Land Use Regression Model for Exposure Assessment to Pollutant Gases in Rio de Janeiro City, Brazil

BACKGROUND AND AIM: Air pollution is a public health problem. The data from WHO show that nine out of 10 people breathe air containing high levels of pollutants and the exposure to fine particles caused about 4.2 million deaths in 2016. Therefore, the aim of this study was to elaborate a model for long-term exposure assessment to air pollution gases. METHODS: This study was developed in Rio de Janeiro city, Brazil; it has 1,200.255 km² large, about 6.7 million residents and located in the southeastern region of country. The information of O₃, SO₂ and predictor variables were obtained from government agencies. The potential predictor variables have been used: temperature, relative humidity, vehicular traffic, Census, altitude, vegetation cover, land use, rock masses, hydrographic and hydrographic sub-basins, urban zoning and road network. Linear regression models were specified using the supervised stepwise procedure for the development of Land Use Regression models. Cook-D statistics were used to detect influential observations. The overall model performance was evaluated by leave-one-out cross validation (LOOCV). RESULTS: The annual average of O₃ and SO₂ was 117.68 (SD=±36.83) and 8.796 (SD=±4.93) μg·m−3, respectively. The final model for O₃ included four predictor variables: ADHHOLD_300, SQRALT, GREEN_5000 and DPOP_5000. The adjusted R² value was 0.83 and p-value=2.359E-05. The performance evaluated by LOOCV presented R²=0.74, RMSE=18.26 and MAE=14.26. For SO₂ included three predictor variables: SQRALT, INDUSTRY_100 and PORT_5000. The adjusted R² value was 0.91 and p-value=0.002. The performance showed R²=0.52, RMSE=3.60 and MAE=2.73. CONCLUSIONS: It was possible to elaborate a model applicable to areas where there is no air quality monitoring. The model allows an evaluation of the impact on the health of exposed populations, serving to support decision-making and the development of public policies and investments, in the medium and long term. KEYWORDS: Ozone, Sulfur dioxide, Air pollution.