Community-based monitoring (CBM) encompasses different levels of citizens’ participation going from opportunistic sensing to crowd sourcing, participatory data collection and “citizen science”. In the urban environment with its complex physical and social interactions CBM can help in collecting data at a spatial and temporal scale that corresponds best to inhabitants personal environment and activities. It can also enhance people’s understanding of the urban environment, and as such be a part of collaborative decision-making processes.
A crucial issue is the availability and quality of low-cost sensing devices. This is i.e. the case for air quality monitoring. Portable instruments are available to measure components such as UFP and black carbon. They are however expensive, and large scale crowd-sourcing strategies are therefore not obvious. Still they can play a role in participatory monitoring exercises.
Several projects have developed portable devices, integrating low-cost gas sensors, GPS and mobile phones. However, quality issues, such as sensitivity, cross-interference, sensor drift, and susceptibility to temperature or humidity, are often overlooked.
We will present the EveryAware SensorBox, that will be used in community-based air quality monitoring in Belgium, Italy and the UK. To overcome the quality issues of single sensors, it is conceived as a portable multisensor array or “e-nose”. The prototype SensorBox contains a range of commercially available electrochemical and metal oxide gas sensors. Preliminary tests were carried out to select sensors based on their response to traffic related air pollution, using simultaneous mobile measurements of ultrafine particles (UFP) and black carbon (BC) as indicators, and to environmental conditions. On-field calibration methods using non-linear modeling, such as Gaussian Process regression, as well as appropriate data collection strategies will be presented. By the time of the conference we expect to be able to show the first results of field tests with the EveryAware SensorBox.
Authors: Jan Theunis (1), Bart Elen (1), Stefano Ingarra (2), Andrea Molino (3), Joris Van den Bossche (1/4), Matteo Reggente (1/5), Vittorio Loreto (2)
(1) VITO Flemish Institute for Technological Research, Boeretang 200, 2400 Mol, Belgium, firstname.lastname@example.org, Telephone: +32 14335372
(2) Fondazione I.S.I. – Istituto per l’Interscambio Scientifico, Via Alassio 11/c, 10126 Turin, Italy
(3) CSP – Innovazione nelle ICT s.c. a r.l., Via Alassio 11/c, 10126 Turin, Italy
(4) Ghent University, Faculty of Bioscience Engineering, Coupure Links 653, 9000 Ghent, Belgium
(5) Ghent University, Department of Information Technology (INTEC), Sint-Pietersnieuwstraat 41, 9000 Ghent, Belgium
(6) Sapienza University of Rome, Physics Department, Piazzala Aldo Moro 2, 00185, Rome, Italy
Where: this paper has been presented at the Innovation for Sustainable Production 2012 (Bruges 6-9 May 2012)