Air quality model validation in urban area: a new approach using a wireless pervasive sensor system.

Abstract

This paper describes the implementation in Palermo of a new approach developed at Newcastle University (UK), in the context of the MESSAGE project (Mobile Environmental Sensing System Across Grid Environments), using a novel inexpensive pervasive sensor system. The new system infrastructure shows how additional data from the new environmental sensors, namely “mote” can improve the assessment of the impact of traffic and congestion on air quality. After an evaluation and validation process of the sensors against the precision system (City Council Environmental Cabin) the sensors where deployed in Palermo and using the data, actual traffic flows and pedestrian traffic light information (all synchronised at one minute resolution), a simulation model was set up to predict air quality, and potentially noise and exposure levels for pedestrian. The proposed approach has been evaluated using a case study area in Palermo, namely a street canyon, in Libertà Avenue. Analysis, calibration and validation of data from two static pervasive sensors, will be presented in this paper. Next the process by which the traffic characteristics (fleet composition, flow, queue length) collected over two months survey were used to validate parameters of the AIMSUN traffic simulation model (flows, queues, congestion states) will be shown. The validated microsimulation model subsequently was used to predict tailpipe emissions taking into account the traffic flow regimes (free flow, unstable and congested) and pollution concentrations which were estimated within the street canyon using the OSPM (Operational Street Pollution Model) model.These estimates will be compared with the pervasive sensor measurements of air pollution, CO, Carbon Monoxide and NO2, Nitrogen dioxide. Results presented in this paper will highlight the benefits of pervasive sensors in the model validation and how they can compliment legacy systems through their flexibility in covering detection gaps in existing urban networks. Moreover, the proposed approach will have the capacity to assess the changes caused in previously unmonitored areas, resulting in more realistic assessments of the impacts of transport policies and strategies. Finally, an evaluation of emissions impact from new pedestrian traffic light scenarios is presented.