Members Area News Site Map Contact עברית
The Technion Center of Excellence in Exposure Science and Environmental Health
Join our mailing list
Name Email Affiliation

Development of air pollution data-driven models

Uri Lerner, Ph.D. 

Exposure to air pollution has a proven impact on public health. Thus, arises the need for an accurate way to measure the level of exposure to various pollutants. Currently monitoring is obtained by stationary AQM (Air Quality Monitoring) stations, which are spread thinly due to size and cost. Recent developments in sensory and communication technologies have made small and relatively low-cost sensing units feasible. Due to their lower cost, these sensors can be spread more densely and provide higher-resolution spatial data.
This research aims at quantifying the sensor’s velocity impact on the measurements. To obtain this, a 10-hour field campaign was conducted in Neve Sha'anan neighborhood in Haifa, Israel. Three cars were mounted with Airbase’s "CanarIT"1 units, measuring pollutants and environmental parameters, and were driven in a specific route. At any given time, one car was parked along the path. To explore the effect of travel speed on the measured values, the mobile sensor’s output was compared with that of the stationary one.
The results show that temperature, relative humidity and noise levels were not affected by speed. When the sensor was mounted perpendicular to the travel direction, O3 measurements showed no influence of speed variance. However, when the sensor was front-facing, a 4 to 10 folds increase in measured values was observed, for travel speeds of over 30 km/h, when compared to those lower than 15 km/h.
When increasing a car’s travel speed, the variability increases. Additionally, an increase of O3 mobile sensor’s reading was observed. Thus, there is a clear influence of travel speed, when monitoring O3 levels using a front-facing sensor. This is the first report of any influence of varying speeds on sensors' ability to accurately measure O3 levels. Any future design of experimental protocols for using small-AQM units should take this phenomenon into account.