Abstract
Background: Land use regression models can have limited temporal representativeness or spatial resolution if mobile monitoring or fixed site data are used exclusively. Our objective was to build hourly spatial and temporal nitrogen oxides (NOx) models for 2003-2015 in Boston and Chelsea (MA, USA) by combining measurements from a mobile platform and a fixed reference site.Methods: NOx measurements were made with a chemiluminescence analyzer mounted in a mobile platform. Monitoring sessions (3-6 hours long) occurred on 49 days in Boston and 46 days in Chelsea between 2011 and 2015. Sessions occurred on all days of the week and in all seasons. We obtained hourly NOx measurements from a federal monitoring site in Boston. We calculated spatial factors, defined as the location-specific ratio between each 10-second mobile NOx value and the corresponding hourly mean fixed site NOx value. To model the spatial factors, we tested 43 covariates including transportation network and land use variables with data obtained from public sources. We prioritized covariates that most increased the adjusted-R2 and made physical sense. We multiplied the modeled spatial factors by the fixed site hourly mean to estimate NOx at <200-m and 1-h resolution.Results: The models were stable with two cross validation methods. The most important predictors were distance from major roads, open space, and residential areas; presence of bus/train stops; and an interaction term for wind speed and being downwind of an airport. The models over-predicted at three validation sites, especially when concentrations were low and in places farther from major roads; however, the adjusted-R2 values for hourly predictions were high (0.53-0.62) and the model captured seasonal and diurnal trends well for both study areas for 2003-2015.Conclusion: Our modeling approach is an efficient way to develop spatially and temporally resolved exposure estimates for use in concurrent and retrospective epidemiology studies.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
