MESSAGE includes a detailed representation of a number of air pollutants including sulfur dioxide (SO2), nitrogen oxides (NOx), carbon monoxide (CO), volatile organic compounds (VOCs), black carbon (BC) and organic carbon (OC). Sectors included are power plants, industry (combustion and process), road transport, households, waste, agriculture, and large-scale biomass burning.
For all fossil fuel and biomass combustion in residential, industrial, transportation and electricity sectors, information from the GAINS model (Amann 2004, Amann et al. 2011) is used to represent various levels of air pollution legislation and control measures until 2030. The methodology is based on aggregation of GAINS related activity (e.g., various fuels) and technology (e.g., type of combustion technique, penetration of control technology) into MESSAGE fuels and sectors using emission factors (methodology described in (Riahi et al. 2011, Riahi et al. 2012). It should be the noted that the energy scenarios underlying the GAINS and MESSAGE models are independent, i.e., no attempt has been made to link the energy system activities in the two models. The linkages are only established at the level of emission abatement measures. Beyond 2030, pollutant emission factors in MESSAGE are scenario dependent and their evolution depends on assumptions on the continued implementation of air pollution controls, depending on continued economic growth and increase in per-capita income levels. However, emission factors are assumed to never decline below currently available information on the 'maximum feasible reduction' levels (MFR) possible in 2030 based on full implementation of all best available technology (BAT) measures as suggested by the GAINS model.
MESSAGE additionally includes assumptions on air pollution from waste, international shipping, agriculture (including agriculture waste burning), waste and large-scale biomass burning. Estimates for the year 2000 are based on (Granier et al. 2011) and future emissions are derived using a number of assumptions detailed in (Riahi et al. 2011). MESSAGE also provides spatial estimates of air pollutants using data described in Granier et al. (2011) and an exposure-driven algorithm for the downscaling of the regional air-pollutant emissions projections based on (Riahi et al. 2011). These have been used to estimate detailed global and regional health impacts of outdoor air pollution (Rao et al. 2012, Rao et al. 2013).