Atmospheric pollution source apportionment testing is a science-based analytical process that identifies and quantifies the contribution of emission sources—such as coal combustion, vehicle exhaust, industrial processes, and fugitive dust—to ambient air pollution. By combining rigorous sampling, comprehensive chemical speciation in accredited third-party laboratories, and validated receptor modeling techniques, source apportionment delivers actionable data for air quality management and emission control strategies.
Technical Overview
Atmospheric pollution source apportionment testing is a systematic analytical discipline designed to determine the origin and relative contribution of various pollution sources to ambient particulate matter and gaseous pollutants. The methodology rests on the receptor modeling principle: the chemical composition measured at a receptor site (an ambient monitoring location) is a mass-weighted mixture of the emissions from all contributing sources. By characterizing both ambient samples and representative source profiles, quantitative source contributions can be resolved statistically.
A complete source apportionment program typically covers primary emission categories, including coal combustion, motor vehicle exhaust, industrial processing, secondary sulfate and nitrate formation, biomass burning, cooking, and resuspended dust from roads, construction sites, and bare soil. The resulting contribution estimates support science-based policy development, heavy pollution episode response, and regulatory compliance verification, and are aligned with ambient air quality benchmarks such as GB 3095 and corresponding international measurement standards.
Testing Samples
Ambient Receptor Samples
Ambient samples are collected at representative monitoring sites within the study area, typically urban, suburban, industrial, transport-influenced, and regional background locations. Common receptor sample types include:
- Fine particulate matter (PM2.5) collected on quartz-fiber and polytetrafluoroethylene (PTFE) filters using medium- or high-volume samplers
- Inhalable particulate matter (PM10) collected in parallel to enable coarse–fine fraction comparison
- Total suspended particulates (TSP) for studies of deposition and large-particle transport
- Volatile organic compounds (VOCs) captured in pre-evacuated canisters or on multisorbent tubes for ozone and secondary organic aerosol precursor analysis
- Dust fall (deposition) samples for evaluating the impact of resuspended and wind-blown dust
Source Profile Samples
Representative samples of the major emission sources in the study region are collected to establish local source profiles. These typically include:
- Coal combustion emissions from power plants, industrial boilers, and residential stoves
- Motor vehicle exhaust characterized through tunnel, roadside, or chassis dynamometer sampling
- Industrial process emissions from steelmaking, coking, cement, and non-ferrous metal facilities
- Fugitive dust from paved roads, unpaved roads, construction sites, and exposed soil
- Biomass open burning and residential wood or straw combustion
- Cooking emissions from commercial and residential kitchens
- Ship and port-related emissions in coastal or riverine study areas
Testing Items
Source apportionment relies on a broad chemical speciation suite, because each source category leaves a distinctive compositional signature in the atmosphere. Core testing items include:
- Water-soluble inorganic ions: fluoride, chloride, nitrite, nitrate, sulfate, sodium, ammonium, potassium, magnesium, and calcium, which quantify secondary inorganic aerosol formation and identify combustion and dust markers
- Carbonaceous components: organic carbon (OC) and elemental carbon (EC), including thermally resolved carbon fractions that distinguish gasoline, diesel, and biomass burning signatures
- Elemental composition: more than twenty major and trace elements, including aluminum, silicon, titanium, vanadium, chromium, manganese, iron, nickel, copper, zinc, arsenic, selenium, cadmium, antimony, barium, and lead, which serve as tracers for dust, metallurgy, fuel oil, and brake wear
- Organic tracers: polycyclic aromatic hydrocarbons (PAHs), hopanes, steranes, levoglucosan, and cholesterol, which are diagnostic molecular markers for vehicle exhaust, biomass burning, and cooking
- VOC speciation: alkanes, alkenes, aromatic hydrocarbons, and halogenated compounds following photochemical assessment monitoring species lists, supporting ozone control and reactivity-based analysis
- Optional isotopic analysis: stable carbon isotopes and lead isotope ratios for advanced source discrimination in complex urban environments
Testing Methods
Sample Collection
Particulate samples are typically collected as 24-hour integrated samples over a full calendar year to capture seasonal variation, with intensified sampling during heavy pollution episodes. Gravimetric mass determination is performed under controlled temperature and humidity conditions in accordance with EN 12341, EN 14907, GB/T 15432, and HJ 618. Stack and fugitive source sampling follows methods consistent with ISO 9096, ISO 23210, and ASTM D3685/D3685M, while VOC sampling follows canister and sorbent-based procedures aligned with HJ 759 and HJ 734.
