are calculations of one underlying signal rather than two parallel ambient samples.
If a report presents a 9.0 ± 0.2 µg/m³ headline based only on short-term random fluctuations while CE remains poorly constrained between 0.60 and 0.75, it underrepresents the model uncertainty. A better report keeps the measurement noise, calibration uncertainty and CE sensitivity separate. Some errors may be correlated across species because they share instrument factors. Combining every uncertainty by naive addition or pretending all uncertainties are independent can be wrong. The aim is a defensible uncertainty statement with named assumptions rather than a cosmetically tidy margin.
Worked example: mass ratios require matching windows
Suppose sulfate rises from 3 to 5 µg/m³ across an hour, but organics are available only as an average over a wider three-hour interval. Dividing the sulfate peak by the three-hour organic mean will not give a valid instantaneous sulfate-to-organic ratio. If only comparable interval means are available, recompute both over the same windows and describe how missing subintervals were treated. Averages of ratios and ratios of averages are generally not equal, especially when denominators vary substantially.
An illustrative two-point calculation makes the distinction visible: in two equal-duration periods sulfate values are 2 and 6 µg/m³, and organic values are 2 and 12 µg/m³. The individual ratios are 1 and 0.5, whose simple average is 0.75. The ratio of the equal-duration average masses is [(2+6)/2]/[(2+12)/2] = 4/7 ≈ 0.571. These answer different questions. Before writing ‘sulfate was three-quarters of organic mass’, state whether the intended claim concerns the average instantaneous ratio or bulk summed mass over the full period.
A diagnostic path when AMS and SMPS disagree
AMS and SMPS measurements can complement one another because AMS emphasizes chemically resolved non-refractory mass while SMPS infers number size distributions according to electrical mobility. Turning SMPS number distributions into mass requires a particle shape and effective-density model, as well as size-range alignment. Even perfect agreement should not be interpreted as chemical proof: two estimates can agree through compensating errors. Substantial disagreement can be informative if the instruments measure genuinely overlapping populations and their processing choices are independently documented.
Start with clock synchronization, inlet position and sampling conditions. Next compare size transmission ranges, the SMPS inversion and multiple-charge correction, the assumed effective density, AMS CE and whether refractory mass could matter. Check for a changing plume, humidity effects and particle growth. If possible, compare integrated AMS size distributions and the mobility distribution using a carefully stated conversion, not visual inspection of incompatible x-axis labels. Report what remains unmatched after these checks. This is more useful than saying one instrument is ‘wrong’ because its line does not sit on top of the other.
A diagnostic path when filter chemistry disagrees
A filter-ion chromatography system may report sulfate, nitrate and ammonium over a long sample. The AMS can provide higher time resolution, but the comparison requires synchronized mass windows and awareness of losses, blanks and analysis differences. Filter nitrate can be affected by semi-volatile partitioning; extracted ions represent operationally collected and recovered material, while AMS response depends on particle transmission and vaporization. A disagreement in ammonium equivalents may also suggest non-ammonium cations, which are not always well characterized by a standard AMS component set.
If the filter sulfate is 5.0 µg/m³ and the time-matched AMS sulfate is 4.3 µg/m³, the AMS/filter ratio is 0.86. That ratio is a useful diagnostic number, not a calibration verdict by itself. Ask whether the filter sample covers the same size fraction, whether field blanks were subtracted, whether the AMS instrument had valid calibration and whether the filter was subject to contamination or loss. Repeat the comparison across many paired samples and inspect residuals before deciding whether there is a persistent instrumental bias.
Humidity, phase state and sample conditioning
Relative humidity can influence aerosol water uptake, particle phase and potentially the conditions affecting collection efficiency. An instrument may measure a dried or otherwise conditioned sample that differs from wet ambient particle properties. Water content reported in a different system may not be directly included in the AMS non-refractory species sum. A humidity-dependent increase in light scattering and a nearly flat AMS dry-mass trace can coexist without contradiction, because scattering responds strongly to particle size and refractive properties as well as dry chemical material.
When translating humidity-related data, name the conditioning at the inlet and the conditions used by the optical or chemical system. An outdoor optical sensor and a laboratory-calibrated AMS should not be presumed to share the same aerosol-water definition. In a very humid period, distinguish genuine uptake of condensed water from changes in dry chemical components and from changes in instrument transmission. The source of uncertainty is a mismatch of physical observables, not an awkward footnote to be dropped for a cleaner story.
