A DIMENSÃO SOCIAL DAS EMISSÕES DE CO2 DO TRÁFEGO DE VEÍCULOS: modelos inteligentes de alta definição na cidade do Rio de Janeiro.

CO2 emissions from urban areas have been perceived as a central point in the Global Carbon Cycle, although we still understand little about their social dimension in space and time. The work aims to model CO2 emissions from road traffic and evaluate the relationship between socio-economic characteristics of the population living in the city of Rio de Janeiro between 7 and 31 January 2021. A mapping model was used to estimate CO2 emissions at the street level, using Machine Learning algorithms, mobility data, Geographic Information System, and R. Emissions were aggregated to planning areas and regions, as well as to the IPEA digital mesh containing socioeconomic indicators, such as population density, average per capita household income and number of jobs. The results showed that the ML model effectively represents the spatio-temporal variability of hourly vehicle flows and average speed at road level in the city of Rio de Janeiro, presenting an RMSE of 138 veh h-1 and 7.0 km h-1, respectively. The model estimated high emissions on Avenida Brasil, the city's main expressway, with a maximum of 4,750 tCO2 km-2. The biggest emissions occur on the expressways that connect the city towards the Center and the edge of the South Zone and Barra da Tijuca, old and new centralities, where there is a greater job offer. [ABSTRACT FROM AUTHOR]