Instrumental Analysis
Chemical speciation is performed using validated instrumental techniques in accredited third-party laboratories:
- Ion chromatography for water-soluble inorganic ions extracted from filter sections
- Inductively coupled plasma mass spectrometry (ICP-MS) and inductively coupled plasma optical emission spectrometry (ICP-OES) following acidic digestion for elemental analysis, with non-destructive X-ray fluorescence (XRF) as a complementary technique
- Thermal-optical transmittance or reflectance analysis for OC/EC determination in accordance with EN 16909
- Gas chromatography–mass spectrometry (GC-MS) for PAHs, hopanes, steranes, and levoglucosan in accordance with HJ 646, and for VOC species in ambient and source samples
Receptor Modeling
Quantitative source contributions are resolved using established receptor models, selected according to data availability and study objectives:
- Chemical Mass Balance (CMB): a least-squares fitting model that matches ambient species concentrations with locally measured source profiles, well suited for episode analysis and secondary sulfate and nitrate estimation
- Positive Matrix Factorization (PMF): a multivariate factorization model that decomposes the ambient data matrix into factor profiles and contributions without requiring pre-measured profiles, constrained by realistic uncertainty weighting
- UNMIX: a geometric multivariate model used as an independent cross-check of resolved factors
Model robustness is verified through bootstrap resampling, displacement analysis, factor mapping against source profiles, and comparison of model-predicted mass with gravimetric mass. Secondary organic carbon is estimated using the minimum OC/EC ratio approach or elemental tracer methods. Deliverables typically include annual, seasonal, and episode-based contribution estimates with quantified uncertainty.
Applications
Source apportionment testing provides decision-support evidence for a wide range of air quality management scenarios:
- Urban and regional PM2.5 management planning, prioritizing control measures against the largest contributing sources
- Heavy pollution episode analysis and emergency response, identifying the drivers of rapid particulate accumulation
- Ozone control strategy development through VOC source apportionment and photochemical reactivity ranking
- Industrial park and enterprise-level emission accountability, quantifying the impact of specific processes on local air quality
- Post-implementation evaluation of clean air action plans, coal-to-gas conversion, and vehicle emission control programs
- Scientific research, emission inventory validation, and regional joint prevention and control studies
Quality Assurance and Standards Compliance
Credible source apportionment depends on rigorous quality assurance throughout the workflow. Field blanks, duplicate samples, and filter conditioning controls quantify sampling artifacts; certified reference materials, calibration verification standards, and inter-laboratory comparisons validate instrumental accuracy; and species-specific detection limits are incorporated into model uncertainty inputs. Sampling and analytical procedures follow recognized frameworks from ISO, EN, ASTM, and GB standard systems, and studies are commonly executed by accredited third-party laboratories under documented quality management systems consistent with ISO/IEC 17025.
FAQ
What is the difference between source apportionment and an emission inventory?
An emission inventory estimates what is released into the atmosphere from documented activities, whereas source apportionment measures what is actually observed in the air and mathematically attributes it to sources. The two approaches are complementary: receptor-based apportionment validates inventory assumptions and captures secondary formation, resuspension, and long-range transport that inventories may underestimate.
How long does a source apportionment project take?
A comprehensive annual study typically requires twelve months of continuous sampling to capture seasonal source variation, followed by several months of laboratory analysis and modeling. Shorter episode-focused studies can be completed within weeks when rapid diagnostic answers are needed during a pollution event.
Which receptor model should be used, PMF or CMB?
PMF is preferred when local source profiles are incomplete, because it resolves factors directly from the ambient data matrix. CMB is preferred when high-quality local source profiles exist and precise episode-level attribution is required, for example distinguishing coal burning from vehicle exhaust during a winter haze event. Best practice applies both models and reports results where they agree.
Why are both quartz and PTFE filters used for particulate sampling?
Quartz-fiber filters are suitable for carbonaceous (OC/EC) and organic tracer analysis, while PTFE filters provide the lowest blanks and best gravimetric performance for mass and elemental analysis. Parallel sampling on both substrates ensures that every chemical species is measured on an appropriate medium.
What sample frequency is required for valid results?
Typical programs collect 24-hour integrated samples at least every third to sixth day throughout a full year, supplemented by daily or continuous sampling during identified heavy pollution episodes, ensuring sufficient data points for statistically robust factor resolution and seasonal representativeness.