Temporal attribution: what a spike can and cannot show
Suppose an AMS nitrate trace rises abruptly at the same time as a nearby roadway traffic index. That is a temporal association, not an established source fingerprint. Nitrate formation depends on multiple atmospheric precursors and chemical pathways, and a roadway can coincide with evolving mixing height or arrival of a regional plume. The correct next step is to inspect meteorological trajectories, nitrogen-oxide observations where applicable, supporting carbon or aerosol markers and appropriate chemical-transport context. A brief spike in nitrate can also be influenced by conditions controlling gas-particle partitioning.
For narrative clarity, separate three sentence types. Observational: ‘Measured particulate nitrate increased between two time stamps.’ Interpretive: ‘A change in transport or partitioning could help explain the rise.’ Attribution: ‘A specific emission source produced the measured rise.’ The last requires much stronger evidence. If no independent source information exists, stop at the first or cautiously second statement. Good scientific translation often consists of resisting the final step, not adding dramatic language.
Quality checks before using the data in a report
Begin with instrument time stamps, data availability flags, calibration and service records, sample flow, baseline stability, detector response, component reconstruction, collection-efficiency method and any periods of inlet contamination. Confirm the plotted interval reflects valid data rather than interpolated gaps. Flag sustained zero-like signals when the instrument was not operating; these are missing observations, not necessarily clean air. When statistics are aggregated, explicitly identify what fraction of possible intervals contributed and whether invalid periods were excluded from the denominator.
Then look for cross-checks: do measured components rise or fall coherently with an independent total mass estimate? Does a calibration challenge make sense? Are negative baseline-corrected values handled by a documented policy? Is the same CE assumption applied consistently to comparable datasets, with exceptions justified? Are seasonal changes confounded by hardware changes? A reader does not need access to the instrument firmware to ask these questions. Each check helps convert a graph from a decorative picture into evidence whose reliability can be tested.
Translating the most misleading phrases into defensible English
‘The AMS measured all fine particles’ should become ‘The AMS estimated a specified non-refractory aerosol fraction within its transmission and detection range.’ ‘The organic aerosol was 60% of air pollution’ should become ‘Organic mass accounted for 60% of the summed AMS components during the stated averaging period.’ ‘The nitrate vanished’ should become ‘The measured particulate nitrate concentration declined, with partitioning and sampling conditions requiring evaluation.’ These rewrites do not weaken science; they protect the boundaries between what was measured and what was inferred.
‘Instrument A proves instrument B is inaccurate’ should become ‘The instruments differ by a measured amount, and their definitions, size responses, calibrations and uncertainty must be checked.’ ‘A sharp sulfate increase proves a named source’ should become ‘A sulfate increase was observed; source attribution requires independent evidence.’ The precision of terms is part of interpretation, which is why the eduKate vocabulary precision hub provides a useful general-language parallel. A word can be nearly right yet change the scientific claim. The goal is accuracy that a non-specialist can follow without being misled.
Independent transfer problem: chemical percentages
Try this before reading the solution. An illustrative one-hour mean has organics 6.0, sulfate 2.5, nitrate 1.0, ammonium 1.2 and chloride 0.3 µg/m³. Calculate the summed AMS non-refractory concentration and the percentage contributed by organics. Then explain why the percentage must not be labelled the organic share of PM2.5. The arithmetic is straightforward; the conceptual test is selecting the correct denominator and naming the measurement’s physical scope.
Answer: the sum is 11.0 µg/m³ and the organic contribution is 6/11 × 100 ≈ 54.5%. The valid claim is that organics made up about 54.5% of this AMS non-refractory component sum over the hour. PM2.5 covers another size fraction and may contain refractory and other components not represented in the sum. Even an independently measured PM2.5 value should be time-matched before calculating an organic share relative to it. This is a simple example of why the denominator is often more important than a calculator.
Independent transfer problem: charge and neutralisation
An invented AMS record gives sulfate 9.6, nitrate 6.2 and ammonium 5.4 µg/m³. Convert the measured ion masses into microequivalents per cubic metre and assess a restricted ionic balance. Use approximate molar masses sulfate 96.06, nitrate 62.00 and ammonium 18.04 grams per mole. Do not assume that any apparent balance reveals an exact particle salt or excludes all other aerosol ions. The task is to make the charge arithmetic explicit.
