The collective realization about our acceptance of polluted air has been one of the most spectacular realities unveiled by the new coronavirus pandemic (COVID-19). As personal vehicle miles fell dramatically, skies in the world’s largest cities turned a shade of blue that astonished many residents. While skies cleared, researchers discovered that exposure to poor air quality could increase susceptibility to the coronavirus.

As parts of the world return to a resemblance of pre-lockdown status, there is discussion about not letting this improvement in air quality slip quietly into the history books. With urbanites around the world expressing a strong and renewed interest in improving air quality, cities might find themselves in need of better air quality data on which to base new policy decisions.

Luckily, the past five years have seen tremendous leaps forward in air quality sensing for smart cities. When gathering air quality data, smart cities should consider the following recommendations:

1. Increase data resolution

To improve air quality, cities must first achieve greater data resolution. Historically, air quality for an entire city has been measured at one or two geographic points for regulation compliance. However, concentrations of pollutants can change significantly over just a few city blocks, so measurements must be collected from many strategically chosen locations.

The past few years have seen the development of air quality sensors that are smaller, more automated and much less expensive than the regulatory sensors of old. These new sensors can be mounted to existing infrastructure and constantly communicate levels of up to a dozen pollutants through the internet of things (IoT). True, they may not have the accuracy of their larger brethren, but today's IoT sensors are getting pretty close.

Cities have options beyond physically deployed sensors, though they come with their own drawbacks. Cities can contract mobile surveying, which will give a much larger profile across geographic space at the cost of a less complete profile over time. Typically, these surveys involve sensors mounted to cars, but there are examples of cities using bike-share bikes, municipal trucks and even drones for mobile measurement.

Alternatively, cities could purchase access to one of the air quality models being pioneered by environmental data companies. These companies take data from existing air quality sensors and fill in the gaps with fluid dynamics models that account for weather conditions, traffic patterns and historic conditions to predict air quality for any point. Given the small pool of air quality data this method draws from, there are concerns about its accuracy. However, it is an inexpensive option.

2. Understand air quality sensors' ROI

Once a city has settled on a measurement technology, they must face the quintessential question for any municipal project: how do we pay for this? This question is a bit more difficult to answer for air quality monitoring than, say, smart street lighting. For the latter, you can model the energy savings and calculate a clear timeline for ROI, which makes the decision to cut the check much easier.

Cities have struggled to draw such a clear picture for air quality monitoring. One way to envision ROI is to quantify the macroeconomic gains when the people in a city spend less time sick due to asthma, emphysema and other respiratory conditions. Air pollution has a quantifiable economic cost: roughly $900 billion or 5% of US GDP annually, according to a 2019 study in the Proceedings of the National Academy of Sciences.

There may be other ways for cities to realize returns on their air quality sensing expenditures. As previously mentioned, air quality modeling companies depend on physically sensed data as the foundation of their modeling projections. The more data points they can gather, the more accurate their models will be. Cities could use the market for licensing this kind of data to gain a financial return on their IoT air quality network.

Other opportunities exist around data licensing, such as real estate development companies that are interested in proving good air quality as part of their project proposals.

3. Find innovative data uses

Once an air quality sensor network is in place, cities can use the data for much more than determining whether the air is good or bad on a given day. For example, the city of Leeds in the U.K. is combining air quality sensing and geofencing to automatically switch the city’s fleet of hybrid vehicles to electric-only mode when they enter areas with poor air. This approach helps dissipate these bad air patches, and vehicles can resume normal hybrid operation once they are out of the bad patch.

The city of Coventry in the United Kingdom is using air quality sensors to display messages that encourage drivers and pedestrians entering areas with high air pollution levels, to consider alternative routes. It's not difficult to imagine versions of this system where, instead of merely suggesting alternative routes, the behavior of traffic infrastructure is automatically altered to limit the number of drivers in a city’s most polluted areas.

Although it might be difficult to define an air quality sensor network's place in the smart city, innovative communities are showing that place does exist. The coronavirus crisis has demonstrated the importance of improving air quality instead of dismissing it as something that is nice to have. As technologies that measure and act on air quality data become more accessible, cities need to exhibit innovative thinking about how they use these tools for the betterment of urban life.