Measurements at appropriate spatial and temporal scales are essential for understanding and monitoring spatially heterogeneous environments with complex and highly variable emission sources, such as in urban areas. However, the costs and complexity of conventional air quality measurement methods means that measurement networks are generally extremely sparse. In this paper we show that miniature, low-cost electrochemical gas sensors, traditionally used for sensing at parts-per-million (ppm) mixing ratios can, when suitably configured and operated, be used for parts-per-billion (ppb) level studies for gases relevant to urban air quality. Sensor nodes, in this case consisting of multiple individual electrochemical sensors, can be low-cost and highly portable, thus allowing the deployment of scalable high-density air quality sensor networks at fine spatial and temporal scales, and in both static and mobile configurations. In this paper we provide evidence for the performance of electrochemical sensors at the parts-per-billion level, and then outline results obtained from deployments of networks of sensor nodes in both an autonomous, high-density, static network in the wider Cambridge (UK) area, and as mobile networks for quantification of personal exposure. Examples are presented of measurements obtained with both highly portable devices held by pedestrians and cyclists, and static devices attached to street furniture. The widely varying mixing ratios reported by this study confirm that the urban environment cannot be fully characterised using sparse, static networks, and that measurement networks with higher resolution (both spatially and temporally) are required to quantify air quality at the scales which are present in the urban environment. We conclude that the instruments described here, and the low-cost/high-density measurement philosophy which underpins it, have the potential to provide a far more complete assessment of the high-granularity air quality structure generally observed in the urban environment, and could ultimately be used for quantification of human exposure as well as for monitoring and legislative purposes.

Abstract. The Atmospheric Pollution and Human Health in a Chinese Megacity (APHH-Beijing) programme is an international collaborative project focusing on understanding the sources, processes and health effects of air pollution in the Beijing megacity. APHH-Beijing brings together leading China and UK research groups, state-of-the-art infrastructure and air quality models to work on four research themes: (1) sources and emissions of air pollutants; (2) atmospheric processes affecting urban air pollution; (3) air pollution exposure and health impacts; and (4) interventions and solutions. Themes 1 and 2 are closely integrated and support Theme 3, while Themes 1–3 provide scientific data for Theme 4 to develop cost-effective air pollution mitigation solutions. This paper provides an introduction to (i) the rationale of the APHH-Beijing programme and (ii) the measurement and modelling activities performed as part of it. In addition, this paper introduces the meteorology and air quality conditions during two joint intensive field campaigns – a core integration activity in APHH-Beijing. The coordinated campaigns provided observations of the atmospheric chemistry and physics at two sites: (i) the Institute of Atmospheric Physics in central Beijing and (ii) Pinggu in rural Beijing during 10 November–10 December 2016 (winter) and 21 May–22 June 2017 (summer). The campaigns were complemented by numerical modelling and automatic air quality and low-cost sensor observations in the Beijing megacity. In summary, the paper provides background information on the APHH-Beijing programme and sets the scene for more focused papers addressing specific aspects, processes and effects of air pollution in Beijing.

In the last few years, the interest in the development of new pervasive or mobile implementations of air quality multisensor devices has significantly grown. New application opportunities appeared together with new challenges due to limitations in dealing with rapid pollutants concentrations transients both for static and mobile deployments. In this work, we propose a Dynamic Neural Network (DNN) approach to the stochastic prediction of air pollutants concentrations by means of chemical multisensor devices. DNN architectures have been devised and tested in order to tackle the cross sensitivities issues and sensors inherent dynamic limitations. Testing have been performed using an on-field recorded dataset from a pervasive deployment in Cambridge (UK), encompassing several weeks. The results obtained with the dynamic model are compared with the response of the static neural network and the performance analysis indicates the capability of the on-field dynamic multivariate calibration to ameliorate the static calibration approach performance in this real world air quality monitoring scenario. Interestingly, results analysis also suggests that the improvements are more significant when pollutants concentration changes more rapidly.

No abstract

No abstract

Chemical multisensor devices need calibration algorithms to estimate gas concentrations. Their possible adoption as indicative air quality measurements devices poses new challenges due to the need to operate in continuous monitoring modes in uncontrolled environments. Several issues, including slow dynamics, continue to affect their real world performances. At the same time, the need for estimating pollutant concentrations on board the devices, especially for wearables and IoT deployments, is becoming highly desirable. In this framework, several calibration approaches have been proposed and tested on a variety of proprietary devices and datasets; still, no thorough comparison is available to researchers. This work attempts a benchmarking of the most promising calibration algorithms according to recent literature with a focus on machine learning approaches. We test the techniques against absolute and dynamic performances, generalization capabilities and computational/storage needs using three different datasets sharing continuous monitoring operation methodology. Our results can guide researchers and engineers in the choice of optimal strategy. They show that non-linear multivariate techniques yield reproducible results, outperforming linear approaches. Specifically, the Support Vector Regression method consistently shows good performances in all the considered scenarios. We highlight the enhanced suitability of shallow neural networks in a trade-off between performance and computational/storage needs. We confirm, on a much wider basis, the advantages of dynamic approaches with respect to static ones that only rely on instantaneous sensor array response. The latter have been shown to be best choice whenever prompt and precise response is needed.

No abstract

Low cost, personal air pollution sensors may reduce exposure measurement errors in epidemiological investigations and contribute to citizen science initiatives. Here we assess the validity of a low cost personal air pollution sensor. Study participants were drawn from two ongoing epidemiological projects in Barcelona, Spain. Participants repeatedly wore the pollution sensor - which measured carbon monoxide (CO), nitric oxide (NO), and nitrogen dioxide (NO). We also compared personal sensor measurements to those from more expensive instruments. Our personal sensors had moderate to high correlations with government monitors with averaging times of 1-h and 30-min epochs (r ~ 0.38-0.8) for NO and CO, but had low to moderate correlations with NO (~0.04-0.67). Correlations between the personal sensors and more expensive research instruments were higher than with the government monitors. The sensors were able to detect high and low air pollution levels in agreement with expectations (e.g., high levels on or near busy roadways and lower levels in background residential areas and parks). Our findings suggest that the low cost, personal sensors have potential to reduce exposure measurement error in epidemiological studies and provide valid data for citizen science studies.