Urban air monitoring stations are used to measure city-wide pollution levels (i) for regulatory purposes and (ii) to estimate the average exposure for the population. However, the spatial coverage from one or at most a few regulatory monitoring stations is insufficient to capture the spatial variability in PM concentrations across urban areas. This lack of spatial data represents a serious knowledge gap when performing human inhalation exposure studies. Inadequate spatial resolution may mask the exposure variability in the study population and may lead to potential exposure misclassifications. The intra-urban variation of air quality can be resolved by using land-use regression (LUR) models, which require sampling measurement campaigns at multiple locations along with a set of predictor variables derived from geographic information systems (e.g., various traffic representations, population density, land use, physical geography and climate). However, such sampling campaigns are usually short in time or use mobile monitoring networks due to the high cost for deployment and maintenance of scientific-grade instruments. The recent development of low-cost PM monitors is potentially a viable solution to increase the spatial resolution of air quality monitoring and to perform long-term sampling campaigns at multiple sites. This study aims to produce an hourly-resolved LUR model to predict hourly PM concentrations at individual locations across Monroe County, NY. The study uses spatially-resolved PM data measured with low-cost monitors for the development of the LUR model. Low cost PM monitors (Speck, Airviz Inc. PA, USA) were placed at 23 sites across Monroe County, NY during the 2015/16 heating season (November to March). Data were handled according to the results of previous intercomparison studies with reference instruments to assure a robust and reliable dataset. Coefficients of determination for hourly PM predictions between 0.6 and 0.65 for the various sites, which is comparable with previous LUR studies for daily PM predictions.
MASIOL M (Corresponding)
|Data di pubblicazione:||2017|
|Titolo:||Predicting the Spatial Variability of PM in Urban Areas with Low-Cost Monitors and Land Use Regression Modelling|
|Titolo del libro:||2017 AAAR Annual Conference Abstracts|
|Appare nelle tipologie:||4.2 Abstract in Atti di convegno|
File in questo prodotto:
| AAAR2017 - Masiol - LUR.pdf||Versione dell'editore||Accesso chiuso-personale||Riservato|