Assessment of on-road transportation demand and CO2 emissions for determination of air quality impacts from the Megacity of São Paulo

In this manuscript we assess the capability of using mobility surveys and a high-scale assignment and emission model to study climate change and air quality impacts related to on-road transportation in the Megacity of São Paulo (MSP). Initially, we estimate CO2 emissions of light and heavy vehicles (LVs and HVs) at a spatial scale of 500m and temporal scale of an hour, using transport demand modeling. The estimates are based on origin and destination trip pairs and the height of the planetary boundary layer (PBL). These estimates, performed for the years 2007 and 2012, depend also on intermediate variables as dilution rates (D) and surface particulate-matter concentrations (PM). Secondly, we assess the changes in CO2 vehicle emissions from the MRSP over the period 2007-2012 (4% year-1). Consequently, CO2 emission inventories merge trip-based surveys, traffic assignments and road network database with air pollution monitoring data. Despite the difference of the methodologies, we use a road link bottom up vehicle activity based approach, the assessed emissions agree with the State's Emission Inventory. This paper shows that the CO2 emissions from LDVs and HDVs in the MSP in 2007 and 2012 were 8,477 and 10,075 tCeq day-1 (58% LVs and 42% HVs), respectively. CO2 emissions from vehicles show spatial patterns consistent with passenger and freight transport trips and road network assignments. Temporal profiles (diurnal, weekly and monthly) were estimated using traffic counts and congestion surrogates. The profiles were compared with average road-site (Western of MSP) and background (Jaraguá Peak) CO2 measurements available for 2014. On-road measurements showed one peak associated to the morning peak hour of vehicles (437±45 ppm) and another night peak (435±49 ppm) related to the low PBL (313 m) and D (329 m2 h-1). From on-road measurements, background values (414±2 ppm) were subtracted to estimate excess CO2 (12±8 ppm) directly attributed to vehicles. The inventory reflects the relationships between traffic patterns and emissions, and the developed methodology could be used to evaluate the impacts of forthcoming urban transport and emission control policies. In the future, our estimates will be verified with ground measurements of CO2 concentrations over a bigger monitoring network in the MSP.