Air Quality Monitoring - Stage 4

01 Conduct offline (filter-based) and real time chemical speciation

By Stage 4, the laboratory facilities are established, and it is important to advance the offline characterisation of filter samples, particularly by analysing primary organic markers associated with specific sources such as wood burning, traffic emissions, cooking etc (Srivastava et al., 2025). This can be achieved using various analytical techniques (e.g. HPLC, GC–MS, or LC–MS), which provide detailed chemical speciation. However, implementing this approach may require upgrading the existing analytical facilities by investing in specialised instruments and supporting equipment. If the resources are limited to upgrade the analytical facility in your jurisdiction, the analysis could be contracted to an external laboratory, either within the region or internationally. In such cases, this would typically involve a financial cost, although it may also be possible to establish collaborative arrangements such as data sharing partnerships or co-authorship on resulting publications to support the work.

Real time chemical speciation is important for advancing monitoring capabilities to better capture the chemistry of high pollution episodes or haze events. This could involve incorporating high time-resolution instruments (e.g. Aerosol Chemical Speciation Monitor (ACSM)), which provides detailed insights into PM composition. While high time-resolution instruments are expensive (typically exceeding USD 300,000), they offer significant value in understanding PM composition and identifying emission sources (for more information, see Source Attribution Guidance Stage 4, Step 6). For example, high time-resolution instrument provides real-time measurements of the major non-refractory components of fine PM, including organic aerosol, nitrate, sulphate, ammonium, and chloride. This information can play a key role in designing effective mitigation strategies. Therefore, despite the cost, it would be good to consider deploying such instruments at least at major monitoring stations (ACTRIS, 2021) where they can deliver the greatest benefit. 

02 Add new pollutants to the monitoring network

To enhance and expand long-term air quality monitoring, it may be beneficial to introduce additional instruments such as monitors for methane (CH4) and ammonia (NH3) alongside key pollutants like PM2.5/PM10, NOx, CO2, SO2, and ozone.  

Monitoring methane is highly important because CH4 is a potent greenhouse gas and a significant contributor to global warming. It can also serve as a tracer for emissions from agriculture, waste management, the energy sector, and wetland environments. Long-term CH4 measurements can help identify emission sources, evaluate mitigation strategies, and support the Global Methane Initiative and Global Methane Pledge. Depending on resource availability in your jurisdiction, the most appropriate monitoring approach and instrumentation can be selected (Lin et al., 2023).

Ammonia is a major precursor of secondary inorganic particulate matter which contributes significantly to PM mass. Continuous NH3 measurements can improve understanding of agricultural emissions, nitrogen deposition, and secondary inorganic particle formation. However, NH3 monitoring remains technically challenging due to its highly reactive nature, tendency to adsorb onto sampling surfaces, and sensitivity to environmental conditions such as temperature and humidity (UK Environment Agency, 2024). See Twigg et al., 2022 below to gain information on the use of different NH3 instruments.

At this stage it is also necessary to start continuous monitoring of volatile organic compounds (VOCs), as this helps improve emission inventories for air quality models and enhances our understanding of ozone formation. Instruments such as BTEX (BTEX refers to a group of VOCs consisting of benzene, toluene, ethylbenzene, and xylene) monitors are particularly relevant here, especially considering that benzene is classified as a human carcinogen by the International Agency for Research on Cancer (IARC) and BTEX compounds are recognised as hazardous air pollutants by the U.S. Environmental Protection Agency.  

Adding VOCs, CH4 and NH3 would strengthen the long-term monitoring goals, providing valuable information for air quality management through direct information on emission sources, climate assessments, and evidence-based decision-making. 

03 Implement continuous black carbon and metals monitoring

Black carbon (BC) has significant impacts on both climate and human health (European Environmental Agency, 2013). The main sources of BC are the incomplete combustion of fossil fuels and biomass, as well as emissions from vehicle exhausts. Adding BC measurements into your monitoring network is strongly recommended, as they can provide valuable information on the contributions of traffic and biomass-burning emissions to PM mass (see Source Attribution Guidance Stage 4, Step 4).

