Predicting traffic-related air pollution in Los Angeles using a distance decay regression selection strategy

Abstract

Land use regression (LUR) has emerged as an effective means of estimating exposure to air pollution in epidemiological studies. We created the first LUR models of nitric oxide (NO), nitrogen dioxide (NO₂) and nitrogen oxides (NO_X) for the complex megalopolis of Los Angeles (LA), California. Two-hundred and one sampling sites (the largest sampling design to date for LUR estimation) for two seasons were selected using a location-allocation algorithm that maximized the potential variability in measured pollutant concentrations and represented populations in the health study. Traffic volumes, truck routes and road networks, land use data, satellite-derived vegetation greenness and soil brightness, and truck route slope gradients were used for predicting NO_X concentrations. A novel model selection strategy known as “ADDRESS” (A Distance Decay REgression Selection Strategy) was used to select optimized buffer distances for potential predictor variables and maximize model performance.

Final regression models explained 81%, 86% and 85% of the variance in measured NO, NO₂ and NO_X concentrations, respectively. Cross-validation analyses suggested a prediction accuracy of 87–91%. Remote sensing-derived variables were significantly correlated with NO_X concentrations, suggesting these data are useful surrogates for modeling traffic-related pollution when certain land use data are unavailable. Our study also demonstrated that reactive pollutants such as NO and NO₂ could have high spatial extents of influence (e.g., >5000 m from expressway) and high background concentrations in certain geographic areas. This paper represents the first attempt to model traffic-related air pollutants at a fine scale within such a complex and large urban region.

Introduction

With a population of 16.7 million, the Los Angeles Metropolitan Area (hereafter referred to as LA) is the largest urban area in the state of California and the second-largest in the United States. LA is also consistently ranked as one of the most polluted metropolitan areas in the US, partially due to heavy reliance on automobiles for transportation, and because more than 40% of all goods transported into the United States move through the Long Beach/Los Angeles port complex, which generates thousands of truck trips per day (Hricko, 2008). The LA Basin is also susceptible to atmospheric inversions (Bytnerowicz and Fenn, 1996), which trap exhaust from on-road vehicles, airplanes, and locomotives, as well as from shipping, manufacturing and other industrial sources in the troposphere. Additionally, with an average rainfall of only 381 mm/year, Los Angeles experiences minimal pollution removal by precipitation. The high level of auto-dependence in LA is replicating in other major conurbations around the world, particularly in newly industrialized countries (Walsh, 2008). Lessons learned from studying air pollution exposures and subsequent health effects in Los Angeles may therefore have widespread applicability to other large urban areas.

This paper reports the first attempt to model NO, NO₂ and NO_X concentrations in LA (Fig. 1) using a land use regression (LUR) approach. The LUR was developed as part of a study sponsored by the California Air Resources Board (CARB) to examine the impacts of outdoor air pollution on respiratory health in children living in LA. The LUR method seeks to predict pollution concentrations at a given site based on surrounding land use, road network, traffic, physical environment and population characteristics using a series of buffers (Su et al., 2009). Geospatial modeling techniques including LUR are an attractive alternative to ambient (government) air monitoring data for assessing traffic pollutant exposures, since they can be applied to large populations and account for neighborhood-scale variations in pollutant concentrations. To our knowledge, LUR to date has not been applied to large urban areas with population over 10 million for traffic-related air pollution (NO_X) but only to estimate more spatially homogenous air pollutant concentrations such as fine particulate matter (Moore et al., 2006; Ross et al., 2007).