Air Quality Monitoring - Stage 2

01 Establish a network

Expanding from a single continuous air quality monitor to a multi-site network requires careful planning and execution. It involves several stages and for each new monitor, consideration should be given to choice of technology, calibration, operation, maintenance, and data management and these are discussed in subsequent steps below. The first step for network expansion involves establishing a network plan with consideration of site selection that will help to achieve your monitoring objectives (e.g. whether to move beyond baseline assessment and conducting compliance monitoring in specific areas).  Location selection should consider site representativeness, staff access, power and communication infrastructure, equipment and staff safety, and finally network architecture that includes a map of your network, with instrumentation identified, communication methods listed and data storage and access methods to enable scaling up as your network expands. High-traffic areas and industrial zones (where pollutant concentrations are likely to peak), and community locations (representative of public exposure to air pollution) represent prime sites for consideration.  The following references provide examples of network strategies that may be useful as you plan your growing network. While References 8 and 9 below touch on low-cost sensors (See Step 4), they also have general guidance for air monitoring network strategies. Stage 1 guidance provided resources for the actual deployment of additional PM2.5 continuous monitors, so that can be referenced for those additional monitors. 

02 Install continuous reference-grade gaseous pollutant monitors at select sites

The integration of monitors for gaseous air pollutants (e.g. ozone, nitrogen dioxide, sulfur dioxide, carbon monoxide, etc.), also called “gas analyzers”, should focus on optimal placement within the existing monitoring framework. It’s advisable to collocate gas analyzers with established fine particle monitors, as this facilitates the comparative analysis and representative sampling of particulate pollution and gaseous pollutants. A modular approach using racks that accommodate multiple sensors will help streamline the setup, ensuring that all monitors have access to necessary power and data transmission infrastructure. These monitoring systems should enable real-time tracking of gas concentration levels, making it possible to identify pollution spikes and relate them to meteorological conditions such as temperature and wind patterns, as well as aid in source attribution efforts.  

Implementing gaseous pollutant monitoring will necessitate increased human resources for effective management. Gas analyzers require more frequent attention (e.g. periodic calibration checks) compared to particulate monitors, translating into an extra technician dedicating about five to ten hours per month for each location. Ensuring zero air and calibration gas cylinders are available at the monitoring site (and are secured safely) will also require additional logistical coordination. Data collection and verification could require additional staffing, potentially necessitating the equivalent of one full-time employee for every five monitoring sites as well as training these staff on the operation of new systems and how to interpret new data.

Modern urban monitoring networks usually incorporate gas monitors for nitrogen oxides (NO/NO2), ozone (O3), sulfur dioxide (SO2), and carbon monoxide (CO) with each PM monitor, as part of a comprehensive Air Quality Monitoring Station (AQMS). However, given the added cost, logistical challenges, and effort required for gas monitoring, it is prudent to at least deploy between one to three gas analyzer racks for every five to ten monitoring sites. Further, not all gases need to be monitored at every location. A balanced approach might involve placing monitors for nitrogen oxides (NO2 and NO) near high-traffic urban areas (commonly called “curbside” or “roadside” locations) and near industrial areas. Ozone (O3) monitors could be placed at “urban background” locations, where concentrations are usually higher, especially during warmer months. Carbon monoxide (CO) monitors can be installed at roadside and near waste burning sites, and volatile organic compounds (VOCs) should be addressed based on industrial activity and urban density.  Initial setup costs can vary significantly, with gas analyzers ranging from $5,000 to $100,000 depending on species and their associated measurement technology. Ongoing operational costs, which encompass calibration, maintenance, and potential replacement parts, can average between $1,500 to $5,000 annually per site. Additionally, training costs for personnel will contribute to the overall financial requirements.

However, the benefits of enhanced gas monitoring capabilities often outweigh the associated costs if your jurisdiction has an engaged public health community that will use the data to improved public health outcomes based on a comprehensive assessment of gaseous pollutant exposure, which allows for targeted interventions and regulatory compliance. The availability of data can drive science and then informed policy decisions, fostering community engagement by making air quality more transparent and actionable.

Finally, selecting the particular gases to measure is critical in assessing ambient air quality-related health impacts. Nitrogen dioxide (NO2), ozone (O3), carbon monoxide (CO), sulfur dioxide (SO2), and volatile organic compounds (VOCs) are particularly vital. Each of these pollutants poses specific health risks (for example, see the US EPA reference below) and is significant for urban air quality, making their inclusion in your monitoring strategy essential for effectively understanding and mitigating public health risks. 

