What technical aspects to consider when choosing an air quality sensing device?
Description
Thanks to the fast changes in air sensor technology and the emergence of many environmental monitoring projects and initiatives, using sensing devices for environmental monitoring has never been more affordable. Nowadays, there is a very wide range of sensing devices to choose from: from off-the-shelf commercial devices to do-it-yourself (DIY) designs. Among these options we will very likely find a device for each need. This variety also means that choosing the right device is not so simple, especially when accounting for all the variables at play, both from the technical and non-technical point of view. The aim of this section is to provide some guidance, from the technical point of view, on how to choose an air quality sensing device, what factors to consider, and ultimately provide a simple checklist for us to review before embarking on the monitoring project at hand. This section goes hand in hand with the next one on What non-technical aspects to consider when choosing sensing devices?.
Why is this relevant?
Sensing devices are normally part of a larger system, which very likely involve connectivity to a data platform, and maybe other tools and services such as dashboards for visualization. These devices shall be considered as a part integrated in a whole system, from data gathering to data storage and data visualization. Usually, when we acquire a sensing device, that choice will immediately bind us to a certain data platform solution for transmission and storage of the data and will also impact how we can visualize and access our collected data. Thus, it is important to consider all the elements in the system before selecting the best monitoring option.
How can this be done?
A good first step is to understand the main components of the sensing devices that we have lined up in the chapter What are the main elements of a sensing device?
Once we have a clear understanding of what these parts are, we need to then look at how we would like our device to be deployed on the field. Does it need to be installed for long periods of time? Are there Wi-Fi communications available? Before doing a major investment, you can always order samples and try them in real life situations, covering different locations and types of users. Go through the whole process from installing the devices to your final dashboard or at least to your centralized database to see if you can spot any unforeseen issue.
The following checklist should cover most of the technical aspects you need to consider:
Pollutants measured: evaluate the sensing technology employed by the device, such as electrochemical, optical, or semiconductor-based sensors, to determine its suitability for detecting specific air pollutants of interest. Ensure that the range of air pollutants that the sensor device is capable of detecting, including common pollutants such as particulate matter (PM), nitrogen dioxide (NO2), ozone (O3), sulfur dioxide (SO2), carbon monoxide (CO), and volatile organic compounds (VOCs), are the ones that you need to monitor to answer your research question.
Geolocation (GPS): some manufacturers provide built in GPS chips. They may consume some precious energy but in some cases, they are an important asset. Typically, if your project involves moving objects (bus, cars, boat, UAV...), a GPS will come in handy to be able to geolocate the collected data. If your project involves wearables, most of the time it will come with an app that will retrieve location from the smartphone’s GPS. However, you will have to deal with the diversity of smartphones and specific authorizations.
Data quality: in some cases, a better sensor performance can be directly linked with an increase in cost, but that’s not always the case when dealing with air quality sensing devices. To keep a complex matter simple, better quality is related to two main aspects: sensor and interface electronics construction. Assess the accuracy and precision of the sensor device's measurements under real-world conditions, considering factors such as calibration, cross-sensitivity, and environmental variability. Check if the sensor device has been used by other initiatives or research institutions. See if there is information available besides the one provided by the manufacturer. A good rule of thumb is to aim for a tested, widespread technical solution, ideally already reviewed by independent bodies.
Response time and Detection limit: evaluate the response time of the sensor device, which refers to how quickly it detects changes in pollutant concentrations. Short response times are essential for capturing rapid fluctuations in air quality (critical if the sensor is going to be used as mobile). Determine the lowest concentration of pollutants that the sensor device can reliably detect, as this influences its ability to monitor air quality in areas with low pollutant levels.
Data transmission: evaluate how the sensor device transmits data, whether through wired connections, wireless communication protocols (e.g., Wi-Fi, Bluetooth, LoRa), or cellular networks, and assess compatibility with the existing data transmission infrastructure. Most devices will offer one or two data logging options. The most common one is remote data logging, via Wi-Fi communications, but some other options are available such as GSM/LTE or LoRa. In some cases, remote data logging may not be available, and only local storage is possible, in the form of a (micro) SD card. Wi-Fi is generally widely available in many cities, but keep in mind that certain areas might not be covered, or that the types of Wi-Fi networks available may not be supported for the devices (for instance, networks with captive portals such as eduroam (WPA2-Enterprise) are generally not supported). There might also be other challenges associated with Wi-Fi connectivity such as sensor distance or Wi-Fi terminals being turned off. Sensing devices using GSM/LTE or LoRa are becoming more common, but they require additional components, such as pre-paid phone/GSM plans, or an existing LoRa infrastructure available in the area. GSM/LTE is a good option as it is a quite pervasive solution that will get you covered in almost any location; however, consider the corresponding budget for the whole length of your program to pay for the data plans. LoRa and other radio networks are interesting especially if your territory already offers that type of infrastructure, however, some specific constraints must be considered such as the bandwidth fare use.
