Abstract View
Performance and In-field Calibration for Low-Cost Sensors Measuring Ambient Particulate Matter in Kolkata, India
V. FAYE MCNEILL, Columbia University
Abstract Number: 177
Working Group: Urban Aerosols
Abstract
We present an in-field calibration of low-cost ambient particulate matter sensors (LCS) operating in greater Kolkata, India. Two datasets were examined: a network of 20 Clarity sensors operating from September 2018-April 2019 and a network of Purple Air sensors operating from August 2019-present. LCS were co-located with reference-grade PM monitors (RGM) operated by the West Bengal Pollution Control Board at Rabindra Bharati University (RBU), Victoria Memorial (VICTORIA), and Padmapukur (Howrah, PDM). Data from the co-locations were used to calibrate the LCS network using multiple linear regression and random forest approaches. Measured relative humidity and temperature were significant model features. Cross-calibration between co-located LCS-RGM pairs was tested. Calibrating PDM data using the model developed for the RBU co-located pair resulted in a 41.5% reduction of the RMSE. The cross-calibration yielded correct identification of days violating the 24h national ambient air quality standard for PM2.5 using the LCS network 85% of the time. The corrected data identifies days with an air quality index of “moderately poor” or worse 100% of the time, and “poor” or worse 97% of the time. This suggests that LCS can be a useful supplement to RGM networks for air quality management.