Solution: sulfate demands 2 × 9.6/96.06 ≈ 0.200 µeq/m³ and nitrate demands 6.2/62 ≈ 0.100 µeq/m³; together they demand about 0.300. Ammonium supplies 5.4/18.04 ≈ 0.299 µeq/m³. The restricted balance is close, within the rounding of these invented inputs. The conclusion is limited to the chosen ions and estimated concentrations. Other acids, cations, gas partitioning, sampling bias and instrument uncertainties may still matter. Near-neutrality is not evidence for a unique molecular composition.
Independent transfer problem: alternative corrections
A calibration-scaled AMS organic signal is 3.6 µg/m³ before the collection-efficiency adjustment. A report used CE = 0.40; a review argues a value of 0.60 is also plausible. Calculate the two corrected numbers and the relative difference. Then identify what evidence could help choose the appropriate assumption and how a final public-facing report should represent the uncertainty if neither value can be established definitively.
Solution: at CE = 0.40, the result is 3.6/0.40 = 9.0 µg/m³. At CE = 0.60, it is 6.0 µg/m³. The former exceeds the latter by 3.0 µg/m³, equal to 50% of 6.0. Support could include chemical composition, particle phase and independent size-matched mass comparisons, coupled with published calibration methods and instrument conditions. Without a justified choice, present a sensitivity range or conditional interpretation rather than treating the higher number as a measurement independent of the CE model.
Independent transfer problem: averaging and missing data
An instrument records three valid 10-minute non-refractory concentrations of 8, 10 and 12 µg/m³ and then has three missing 10-minute intervals. What is the mean of the observed half-hour, and can it honestly be called a complete one-hour mean? If an automated dashboard writes zero for missing samples, how would the naive hour average differ? This question tests whether the reader can distinguish measurement availability from environmental concentration.
The mean of the three valid records is (8 + 10 + 12)/3 = 10 µg/m³ for the observed half-hour. There is insufficient information to determine the complete hour’s true average. Replacing missing values with zero would give 30/6 = 5 µg/m³, a numerical artifact. A careful summary records 50% temporal coverage and withholds or qualifies the hourly mean according to the relevant data completeness rule. Do not invent an hour of clean air merely because a logger stopped writing values.
Frequent questions about AMS interpretation
Is an AMS a replacement for a PM2.5 filter? No. Its chemical time resolution is valuable, but the mass definition, size transmission, handling of refractory components and operational calibration differ. Are non-refractory and volatile identical? Not necessarily; non-refractory here describes response under the AMS vaporization method, not a single atmospheric volatility class. Can its nitrate, sulfate and organic time series directly identify a source? They provide important clues, but source attribution needs other evidence and an explicit uncertainty assessment.
Does a high-resolution instrument automatically resolve every molecule? No. Better ion separation reduces some ambiguities, but fragmentation, isomers and unresolved chemical mixtures still impose limits. Should a negative background-corrected mass be simply deleted? Not automatically; follow the documented detection-limit and quality-assurance policy. If AMS and SMPS size distributions have different peak locations, which is wrong? Neither necessarily; compare vacuum aerodynamic and mobility diameters, effective density and transmission ranges. These questions are worth resolving before building a complicated attribution model.
From instrument numbers to reliable public explanation
A reliable summary can fit four sentences: define the sampled aerosol fraction; report the numerical observation with its units and averaging interval; state the key calibration or CE assumption; and distinguish a supported interpretation from an unproven source claim. Longer specialist analysis should then show how the result changes under plausible measurement uncertainty. This structure respects the reader while leaving room for advanced detail, rather than turning a simple graph into an unsupported sweeping claim.
This article’s numerical examples are constructed for teaching, not observations from a real monitoring site. The linked methodological sources concern physical measurement and calibration; they are not evidence for any invented data point. You can continue with the verified SMPS mobility-diameter interpretation, aerosol optical scattering interpretation, and filter-based organic and elemental carbon interpretation. Their complementary measurement boundaries make it easier to see why matching numbers do not always measure the same thing.
Original methodological sources and further reading
US EPA, Ambient Air Monitoring Quality Assurance, programme procedures and guidance
Google Search Central, Creating Helpful, Reliable, People-First Content, editorial guidance