03 Establish collaboration agreements with hydrometeorological services (weather)

Establishing a collaborative relationship between air quality protection agencies and a jurisdiction’s hydrometeorological services department is essential for optimizing data sharing and resource management. Start by scheduling a series of joint meetings to discuss mutual goals, such as improved air quality monitoring and data integration.  

Identify common areas of interest and opportunities for collaboration, such as shared sensor networks or co-located monitoring stations. Develop a data-sharing agreement outlining the types, frequency, and protocols for data exchange, ensuring both parties prioritize transparency and accessibility.

To reduce the human resources burden, consider cross-training staff from both agencies to operate and maintain equipment, facilitating knowledge transfer and creating a more versatile workforce. Implement joint maintenance schedules for shared equipment, which can improve efficiency and reduce operational costs. Regular communication and feedback loops will further strengthen this partnership, fostering ongoing collaboration and innovation in monitoring efforts. 

04 Integrate low-cost sensors

Regulatory or reference-grade monitors can be expensive and resource-intensive to maintain. Hence, integrating lower-cost sensor-based ambient air pollution monitors (for both PM and gaseous species) into a monitoring network is an opportunity to enhance air quality assessment while addressing both spatial and temporal coverage. However, this integration requires a balanced approach, emphasizing careful calibration and co-location with established regulatory reference-grade monitoring stations.  

To effectively incorporate low-cost sensors, the first critical step is robust calibration. Low-cost sensors often exhibit variability in accuracy and reliability compared to reference-grade devices. Therefore, before deployment, it is essential to calibrate these sensors systematically against certified reference monitors. This usually involves co-locating low-cost sensors with reference-grade monitors in the target setting (e.g. urban background and/or roadside locations), allowing for direct comparisons of their data over a defined period (typically a few weeks), to develop correction algorithms or factors to apply to the readings of the low-cost devices, thereby improving their data quality to “near-reference” levels. Once calibrated, the low-cost sensors can be deployed at various locations, particularly in urban areas where traditional monitoring networks may be sparse, but also to assess baseline conditions in peri-urban and rural locations.  

The integration of low-cost sensors offers several significant advantages. First, when deployed as part of a spatially dense monitoring network, these sensors can effectively identify air quality hotspots—areas with high pollution levels that are not captured by existing sparse monitoring networks. By pinpointing such areas, regulatory agencies and local governments can tailor interventions to mitigate pollution exposure, improving public health outcomes.

Second, low-cost sensors can serve as vital tools for assessing population exposure. By deploying networks of these sensors in residential and urban areas, researchers can obtain real-time data reflecting community exposure to harmful pollutants. Such information can be instrumental in determining future placements for regulatory-grade monitoring sites, ensuring they align with areas of highest concern.

Public awareness is another crucial benefit of low-cost sensors. Their relatively low cost and ease of deployment allow communities, civic organizations, and schools to participate in air quality monitoring. Engaging communities through citizen science initiatives fosters transparency and education regarding air quality issues, empowering residents to advocate for improved air quality measures based on localized data.

Despite their numerous advantages, low-cost sensors also have inherent limitations. One notable challenge is their reduced accuracy and reliability compared to reference-grade monitors, especially in variable environmental conditions. For example, unlike reference-grade gas monitors that are operated at a steady temperature with controlled shelters, sensor-based systems are usually exposed directly to the ambient fluctuations in temperature, which can introduce errors relative to a reference system. PM sensors are also susceptible to humidity changes (while reference monitors control the moisture content of incoming air), leading to data variability. Users must remain cautious and aware that while low-cost sensors can provide valuable insights, they should not be used as stand-alone sources for regulatory compliance or long-term trend analysis. The WMO guidance below provides guidance on how to correct for these and other effects.

Additionally, the expected lifespan of low-cost sensors tends to be shorter than that of traditional monitoring equipment. Many low-cost sensors may only function effectively for one or two years, depending on the environmental conditions and usage. Given their anticipated degradation over time, a systematic monitoring schedule must be established to assess sensor health and performance and to have a clear decommissioning plan in place to remove faulty and obsolete sensors.

There are now many lower-cost sensor-based monitoring solutions in the market. Efforts like AQ-Spec, Afri-SET, and the Airparif Microsensors Challenge conduct independent, standardized testing across the world and these results can provide an initial assessment of sensor data quality that an air quality network manager can review before deciding on a potential sensor-based solution(s) to implement.  

Air quality managers who are seeking integrated planning capacity can also explore the WMO IG3IS website referenced below to learn about low-cost GHG monitoring network potential.  

