Numerical simulations on the effect of traffic restriction on the air quality of Sao Paulo city – Brazil

Sao Paulo is the largest city in Brazil, with 11 millions inhabitants and more than 5 millions vehicles. Therefore, the major source of air pollutants are vehicles, that were responsible for the emission of 160000 tons of carbon monoxide (CO), 37700 tons of hydrocarbon (HC), 55300 tons of nitrogen oxides (NOx), 4300 tons of sulfur oxides (SOx), and 2000 tons of particulate matter (PM) only last year (2014), according to the Environmental Agency of the state of Sao Paulo (CETESB). The situation has been worse in the past, since the concentration of some pollutants have decreased due to the implementation of several public policies that targeted vehicular emissions. In 1986 the Vehicular Air Pollution Control Program (PROCONVE) was created by the National Environmental Council of Brazil (CONAMA), and this program established several phases along the following years with decreasing limits for emission factors of new vehicles. This program had a positive effect on the air quality of Brazilian cities, causing a decreasing tendency on the concentrations of many pollutants, particularly CO. Nevertheless, the number of vehicles in Sao Paulo is increasing and in 2013 presented an average growing rate of 2.7 %, as the pollutants concentration are presenting a stabilizing tendency lately. Locally, some policies were implemented to enhanced mobility and also had an impact on pollutants concentration. In the city of Sao Paulo, vehicles are restricted in a central zone of the city for one weekday, based on the last digit of the vehicle plate, during rush hours (from 0700 LT to 1000 LT and from 1700 LT to 2000 LT) since October 1997. Heavy-duty vehicles (trucks) are also restricted in the central zone and some highways from 0400 LT to 2200 LT or from 0500 LT to 2100 LT during weekdays and from 1000 LT to 1400 LT on Saturdays. An inspection program, the only air quality focused policy, was created in 2009 to condition the renovation of the vehicular license on the attainment of standard emissions; nevertheless this program was discontinued in 2014. Recently, a total of 100 km of bicycle paths and lanes and 440 km of dedicated bus lanes were implemented in order to decrease the use of private vehicles as transportation. These local policies, although mostly focused in mobility issues, may have helped changing the diurnal cycle and spatial distribution of pollutant concentrations, since the vehicles under restriction may have changed paths and hours to avoid restricted zones and periods. Therefore, the aim of this work is to simulate the influence of the local public policies in the diurnal evolution and spatial distribution of carbon monoxide, particulate matter and ozone in the city of Sao Paulo. The simulations will be conducted using first the Weather Research and Forecasting model (WRF) with the urban canopy parameterization to generate the atmospheric fields. Then, the Community Multi-scale Air Quality model (CMAQ) will use the atmospheric fields and emission information to simulate the concentrations of the pollutants. The simulations will then be compared to the observations performed by the Environmental Agency monitoring network. Two cases will be studied: one simulating the restriction zones and hours as implemented today and other simulating how the pollutants concentration field would be if these restrictions have not been implemented. The analysis of the observational network shows a change in the diurnal cycle of carbon monoxide and particulate matter from 1997 to 2012: CO diurnal cycle has 2 peaks during morning and evening rush hours, however recently the evening rush hour lasts longer then in 1997; PM diurnal cycle has 3 peaks, during morning and evening rush hours and around 0200 LT, and the morning rush hour was the highest peak in 1997, however now the morning rush hour is the lowest peak and the early morning peak is the highest. These changes may be related to a new behavior of the drivers that want to avoid restriction zones and hours. A tendency of homogenization of the spatial distribution of the pollutants has also been noted lately, and it may be caused by increasing urbanization of the peripheral areas of the city or by the change in vehicles routes to avoid restricted (central) zones. Therefore we hypothesize that the simulations will be able to determine the role of the restriction zones and hours on the diurnal cycle and spatial distribution of the vehicular pollutants, helping assess the effect of local policies on pollutants concentration in the city of Sao Paulo.