Abstract
Brazil's significantly large fleet of ethanol-powered vehicles makes it unique with regard to mobile source emissions. An accurate estimation of emission rates is best obtained from studies reflecting real-world vehicular conditions, such as tunnel investigations. Although laboratory-based estimations are standardized, they may not reflect reality. This study evaluates two methods for estimating emission factors that arise from the circulating fleet of vehicles that travel through the Rebouças Tunnel in Rio de Janeiro. The two monitoring points were located in the L1 gallery of the tunnel. Thirteen sampling campaigns took place between 2017 and 2020. The Pearson correlation method yielded emission rates for total aldehydes of 22.1 ± 6.9 mg km− 1, while the fuel consumption testing method resulted in rates of 17.7 ± 7.8 mg km− 1. Acetaldehyde predominated in both methods, as it is a major constituent of the Brazilian fuel matrix. Benzene emerged as the most abundant BTEX compound, with 3.7 ± 3.2 mg km− 1, when the Pearson method was used, and 2.5 ± 0.3 mg km− 1 with the fuel consumption testing method. Naphthalene, which was mainly collected in the gas phase, showed rates of 1.4 ± 1.0 mg km− 1 (Pearson) and 0.9 ± 0.4 mg km− 1 (fuel consumption). Despite these methodological variations, our findings converge, which suggests that the choice of method should depend on available resources. The Pearson method, which requires a vehicle counting method for traffic flow, obviates the need for CO and CO2 concentration measurements, that are essential for determining fuel consumption.