부산대학교 도서관

Using low-cost air quality sensors (PM2.5, NO2, CO), air pumps, and a Raspberry Pi computer, we constructed a system by which air quality in tree canopies could be interrogated and quantified. The system involves pumping air into a sensor-containing box alternatively from tree canopy air and ambient air; repeating often enough to document if there are concentration differences between these two sources. By using the same set of sensors for air analysis from two sources, we eliminate issues such as sensor offset or drift and/or sensitivity to environmental conditions. True differences between tree canopy air and ambient air can be verified only after it has been established that the concentration difference between co-located inlet tubes is negligible. We've documented co-location results, described data summary protocol and as proof of concept, we show true differences in PM2.5(production) and CO (consumption) between ambient air and tree canopies on the University of Southern California's campus. In one tree tested, NO2between tree canopy and ambient air fluctuated as a function of day/night indicating periods of production and consumption. This system can be applied to document which tree species modify air quality, and how much, and can thus help urban forestry decision-makers when choosing tree planting under various environmental conditions.