05 Use passive diffusion sampling

In contrast to reference-grade gas analyzers, passive diffusion tubes are simple, cost-effective devices that offer a practical means to assess the levels of various gaseous pollutants, such as nitrogen dioxide (NO2), ozone (O3), volatile organic compounds (VOC) and sulfur dioxide (SO2), without the need for electrical power or a temperature-controlled shelter. Their operation relies on the principle of diffusion; as the target gases enter the tube, they react with a sorbent material, allowing for quantification after a defined exposure period.

Among the significant advantages of passive diffusion tubes are their cost-effectiveness and their general ease of deployment. They can be placed in various locations, including residential areas, industrial zones, and near traffic corridors, to assess air quality without the need for complex infrastructure. This makes them particularly valuable in underserved communities or in regions where environmental monitoring resources are limited. Their deployment is straightforward, as they can be installed by community members or volunteers, promoting citizen engagement in air quality assessment.

In addition to their utility for general monitoring, passive diffusion tubes are especially effective for hotspot identification. By strategically placing these devices in areas suspected of high pollution levels — such as busy roadways or industrial facilities — researchers can gain insights into localized air quality issues. The data collected from these hotspots can inform regulatory bodies and local authorities about pollution levels, enabling targeted intervention efforts to reduce exposure to harmful emissions.

However, users must also recognize the limitations associated with passive diffusion tubes. One primary concern is the requirement for access to an accredited analytical chemistry laboratory for the analysis of the collected samples, in addition to the added expense of cold storage of the samples and transport in cold conditions. After the exposure period, which typically lasts one month, the tubes need to be sent to a laboratory for precise analysis and quantification of the pollutants at additional cost. This dependency on laboratory facilities can lead to delays in obtaining results, which may hinder timely decision-making, as can the lack of real-time reporting, particularly in rapidly changing environmental conditions. Diffusion sampling, especially for NO2, may not perform well at low humidity, and variable humidity and temperature can also influence the uptake rate (rate at which NO2 is adsorbed). 

06  Conduct analysis on historical data and make data publicly available

Ambient air quality data is useful only to the extent that it is analyzed and used.  By Stage 2, your jurisdiction should have collected at least one year of data that is crucial for informing both the public and decision-makers about environmental conditions and health implications. Begin by making your validated data publicly available so that academic partners and civil society organizations can help with the data analysis burden.  Aggregating data from various sources like monitoring stations and low-cost sensors can help to identify the right data for the right purpose. Agencies need to ensure the highest data quality through validation techniques to maintain accuracy for compliance purposes, but other data may be appropriate for citizen science initiatives and to support hotspot identification. Aggregation platforms and algorithms exist for different types of analysis so you may want to get in touch with academic partners or third-party aggregators like OpenAQ to help with some analyses.

Common analysis methods include Descriptive Statistics, which summarize the key features of the dataset (e.g. mean, median, and standard deviation), helping to identify spatio-temporal trends in air quality. Time Series Analysis is useful for assessing seasonal patterns and trends, allowing for informed predictions about future air quality and correlations with meteorological parameters.

Another important technique is Spatial Analysis, which maps pollutant concentrations to highlight pollution hotspots. This method also helps visualize geographic disparities in air quality, aiding targeted interventions.  

Comparative Analysis can be employed to assess air quality against standards (e.g., National Ambient Air Quality Standards) or historical data, illustrating progress or regression.

These analyses should be accompanied by Visualization Techniques showing spatial patterns (e.g. heat maps) and temporal trends (diurnal, weekday vs weekend, monthly or seasonal) and correlations of air pollutant levels with meteorological data and other relevant information for additional insights. 

07 Prepare an operational budget to ensure continuity

Identifying key budget requirements for air quality monitoring network operations and maintenance is essential for long-term sustainability. Start by conducting a comprehensive inventory of all network components, including monitoring stations, sensors, data analysis tools, and support infrastructure. Classify expenses into fixed and variable costs, accounting for initial capital expenditures, ongoing operational expenses, maintenance, and personnel costs.

Next, project future needs by assessing the expected lifespan of equipment and estimating replacement cycles. Include costs for calibration, repair, and training for staff involved in data collection and analysis. Develop a detailed budget proposal (beyond what you prepared in Stage 1, Step 1) that outlines both current and anticipated future expenses, including contingency funds for unforeseen repairs or upgrades to highlight the importance of sustained funding. Present this information clearly to decision-makers to advocate for the necessary funding, ensuring the continuous operation and reliability of the air quality monitoring network.  Chapter 2.4 of the Clean Air Asia guidance linked below can provide context for thinking about budgets that enable sustainable operation of a network and some preliminary (but dated) cost estimates to get started.