Internal data logging: in some cases, devices will offer internal data storage in the form of a (micro) SD card, or similar. If the device has remote data logging capabilities, this can be seen as a backup solution, but in some cases, it can be the only way to access the data. In general, this feature can be helpful if communication is lost, or it can be an option to store data with relatively low energy consumption, if you operate on solar panel. Make sure you understand how this data is stored as well, for instance, the file format or the capacity of the support.
Robustness and durability: consider the environmental conditions in which the sensing device will be deployed, including temperature, humidity, and exposure to outdoor elements, to ensure durability and reliability in harsh environments (extreme temperatures, rain, snow, etc.). Make sure the equipment and its casing are made for this if you want a reliable and durable installation. For instance, electronic devices or boards typically have specified temperature ranges within which they can operate effectively. In environments with more extreme temperatures, such as a village atop a mountain where temperatures can fluctuate significantly, standard consumer-grade devices may not suffice. In such cases, specialized electronic devices designed for military or industrial use, with extended temperature ranges and ruggedized construction, become necessary for reliable data monitoring. Moreover, the enclosures or casings housing these sensing devices play a pivotal role in protecting them from environmental hazards. These enclosures must be designed to withstand water ingress, snow accumulation, UV radiation from direct sunlight, and other external factors that could potentially compromise the integrity of the device. For example, in a scenario where sensors are deployed for environmental monitoring in a forest, the enclosures must be waterproof to protect against rain and snow infiltration. Similarly, in agricultural settings where sensors are exposed to direct sunlight for prolonged periods, UV-resistant enclosures are essential to prevent damage from sun exposure.
Power requirements: most sensing devices will require very low power demands. However, this is not to be overlooked, as access to the power grid may not be available everywhere. Determine the power requirements of the device, including battery life and power consumption, to ensure continuous operation in remote or off-grid locations. This is especially important for wearables. If the context requires short deployments (some days up to a week), maybe an autonomous device (with a battery) is the best way to go as it will require smaller efforts. Many sensing devices may work as well with USB power, so a power bank may save your data collection needs for some days, but make sure it provides enough power for the sensing device. Finally, solar panels can also be a good option, but they are generally more expensive, and in some latitudes, they might not provide enough energy in cloudy days or winter months, or in windy locations can be a safety issue (as they can get ripped off) so make sure you estimate the panel size and batteries correctly. In all cases, keep in mind that all solutions may attract more attention and be stolen, especially those that involve solar panels or batteries. There are also some low-tech and passive sampling methods such as NO2 tubes where, by design, energy is not an issue.
Readiness to deploy: the diversity of solutions comes with a different set of requirements on each case. Some air quality sensing devices are fully assembled, tested and calibrated by the manufacturer, which will have an impact on the cost. However, in some other cases, there will be devices that require some more involvement, for instance in terms of assembly of the components (for instance, a kit) or calibration, in which case additional equipment will be necessary. It is important to keep this in mind and, if possible, discuss with the device/monitor provider, check if there are guides and clear instructions for this process and assess the resources available within your team, both in terms of personnel, skills and economic resources. There is no rule of thumb but managing many devices (more than 30-50 units) will require significant effort, depending on the complexity, maturity and experience available. Finally, there may need to be other unforeseen requirements, such as computers and software needed to interact with the devices, smartphones for their configuration and so on. All these, including the usability of each of those, should be factored in when considering the various options available, and adapting them to each specific context. It is also important to check the connection to Sensor Data Platform: Evaluate whether the sensor device is compatible with existing sensor data platforms or cloud-based services for data storage, analysis, and visualization; and if the Data Platform provides what you need. Additionally, assess the availability of application programming interfaces (APIs) facilitating interoperability.
Useful resources
An EPA guide on low-cost air pollution monitors: https://www.epa.gov/indoor-air-quality-iaq/low-cost-air-pollution-monitors-and-indoor-air-quality
South Coast AQMD’s AQ-SPEC program evaluates sensors in the field, under ambient conditions, and in the laboratory under controlled environmental conditions for sensors measuring criteria pollutants (PM2.5, PM10, NO2, ozone, carbon monoxide) and volatile organic compounds (VOC) that include air toxics http://www.aqmd.gov/aq-spec/evaluations.
You might also be interested in….
Last updated
Was this helpful?