| Autor: |
Lu T; Department of Earth Science & Geography, California State University Dominguez Hills, 1000 E. Victoria Street, Carson 90747, California, United States., Marshall JD; Department of Civil & Environmental Engineering, University of Washington, 201 More Hall, Seattle 98195, Washington, United States., Zhang W; Edward J. Bloustein School of Planning and Public Policy, Rutgers University, 33 Livingston Avenue, New Brunswick 08901, New Jersey, United States., Hystad P; College of Public Health and Human Sciences, Oregon State University, 2520 Campus Way, Corvallis 97331, Oregon, United States., Kim SY; Department of Cancer Control and Population Health, Graduate School of Cancer Science and Policy, National Cancer Center, Goyang-si, Gyeonggi-do 10408, Korea., Bechle MJ; Department of Civil & Environmental Engineering, University of Washington, 201 More Hall, Seattle 98195, Washington, United States., Demuzere M; Urban Climatology Group, Department of Geography, Ruhr-University Bochum, Bochum 44801, Germany., Hankey S; School of Public and International Affairs, Virginia Tech, 140 Otey Street, Blacksburg 24061, Virginia, United States. |
| Abstrakt: |
National-scale empirical models of air pollution (e.g., Land Use Regression) rely on predictor variables (e.g., population density, land cover) at different geographic scales. These models typically lack microscale variables (e.g., street level), which may improve prediction with fine-spatial gradients. We developed microscale variables of the urban environment including Point of Interest (POI) data, Google Street View (GSV) imagery, and satellite-based measures of urban form. We developed United States national models for six criteria pollutants (NO 2 , PM 2.5 , O 3 , CO, PM 10 , SO 2 ) using various modeling approaches: Stepwise Regression + kriging (SW-K), Partial Least Squares + kriging (PLS-K), and Machine Learning + kriging (ML-K). We compared predictor variables (e.g., traditional vs microscale) and emerging modeling approaches (ML-K) to well-established approaches (i.e., traditional variables in a PLS-K or SW-K framework). We found that combined predictor variables (traditional + microscale) in the ML-K models outperformed the well-established approaches (10-fold spatial cross-validation (CV) R 2 increased 0.02-0.42 [average: 0.19] among six criteria pollutants). Comparing all model types using microscale variables to models with traditional variables, the performance is similar (average difference of 10-fold spatial CV R 2 = 0.05) suggesting microscale variables are a suitable substitute for traditional variables. ML-K and microscale variables show promise for improving national empirical models. |