BC is commonly measured using a range of optical and spectroscopic instruments that quantify the light-absorbing properties of carbonaceous particles (see ACTRIS, 2021; UK DEFRA). Many monitoring networks, including the UK BC monitoring programme, use such instruments to provide real-time measurements of BC concentrations. Measurements at multiple wavelengths can also help distinguish between different combustion sources, with shorter wavelengths being more sensitive to organic compounds associated with biomass burning, while longer wavelengths may provide a quantitative measure of BC.

Additionally, the monitoring of high-temporal-resolution data on elemental composition using x-ray fluorescence techniques can provide important insights into emission sources (Tremper et al., 2018), including industrial activities and dust, while also contributing to a better understanding of associated health impacts. 

04  Expand monitoring to support satellite remote sensing and air quality models

The concept of using satellite data was already introduced in Stage 3, Step 4, as satellite observations are useful for increasing spatial coverage beyond point-based surface measurements. However, significant discrepancies are still observed when comparing ground measurements with satellite products. This is particularly important given that many air quality models rely on satellite-derived indicators to assess ozone formation and constrain emissions. One of the most widely used indicators for ozone formation chemistry is the formaldehyde (HCHO) to NO2 ratio (e.g. derived from TROPOMI observations) (Acdan et al., 2023). HCHO is commonly used as a proxy for volatile organic compounds (VOCs), while NO2 is a proxy for NOx emissions. The HCHO/NO2 ratio therefore provides an indication of whether ozone production is primarily VOC-limited, NOₓ-limited, or within a transition regime. This information helps determine whether reducing VOC emissions, NOₓ emissions, or both is likely to be the most effective approach for lowering surface ozone concentrations, making this metric valuable for identifying the most effective emission-control strategies.

To address this gap, it is proposed to introduce continuous monitoring of HCHO (check given references for instruments), which would play a key role in validating remote sensing data and improving our understanding of surface ozone formation regimes (UK DEFRA, 2000). However, direct measurements of HCHO remain relatively uncommon within routine air quality networks, despite its importance. Continuous surface measurements would provide a robust dataset for the validation of satellite observations.

In addition, the deployment of a continuous lidar system at a major monitoring site would enable routine measurement of atmospheric boundary layer height (Barlow et al., 2011; RI URBANS, 2025). The boundary layer controls the volume into which pollutants are mixed and is therefore a critical parameter for understanding the dispersion of pollutants. Accurate information on how the boundary layer height evolves throughout the day can be used to constrain dispersion processes within air quality models. Incorporating observed boundary layer heights has been shown to improve the representation of pollutant transport, mixing, and accumulation, subsequently improving the accuracy of air quality models. 

05  Expand monitoring sites to capture impacts on crops and ecosystems 

Previous monitoring stages are mainly focused on understanding emission sources, characterizing exposure/health impacts, supporting decision making and designing mitigation strategies. Building on that knowledge, it is also important to shift the attention towards understanding how pollutants, especially ozone, may affect vegetation and crop yield (WMO, 2023a).  

The first step in this direction is to liaise with colleagues conducting Environmental Impact Assessments to decide where best to locate these additional monitors, focussing on areas predicted to have the highest impacts of ozone on yield or forest tree growth. If possible, collect hourly meteorological data from the same or nearby sites. This data can be used together with hourly ozone data to calculate the Phytotoxic Ozone Dose above a flux threshold of 'y' (PODy) as described in the Environmental Impact Assessment Guidance Stage 4, Step 3, and associated resources. This metric has been proven to be a more biologically relevant metric for plants than AOT40 as it takes into account how wide open the leaf pores are and therefore how much ozone gets into the plant at a given time, rather than simply accounting for the ozone concentration at canopy height.

In addition to ozone, PM can also influence ecosystems, both directly through deposition on leaf surfaces that may block stomatal pores and indirectly by reducing the amount of sunlight reaching the leaves (WMO, 2023b). Although response functions for these impacts are not yet fully developed, PM measurements can still provide useful indications of where such effects are likely to occur.

