RESUMEN
Addressing the challenge of mapping hyperlocal air pollution in areas without local monitoring, we evaluated unsupervised transfer learning-based land-use regression (LUR) models developed using mobile monitoring data from other cities: CORrelation ALignment (Coral) and its inverse distance-weighted modification (IDW_Coral). These models mitigated domain shifts and transferred patterns learned from mobile air quality monitoring campaigns in Copenhagen and Rotterdam to estimate annual average air pollution levels in Amsterdam (50m road segments) without involving any Amsterdam measurements in model development. For nitrogen dioxide (NO2), IDW_Coral outperformed Copenhagen and Rotterdam LUR models directly applied to Amsterdam, achieving MAE (4.47 µg/m3) and RMSE (5.36 µg/m3) comparable to a locally fitted LUR model (AMS_SLR) developed using Amsterdam mobile measurements collected for 160 days. IDW_Coral yielded an R2 of 0.35, similar to that of the AMS_SLR based on 20 collection days, suggesting a minimum requirement of 20-day mobile monitoring to capture city-specific insights. For ultrafine particles (UFP), IDW_Coral's citywide predictions strongly correlated with previously published mixed-effect models fitted with 160-day Amsterdam measurements (Pearson correlation of 0.71 for UFP and 0.72 for NO2). IDW_Coral demands no direct measurements in the target area, showcasing its potential for large-scale applications and offering significant economic efficiencies in executing mobile monitoring campaigns.
Asunto(s)
Contaminantes Atmosféricos , Contaminación del Aire , Monitoreo del Ambiente , Monitoreo del Ambiente/métodos , Material Particulado , Dióxido de Nitrógeno/análisis , CiudadesRESUMEN
Black carbon (BC) has emerged as a major contributor to global climate change. Cities play an important role in global BC emission. The present study investigated the spatial pattern of in-traffic BC at a high spatial resolution in Shanghai, the commercial and financial center in Mainland China. The determinants including road network, social economic status and point-source pollutants, which may influence the BC spatial variability were also discussed. From October to December 2016, mobile monitoring was conducted to assess the BC concentrations on three sampling routes in Shanghai with a total length of 116â¯km. The results showed that the mean in-traffic BC among three sampling routes was 10.77⯱â¯3.50⯵g/m3. BC concentrations showed a significant spatial heterogeneity. The highest BC concentrations were near industrial sources and that those high concentrations were associated with either direct emissions from the industries, freight traffic, or both. With the widely distributed polluting enterprises and high emitting vehicles, the average BC in the low urbanization areas (12.80⯱â¯4.54⯵g/m3) was 57% higher than that in the urban core (7.77⯱â¯2.24⯵g/m3). Furthermore, a land use regression (LUR) model based on mobile monitoring was developed to examine the determinants and its spatial variability of BC measurements which corresponded to 17 predictor variables, e.g. road network, land use, meteorological condition etc., in 7 buffer distances (100â¯m to 10â¯km). The variables of meteorological, socio-economical and the distance to BC point-sources were selected as the independent variables. It was found that the established LUR model could explain a proportion (68%) of the variability of BC. LUR modeling from mobile measurements was possible, but more work related to the effect of traffic regulation on BC could be helpful for informing best model practice.