Sensing a change
When it comes to air quality monitoring, does the new wave of low-cost sensors spell a method change? Rick Gould reports
A decade ago, continuous monitoring for ambient pollutants such as nitrogen dioxide and particulate matter could only be carried out using certified, complex, large, demanding and costly monitoring systems. Many local authorities could only afford a few of these at most, leaving gaps in monitoring data across large areas.
Such systems are still essential, as they provide reference measurements based on validated European (EN) standards. They are also required to comply with the EU Air Quality Directive’s monitoring requirements. However, during the past 10 years, there has been an explosion in the availability of relatively inexpensive, compact systems that appear to do the same job – at the fraction of the cost and in a package that is both easy to deploy and operate. These monitoring devices are known as low-cost sensor systems.
A low-cost revolution?
“A revolution is taking place in air quality monitoring with the introduction of very large spatially dense networks of low-cost sensor systems, which can detect key pollutants with a high time-resolution,” explains Dr Nick Martin of the National Physical Laboratory (NPL). For example, in the summer, more than 100 low-cost sensors resembling tiny cooker extraction hoods started popping up across London as part of the Breathe London project. This provides continuous data on London’s air quality, supplementing traditional reference systems. It has parallels worldwide due to the growth in low-cost sensors and the interest that air quality managers and researchers are taking in them.
Low-cost sensors employ similar techniques to those used in reference systems, but in simplified and compact packages. Their prodigious growth is often matched by the manufacturers’ claims for them, with some producers asserting that they are as good as traditional reference monitors. Does the evidence support this?
“So far, they have performed less well than claimed,” says Martin, who works internationally in NPL’s team for characterising and testing the performance of environmental sensors and other types of monitoring system. Indeed, research worldwide shows that the performance of low-cost sensors is variable; some perform almost as well as reference systems under certain conditions, while other types have been found to be inaccurate, imprecise, inconsistent and unreliable.
"Research shows that performance is variable"
Barriers to success
The Breathe London sensors have performed comparatively well in reported, peer-reviewed research papers when compared with reference systems. But how good do they need to be?
The EU Air Quality Directive specifies two grades of monitoring system: quantitative reference systems and semi-quantitative indicative systems. Low-cost sensors could fill the latter role and Martin is part of the European standards group that is developing technical specifications for such sensors.
“Low-cost systems are presently poorly characterised, with no standardised means of verifying that the data meets the requirements for regulatory reporting or other applications,” he says. “This is a barrier to a market expansion that would benefit EU industry, and it inhibits the adoption of newer technologies. End users and regulators have no way of judging whether their purchases will fulfil their claimed performance, making it difficult to assess the effectiveness of new environmental policy implementation,” he adds. “Standardisation is the only way of overcoming these problems.”
Until then, the Government’s Air Quality Expert Group has provided advice on using low-cost sensor systems at bit.ly/2C2hSFS.
Rick Gould, MIEMA CEnv is a technical advisor at the Environment Agency. He is writing in a personal capacity.