Additionally, it would be beneficial to start monitoring both dry and wet deposition of nitrogen and sulphur to assess their adverse impacts on humans and ecosystems (see Vet et al., 2014, and guidelines developed under EANET and LRTAP). It is also important here to liaise with colleagues conducting Environmental Impact Assessments to determine the best locations for monitoring, focussing on impacts on biodiversity and freshwater ecosystems, thereby maximising the impact of the monitoring programme (see Environmental Impact Assessment Guidance Stage 4, Step 2). Wet deposition is monitored through direct sampling of precipitation, whereas dry deposition is more challenging to measure due to complex surface interactions and is typically estimated using the inferential method, which combines air quality measurements with meteorological data and land-cover information. Follow the given references to design the monitoring set up for dry and wet deposition of nitrogen and sulphur. 

06 Enhance quality assurance/quality control (QA/QC) and validation of data

QA/QC has already been introduced earlier in Stage 3, Step 6; however, the deployment of research-grade monitoring instruments at this stage requires additional attention. These monitors require regular calibration, which should be conducted either by highly qualified trained staff or by a third party subcontracted by the manufacturer. All calibration and maintenance information should be carefully logged and made available during data processing.

One way to assess data quality is by reconstructing PM mass using all measured PM components (e.g. elemental and organic carbon, ions, metals) within your jurisdiction from both offline and online instruments. Comparing measurements of the same PM components obtained using different instruments or analytical methods can also provide a useful means of assessing data quality by evaluating the consistency and agreement between measurement approaches (see Xu et al., 2020; Windell et al., 2025).  

Offline analytical methods should be validated using certified reference materials such as NIST standards for metals and organics and compared with historical values to evaluate the robustness and accuracy of the analytical protocols (NIST, 1992). In addition, installing remote monitoring software for monitoring instruments can enable real-time checks on operational status outside of office hours, while maintaining detailed logbooks is essential for tracking instrument performance, maintenance activities, and data reliability. 

07 Enhance data treatment methodologies

By now, you are probably already aware of data treatment tools (e.g. R and Python) and are also capable of easily using them. It is also important now to expand the data treatment procedures for all newly introduced air quality monitors to ensure consistency and reliability in analysis. You need to make sure all data is prepared in accessible and well-documented formats, with clear explanatory notes detailing collection methods, processing steps, and any limitations. Adopting the FAIR principles (Findable, Accessible, Interoperable, and Reusable) will help maximise the usability of the data, alongside providing a designated contact person to address any queries or issues.  

It is also essential to make sure that data treatment methodologies are aligned with previous stages. For example, missing values or values below the detection limit should be treated similarly across all stages.

In addition, making data treatment scripts available on shared platforms such as GitHub or institutional repositories can promote transparency. Many instrument manufacturers provide their own software packages for processing raw data, which can be useful for initial data handling. For more routine tasks, such as averaging data to another timeline, generating diurnal profiles or visualisation, a wide range of community-developed scripts and packages are available online and can often be adapted for your project-specific needs. 

08 Share the data and conduct outreach 

Data should be made available through an open-access platform in a format that is easily accessible and usable for the general public, stakeholders and advisory groups involved in decision-making (see websites from CEDA, UK DEFRA, and EANET). To support policy development and mitigation strategies, data should be presented in interpretable formats, such as figures showing pollutant time series, daily average concentrations, and long-term trends, including annual variations where relevant.  

Outreach activities are equally important for maximising the impact of the monitoring programme. Findings should be shared with the general public, health professionals and educational organisations to raise awareness of air quality issues and support informed decision-making across sectors.

Regular data submission to open platforms is essential to maintain consistency, transparency, and usability. In addition, data sharing practices should align with established regional and international monitoring frameworks, while allowing unpublished data to be made available upon request for research purposes, thereby promoting collaboration and further scientific analysis. 

09  Plan for an integrated field campaign 

At this stage, you are ready to take more advanced steps by focusing on the implementation of special monitoring projects, including the planning of integrated field campaigns to investigate specific emission sources (e.g. crop burning) or episodic pollution events (e.g. winter haze). These campaigns may run for a few weeks or extend over an entire season, depending on the study objectives, the instruments deployed, and the availability of nearby ground stations for collecting supporting meteorological data. Successful implementation will require careful consideration of instrument readiness, including their condition and calibration status, as well as adequate staffing. It is advisable to assign two to three responsible personnel to oversee operations, ensure continuity, and manage any issues that arise.