Hierarchical network design for nitrogen dioxide measurement in urban environments. (1st May 2020)

• Record Type: Journal Article
• Authors: Weissert, Lena; Miles, Elaine; Miskell, Georgia; Alberti, Kyle; Feenstra, Brandon; Henshaw, Geoff S.; Papapostolou, Vasileios; Patel, Hamesh; Polidori, Andrea; Salmond, Jennifer A.; Williams, David E.
• Is Part Of: Atmospheric environment. Volume 228(2020)
• Publication Date: 2020-05-01
• Subjects: Air quality sensors -- Nitrogen dioxide -- Low-cost sensor network -- Calibration; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2020.117428
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: We present a management and data correction framework for low-cost electrochemical sensors for nitrogen dioxide (NO2 ) deployed within a hierarchical network of low-cost and regulatory-grade instruments. The framework is founded on the idea that it is possible in a suitably configured network to identify a source of reliable 'proxy' data for each sensor site that has a similar probability distribution of measurement values over a suitable time period, and that sensor data can be checked and corrected by comparison of the sensor data distribution with that of the proxy. The framework is rule-based and easily modified. We use the reference network to choose proxies and check proxy reliability. We demonstrate the application of this methodology to low-cost instruments that use an electrochemical NO2 sensor together with a semiconducting oxide-based sensor for ozone (O3 ). The three NO2 sensor response parameters (offset, O3 response slope, and NO2 response slope) which are known to vary significantly as a consequence of ambient humidity and temperature variations, we show can be estimated by minimising statistical measures of divergence between sensor-estimated and proxy NO2 distributions over a 3-day window. We show how the parameter variations and statistical divergence measures with respect to the proxy can be used to indicate error conditions. The major error is due to a diurnally-varying, spatially-correlated offset term that is large for extremes of temperature, which we show can be estimated through its spatial correlation, using sensors co-located at reference sites. With these procedures, we demonstrate measurement at nine different locations across two regions of Southern California over seven months with average root mean square error ±7.2 ppb (range over locations 4–11 ppb) without calibration other than the remote proxy comparison. We apply the procedures to a network of 56 sensors distributed across the Inland Empire and Los Angeles County regions. The results show large variations in NO2 concentration taking place on short time- and distance scales across the region. These spatiotemporal NO2 variations were not captured by the more sparsely distributed regulatory network of air monitoring stations demonstrating the need for reliable data from dense networks of monitors to supplement the existing regulatory networks. Graphical abstract: Image 1 Highlights: Shows how to get reliable data from an extensive low-cost network of NO2 sensors. Shows drift correction methods for electrochemical sensors in hierarchical network. Demonstrates use of statistical comparison to trusted proxy. Land-use similarity used to select reference sites as proxies, which can be distant. Uncertainties low enough to resolve spatio-temporal variations over short distances.