Abstract View
From Low-Cost Sensors to High-Quality Data: the Importance of Collocated Calibration Model Development for the AfriqAir Network
MICHAEL R. GIORDANO, Julien Bahino, Matthias Beekmann, Thomas Bigala, Paola Formenti, Mathieu Cazaunau, Corinne Galy-Lacaux, Jimmy Gasore, Theobald Habineza, Aliaksei Hauryliuk, Vincent Madadi, Carl Malings, Beatrice Marticorena, Stuart Piketh, Albert A. Presto, R. Subramanian, Daniel Westervelt, Veronique Yoboue, OSU-EFLUVE, LISA/CNRS, UPEC, ENPC, UP
Abstract Number: 92
Working Group: Urban Aerosols
Abstract
AfriqAir is a hybrid network of low-cost sensors centered around reference monitors. About 50 Real-time Affordable Multi-Pollutant (RAMP), Clarity, and PurpleAir devices have been collaboratively deployed in Nairobi, Kigali, Accra, Abidjan, Niamey, Kinshasa, Zamdela (near Johannesburg), and other African cities. All devices use Plantower optical nephelometers for PM2.5. Generating high-quality data from low-cost sensors requires calibration models to accurately measure PM concentrations. One way to build such calibration models is by collocating reference-grade monitors with the sensors for a set amount of time before deploying the sensors to the target locations. One outstanding question that arises from this methodology is the applicability of the models developed in one location to another location as differences in ambient aerosol size distribution and chemical composition may greatly impact calibration models. Here we examine this question using data from low-cost sensors deployed in South Africa, Côte d’Ivoire (West Africa), and Uganda and Rwanda in East Africa. Calibration models for each location are developed and compared to calibration models developed in urban OECD locations (Pittsburgh, PA USA; the Greater Paris area, France; and Queens, NY, USA). Results show that the specific models for each location perform the best, but the urban OECD calibration models can do fairly well at reproducing measurements from reference-grade instruments collocated with the low-cost sensors deployed in some developing countries.