Sensors can provide air quality measurements on a display, an app on your phone, a file downloaded off the cloud, a map feature, or direct data transfer from the sensor to a computer. But what does the data mean?

 

  • Number

    Different types of sensors will display data differently but they will commonly report the concentration of an air contaminant. Concentration refers to the amount of a substance per defined space.

    The small sensors may also record other data such as temperature, humidity, pressure, or air quality index. If you’re interested in measuring air quality, it’s important to know which number is for the air contaminant.

    Unit

    For sensors that measure gases such as ozone or nitrogen dioxide, the units of the readings will typically be parts per billion (ppb) or parts per million (ppm).

    The sensors that measure particulate matter (PM) typically display measurements in µg/m3. This means microgram of particulate matter per metre cubed of air.

    For some sensors, it might not be obvious what the unit for the measurement is, so it’s important to confirm the unit before comparing the data to air quality guidelines or indices.

  • Date & Time

    Timestamps

    A timestamp is the date and time assigned to a sensor’s data point. For example, 10/17/2019 09:26 is a timestamp from October 17, 2019 at 9:26 AM. Not all sensors’ clocks are synchronized to the current time or local time zone, and they may not automatically adjust for daylight savings time. Some sensors might also display the time in “UTC” or Coordinated Universal Time, which is a standard time used across the world. If you’re comparing different sets of data, such as comparing your sensor data to data collected by Metro Vancouver’s air monitoring stations, you’ll have to convert the timestamp on all the data sets so they’re in the same time zone.

    If you are reading the concentrations from an app on your phone, pay close attention to the timestamp that is being recorded in comparison to the time on your phone. If you are reading the concentrations from a cloud-based application, pay close attention to the timestamp associated with the reading and the time on your computer.

    The format of the timestamp may be different depending upon how it has been programmed. The display may be Day/Month/Year or Month/Day/Year or Year/Month/Day. The format doesn’t matter, as long as you know which one it is.

    Sampling Frequency

    Sampling frequency refers to how often a sensor takes a measurement in a set period of time. For example, the sensor may record data every second, every 5 seconds, every minute, or some other time interval. The greater the sampling frequency the better representation of ambient air contaminant concentrations. If you wish to analyze the data in more detail, it will be important to know your sensor’s sampling frequency.

  • Comparisons

    Comparing To Metro Vancouver Air Monitoring Stations

    Now that you can understand small sensor data, how can you compare your measurements to Metro Vancouver’s air quality data? Metro Vancouver provides ambient air quality data in near real-time from 31 air quality stations in the Lower Fraser Valley Monitoring Network. This data is displayed on AirMap.

    AirMap is an interactive tool that displays concentrations of several air quality readings just by clicking on the desired air contaminant. It also provides notices on any current Air Quality Advisories and the Air Quality Health Index Rating for each site.

    If you are comparing small sensor readings to AirMap data, the following need to be considered:
    • Location of small sensor – Is your sensor indoors or outdoors? AirMap displays outdoor air quality data, which can be more variable than indoor air quality data. A sensor placed indoors can produce very different measurements from one placed outdoors.
    • Time – is the small sensor data referencing the same time as the AirMap data?
    • Units – are the units the same?
    • Averaging period – Is the sensor data an average for the same period of time (e.g. 1-minute, 1-hour, etc.) as the AirMap data?
    • Meteorological conditions – Is the weather at your sensor’s location similar to the weather at the station you’re comparing your data to? Temperature and humidity can influence air contaminant concentrations.
    • Daily patterns – Air quality can change during the day. For example, sunshine and heat increases ground-level ozone formation, so ozone tends to be higher during the day than at night.
    • Seasonal patterns – Air quality can vary depending on the season. For example, fine particulate matter (PM2.5) can increase during the summer from wildfire smoke, but there can also be localized increases during the winter from smoke from wood-burning. Ground-level ozone will also be higher during the summer than the winter.

    Comparing To Air Quality Objectives and Standards

    Metro Vancouver establishes outdoor air quality objectives to protect public health and the environment. These air quality objectives are some of the most stringent in the world.​

    Outside of Metro Vancouver, British Columbia uses provincially developed ambient air quality criteria to inform decisions on the management of air contaminants, called Provincial Air Quality Objectives(AQOs). Canada also has nationwide air quality standards, called the Canadian Ambient Air Quality Standards (CAAQs).

