Development of a multiple regression model to calibrate a low-cost sensor considering reference measurements and...

Area of Science:

• Environmental Science
• Sensor Technology
• Public Health

Background:

• Limited reference air quality monitors in Lima, Peru, hinder policy development.
• Low-cost sensors offer a scalable solution for enhanced air quality data resolution.
• Improving the data quality of low-cost sensors is crucial for developing countries.

Purpose of the Study:

• To evaluate the performance of low-cost air quality sensors in the Peruvian market.
• To calibrate low-cost sensors using reference monitor data.
• To identify high-performing sensors for informed decision-making.

Main Methods:

• Co-location of a low-cost sensor with a reference air quality monitor.
• Development of a multiple regression model using sensor data, temperature, and relative humidity.
• Analysis of sensor performance for PM2.5 and PM10 measurements.

Main Results:

• The evaluated low-cost sensor technology shows promising performance for measuring PM2.5 and PM10.
• PM2.5 measurements demonstrated a better correlation with reference data compared to PM10.
• Sensor-derived meteorological data (temperature, relative humidity) showed limited correlation with reference data.

Conclusions:

• Low-cost sensors are a viable tool for augmenting air quality monitoring networks in data-scarce regions.
• Calibration against reference monitors is essential for ensuring the reliability of low-cost sensor data.
• Further research is needed to optimize the use of low-cost sensors for both particulate matter and meteorological measurements.