Near reference air quality monitoring is a category that has recently emerged from the confluence of improved sensor technology and new measurement paradigms. In a world transitioning towards next generation air quality monitoring (NGAM) it offers a balance between data quality and ease of use. Near reference monitoring can complement traditional regulatory reference networks by providing higher spatial and temporal resolution data in a less expensive way. The data quality is also high enough to be used to determine exceedances and issue pollution alerts. Unlike low cost sensor networks, near-reference instruments are calibrated with reference gases and methods which anchor data quality to international standards.
Reference – very good but expensive
The traditional approach to air quality monitoring is to establish a network of fixed regulatory stations equipped with certified instruments and operated according to a rigorous set of regulatory protocols. Such a network is expensive to install and operate but well suited for long-term air quality trend analysis. However, the strict quality assurance protocols required when operating reference monitoring stations are often not achieved by new or first time air quality networks. This can be due to a number of challenges such as, insufficient QA and/or technical support resource, site selection unknowns, and lack of access to reliable gas standards. In such cases the data quality is more aptly described as near-reference but with the downside rigidity of a fixed, large footprint network.
Near reference – good enough for most applications
Near reference air quality monitoring is a more flexible approach, still underpinned by excellent instrument performance but taking advantage of compact, lower power measurement technologies to make deployment and operation easier and faster. Near-reference grade instruments match the measurement performance of reference instruments in all but the most demanding situations. Their small footprint enables them to be deployed at sites that require little infrastructure – important for first-time or short-term monitoring. Near-reference instruments, like reference analyzers, are calibrated using dilution calibrators and certified gas mixtures enabling traceability to international reference standards. Data quality is usually determined by the user’s operating protocol not limited by instrument performance. Near-reference equipment is well suited to agencies that are starting to invest in air quality networks because in most cases site selection is a key unknown and the portability of near-reference equipment overcomes siting errors.
Indicative – low cost but limited
The term “indicative” is often applied to all air quality monitoring by non-reference equipment but this is not quite correct. Indicative is a category defined by the EU Directive 2008/50/EC on ambient air quality monitoring. It applies to a measurement which meets a lower data quality objective than reference or near-reference. Indicative data can be used to provide information on the spatial distribution of air pollutants but not for determining exceedances of air quality standards or issuing alerts. At the present time there are no performance test standards for indicative monitoring except for the MCERTS standard for particulate monitoring. A methodology for determining whether a sensor can achieve the EU standard for indicative monitoring has been developed but this required a sophisticated test rig and has not been ratified.
Citizen science – cheap as chips
Over the past few years there has been a rapid increase in the availability of low cost sensors and electronics which has enabled citizens to build and install devices to collect outdoor environmental data. These activities, while excellent at generating interest in the environment and building technology skills, often produce data that is low quality, difficult to interpret and unreliable. Often such devices are not calibrated (or even calibratable). There are excellent resources available that address some of the measurement challenges (calibration, selectivity, sensitivity, interferences, data interpretation) but such devices should be considered educational rather than analytical.
A comparison of the data quality for the different types of monitoring is given in the graph below with uncertainty expressed as a percentage of a limit value  . Uncertainty is a measure of the doubt associated with the ambient measurement. It follows that as the uncertainty decreases there is more confidence in the measurement.
Comparison of data quality expressed as a percentage of uncertainty for different types of monitoring
Near reference data ranks in quality between EPA/EU reference and EPA supplemental monitoring. In some cases where site conditions are benign with moderate climate conditions and low cross-interferent concentrations the data quality will match or even exceed regulatory requirements. At other sites, with challenging multiple emission sources and large meteorological variations, data quality may degrade. But this is also the case for reference monitors.
Ambient monitoring is categorised by both the equipment and protocol of operation because both impact data quality as shown in the table below. A poorly operated certified instrument will not achieve good data quality and cannot be considered “reference”. In fact, one might argue that the operating protocol plays a larger role in determining the quality of the data than the instrument type. Furthermore, many NGAM applications such as near-source or community monitoring call for monitor siting in places that would not meet siting standards for regulatory monitors.
A regulatory quality assurance protocol aims to maintain data quality at a level that meets the data quality objectives of a reference network, e.g. operating with a combined relative uncertainty of 15% or less for the European Union. Such protocols are defined in standards or regulations  and specify methods and maximum intervals for calibration and maintenance as well as traceability requirements for reference gases. These are onerous procedures and can be costly. In most cases automated calibration equipment is used to reduce operating costs but this drives up capital expenditure.
|Instrument type||Regulatory QA protocol e.g. USEPA||Traceable calibration||No calibration|
Impact on data quality for ambient air monitoring categorised by equipment and operating protocol
Near reference monitoring can be undertaken by instruments that can be calibrated against reference standards to ensure traceability of data. Certified and near reference analyzers can participate in this type of monitoring. The impact of not following a regulatory QA protocol is a slight increase in measurement uncertainty due to fewer calibrations and servicing but the benefit is reduced operating costs. In many cases, near-reference monitoring with traceable calibrations is perfectly adequate to produce actionable information and the small improvement in data quality via a reference approach does not warrant the significantly higher operating costs. Operating reference or near-reference equipment without field calibration for short term studies (few months) still results in good data quality because rates of instrument drift are low with such instruments.
This can be seen in the graph below of one month hourly averaged ozone data from an Aeroqual AQM 65 co-located at a US NCORE regulatory monitoring site. The AQM 65 was not field calibrated during this period and the correlation coefficient was 0.986 for the month.
Ozone hourly average data for an Aeroqual AQM 65 co-located with a US NCORE regulatory station
Sensor instruments based on fan or diffusion gas sampling, which cannot be calibrated with standard calibration equipment, are not able to meet near reference data quality but might meet an indicative standard if their factory calibration is transferable to the field. In reality, however, sensor drift would soon affect data quality. Then the inability to re-calibrate makes it challenging to re-establish measurement accuracy. Co-locating at a reference site to calibrate is possible but labour-intensive. Automated data validation/correction for large scale low cost sensor networks is still an area of research although recent work is showing progress.
Near reference monitoring is a new category of air quality measurement that is characterized by:
- Instruments which have low rates of measurement drift and can be calibrated using standard reference gases for traceability;
- Operating protocols which are less onerous than regulatory QA processes;
- Data quality (uncertainty) close to EPA reference analysers but better than supplemental monitoring and significantly better than indicative monitors;
- Compact, lower power and re-locatable instruments for wider range of applications; and
- Lower capital and operating costs compared with reference monitoring.
How can Aeroqual help?
Compact near reference instruments like the Aeroqual AQM 65 provide a flexible ambient air monitoring platform. They can be used in a wider range of monitoring applications than certified instruments while maintaining high data quality with field calibration against reference standards (see image below). This makes them ideal for building ambient air monitoring capability, filling gaps in existing networks, industrial fenceline monitoring, and rapid response applications where the capability to easily relocate instruments is required.
Field calibration of an Aeroqual AQM 65 near-reference station with traceable calibrator and certified gas standards
Combined with the AQM 65 near-reference station Aeroqual offers easy-to-use and feature-rich air monitoring software. Aeroqual Connect and Cloud software platforms allow you to access your data from anywhere, view pollution roses, export data and manage your air monitoring network.
If you need high quality air data with greater flexibility and lower costs compared with reference monitoring then near-reference air monitoring could be a great solution. Check out Aeroqual’s range of near reference instruments:
Get in touch with us for all your air monitoring needs!
Authors: Geoff Henshaw and Paul Pickering. For any feedback or questions contact email@example.com.