    Objectives can be for average concentrations over different periods of time, called averaging periods. Let’s say you want to compare your sensor’s measurements to Metro Vancouver’s 1-hour Ambient Air Quality Objective for ozone. Your sensor measures ozone concentrations every 10 minutes and you’ve collected this data:

    TimeConcentration (ppb)
    9:0011.3
    9:1012.0
    9:2011.9
    9:3013.0
    9:4013.4
    9:5014.0
    10:0014.1
    10:1014.0

    1. Add up the concentrations for the 9:00 hour (11.3+12.0+11.9+13.0+13.4+14.0 = 75.6 ppb)
    2. Divide by the number of data points (75.6 ppb/6 data points = 12.6 ppb)
    3. The average 1-hour ozone concentration for 9:00 is 12.6 ppb, which is well below Metro Vancouver’s Ambient Air Quality Objective of 82 ppb for ozone over 1-hour.

    This example was for a “time beginning” data set, which means the average for 9:00 is calculated using the data after this time, from 9:00 to 9:59. A “time ending” data set for 9:00 would use data from 8:00 to 8:59 to calculate the hourly average. When comparing different data sets, it’s important to know if the averages are time beginning or time ending.

    Sometimes the air quality guideline is based on a “rolling average”, such as Metro Vancouver’s 24-hour Ambient Air Quality Objective for fine particulate matter (PM2.5). This means you average concentrations for the previous 24-hours to calculate the 24-hour average for any given hour.

    When comparing sensor data to guidelines, it is important that you compare data for the same averaging period, such as a 1-hour average to a 1-hour guideline.

    Comparing to the Air Quality Health Index

    Some small sensors may display an air quality index reading instead of a concentration. Two common air quality indices are the Air Quality Health Index used in Canada and the Air Quality Index used in the United States.

    The Air Quality Health Index (AQHI) is developed by the Government of Canada and is the preferred index that Metro Vancouver residents should use to evaluate health risk.

    The Air Quality Health Index is a scale from 1 to 10+ to indicate the level of health risk associated with air quality. The higher the number, the greater the risk and the need to take precautions.

    The AQHI is designed as a guide to the relative risk presented by common air pollutants which are known to harm human health:

    1. Ground-level Ozone (O3)
    2. Fine Particulate Matter (PM2.5)
    3. Nitrogen Dioxide (NO2)

    In the development of the AQHI, a formula that combined these three pollutants was found to be the best indicators of the health risk of the combined impact of the mix of pollutants in the air.

    The Air Quality Index (AQI) is an American system which does not use the same scale or the same air contaminants in the calculation as the Canadian AQHI. The AQI should not be used by Metro Vancouver residents to evaluate health risk.

    It is important to know which index is displayed on the sensor, so you can understand if you need to take any actions to protect your health.

  • Weather Effects

    Temperature and humidity can affect small air sensor measurements, especially for sensors that measure particulate matter. For example, in humid conditions, fine particle (PM2.5) sensors can measure water droplets as particles, or water can deposit on particles to increase their size. Both of these scenarios would overestimate PM2.5 concentrations. The more sophisticated “reference monitors” used at government air monitoring stations have internal systems to ensure changes in temperature, humidity, or pressure do not affect air contaminant measurements.

    Some sensors don’t work well in the cold, so if you’re planning on keeping your sensor outside during our Canadian winters, it’s important to check its temperature ratings or environmental operating conditions, which indicate the range of temperatures and humidity in which the sensor is designed to function. The environmental operating conditions are usually listed in the sensor’s user manual.

    Another way you can learn how a sensor behaves in various weather conditions is by looking up scientific studies that test the performance of these sensors. The South Coast Air Quality Management District (SCAQMD) operates the Air Quality Sensor Performance Evaluation Center (AQ-SPEC), which tests small air sensors against more scientific air monitoring instruments. AQ-SPEC publishes a summary of each sensor evaluation, which can help you figure out if your sensor might be influenced by weather conditions.

  • Correction Factors

    Small sensors behave differently depending on local weather and environmental conditions. Researchers who have worked extensively with these devices have developed a “correction factor” for some small sensors. This adjusts the sensor data so it’s similar to the data collected by reference monitors in a given area.

    Developing a correction factor is not an easy task and is best done by researchers who are experienced working with small sensors. For example, researchers at the University of Northern British Columbia (UNBC) set up several PurpleAir sensors alongside reference monitors in Prince George, BC to see how the data the sensors collected compared to the data from reference monitors. Over several years, UNBC developed formulas to adjust the PurpleAir sensor data so it lined up with reference monitor data, and then created a website to show these adjusted values. They also worked with Environment and Climate Change Canada to refine the correction factor for PurpleAir sensors in western Canada.

    However, a correction factor developed for a sensor in a particular region might not apply to where you live. When looking at sensor data, check for correction factors for your sensor and region, but be careful about using correction factors that might not apply to your area.