Deriving high-resolution urban air pollution maps using mobile sensor nodes

摘要: Up-to-date information on urban air pollution is of great importance for environmental protection agencies to assess quality and provide advice the general public in a timely manner. In particular, ultrafine particles (UFPs) are widely spread environments may have severe impact human health. However, lack knowledge about spatio-temporal distribution UFPs hampers profound evaluation these effects. this paper, we analyze one largest spatially resolved UFP data set publicly available today containing over 50 million measurements. We collected measurements throughout more than two years using mobile sensor nodes installed top transport vehicles city Zurich, Switzerland. Based data, develop land-use regression models create maps with high spatial resolution 100?m?i??100?m. compare accuracy derived across various time scales observe rapid drop sub-weekly temporal resolution. To address problem, propose novel modeling approach that incorporates past annotated metadata into process. way, achieve 26% reduction root-mean-square error-a standard metric evaluate models-of semi-daily believe our findings can help epidemiologists better understand adverse health effects related serve as stepping stone towards detailed real-time assessment.