In this lesson, you will learn:

• Why it is important to calibrate PM_2.5 and NO₂ data from Clarity devices

Now that we’ve covered some important definitions let's walk through why we need to calibrate at all. We’ll take it pollutant-by-pollutant.

Why we calibrate PM_2.5

Clarity Nodes use low-cost optical particle counter (OPC) sensors to measure PM mass concentration. OPCs do not measure PM mass concentration directly but rather measure particle count and size. In order to calculate the actual PM mass concentration, which most people are interested in, the OPC needs information about particle composition. An assumed particle composition factor is programmed into each OPC at the factory.

The assumed particle composition often doesn’t match the composition found at a particular project site because particle composition can vary widely depending on the mix of pollution sources found in a given location. The discrepancy between assumed particle composition and actual particle composition is the main source of measurement error found when monitoring PM mass concentration with OPCs. PM calibration fine-tunes the composition assumptions to match as closely as possible with the particulate composition found at the project site, minimizing measurement error.

Example of raw (grey line) and calibrated (blue line) PM_2.5 data. The raw Clarity PM_2.5 data overpredicts relative to the reference (black line). After calibration, the Clarity data matches the reference data much more closely.

Why we calibrate NO₂

Clarity Nodes, along with most other low-cost sensors, use Electrochemical Cell Sensors (ECS) to measure NO₂ concentration. As the air passes over the ECS, a chemical reacts with the NO₂ present in the air and generates a current, which is then translated to NO₂ concentration using a scaling factor.

Several things can interfere with that measurement and the overall sensor accuracy, the most important being changes in temperature and humidity at the project site. Clarity’s NO₂ calibration corrects for those interferences and scales the raw NO₂ signal to match the output of a reference instrument the device is collocated with. Even with this calibration applied, however, there are limitations to the ECS data. For example, it is often noisy in hot and dry conditions. Hot and dry conditions refer to a temperature over 30 degrees C and Relative Humidity of less than 45%. Additionally, data quality often shifts seasonally, with worse performance during the summer months.

Calibration is especially important for NO_2, and because of the interferences we see, we do not recommend using the raw NO₂ data to understand your air quality.

Time series plot showing uncalibrated (black line) and calibrated (colorful lines) NO₂ data. Calibration corrects for the interference of temperature and relative humidity on the raw NO₂ data.