RESUMO
The COVID-19 pandemic has affected people's lives throughout the world. Governments have imposed restrictions on business and social activities to reduce the spread of the virus. In the US, the pandemic response has been largely left to state and local governments, resulting in a patchwork of policies that frequently changed. We examine travel behavior across income and race/ethnic groups in Los Angeles County over several stages of the pandemic. We use a difference-in-difference model based on mobile device data to compare mobility patterns before and during the various stages of the pandemic. We find a strong relationship between income/ethnicity and mobility. Residents of low-income and ethnic minority neighborhoods reduced travel less than residents of middle- and high-income neighborhoods during the shelter-in-place order, consistent with having to travel for work or other essential purposes. As public health rules were relaxed and COVID vaccines became available, residents of high-income and White neighborhoods increased travel more than other groups, suggesting more discretionary travel. Our trip purpose model results show that residents of low-income and ethnic minority neighborhoods reduced work and shopping travel less than those of White and high-income neighborhoods during the shelter-in-place order. Results are consistent with higher-income workers more likely being able to work at home than lower-income workers. In contrast, low-income/minorities apparently have more constraints associated with work or household care. The consequence is less capacity to avoid virus risk. Race and socioeconomic disparities are revealed in mobility patterns observed during the COVID-19 pandemic.
RESUMO
Predicting PM2.5 concentrations at a fine spatial and temporal resolution (i.e., neighborhood, hourly) is challenging. Recent growth in low cost sensor networks is providing increased spatial coverage of air quality data that can be used to supplement data provided by monitors of regulatory agencies. We developed an hourly, 500 × 500 m gridded PM2.5 model that integrates PurpleAir low-cost air sensor network data for Los Angeles County. We developed a quality control scheme for PurpleAir data. We included spatially and temporally varying predictors in a random forest model with random oversampling of high concentrations to predict PM2.5. The model achieved high prediction accuracy (10-fold cross-validation (CV) R2 = 0.93, root mean squared error (RMSE) = 3.23 µg/m3; spatial CV R2 = 0.88, spatial RMSE = 4.33 µg/m3; temporal CV R2 = 0.90, temporal RMSE = 3.85 µg/m3). Our model was able to predict spatial and diurnal patterns in PM2.5 on typical weekdays and weekends, as well as non-typical days, such as holidays and wildfire days. The model allows for far more precise estimates of PM2.5 than existing methods based on few sensors. Taking advantage of low-cost PM2.5 sensors, our hourly random forest model predictions can be combined with time-activity diaries in future studies, enabling geographically and temporally fine exposure estimation for specific population groups in studies of acute air pollution health effects and studies of environmental justice issues.
