Selection of the stations Sample Clauses

Selection of the stations. Since the objective is to obtain an indicator of the mean exposure to air pollution, only monitoring stations representing background pollution should be included in the study. For PM10 and PM2.5, this means that urban background monitors should be selected; for ozone, urban and suburban background monitors should be selected. Traffic and industrial monitors should never be included in the study. The expertise of the people in charge of the monitoring network should be used to identify the correct stations. As some stations may be incorrectly labelled or may be faulty, we propose using simple criteria to help decide if a station should be included or not: Whenever possible, at least two monitors should be available, which allows averaging the concentrations to obtain a better estimate of the exposure. In practice: - List all the available stations in the area: • Urban background stations for PM10 and PM2.5 • Urban and suburban background stations for ozone - Exclude stations with not enough data • Only stations with less than 25% missing data should be used - Exclude stations that may not be representative of the mean exposure: • Exclude the stations if the interquartile range (P25-P75) of the data from one monitor does not overlap with the interquartile ranges of the other monitors. • Exclude the station if its correlation coefficients with the other stations are lower than 0.6 - Try to have more than one station per pollutant - Validate the choice with an expert from the monitoring network