For CAMS NRT forecasts up to and including January 23rd 2017 (model cycle 41r1)
The formulae used are:
PM10 = RHO * ( 1 * SS1 + 1 * SS2 + 1 * DD1 + 1 * DD2 + 0.55 * DD3 + 0.85 * OM1 + 0.85 * OM2 + 0.85 * SU1 + 1* BC1 + 1 * BC2 )
PM2.5 = RHO * ( 1 * SS1 + 0.4 * SS2 + 1 * DD1 + 1 * DD2 + 0.11 * DD3 + 0.7 * OM1 + 0.7 * OM2 + O.7 * SU1 + 1 * BC1 + 1 * BC2 )
Where
RHO: air density (around 1.2 KG/M3 at altitude 0), which can be computed from pressure P and temperature T as RHO = P/R*T, R is the specific gas constant for dry air, 287.058 J/(kg·K).
SS1: seal salt (bin 1)
SS2: seal salt (bin 2)
DD1: desert dust (bin 1)
DD2: desert dust (bin 2)
DD3: desert dust (bin 3)
OM1: organic matter (bin 1)
OM2: organic matter (bin 2)
SU1: sulphur (bin 1)
BC1: black carbon (bin 1)
BC2: black carbon (bin 2)
Sea salt bins: [0.03-0.5], [0.5- 5], [5-20 microns] radius
Desert dust bins: [0.03-0.55], [0.55-0.9], [0.9-20 microns] radius
Notes:
- SS3 does not contribute to PM10 and PM2.5, on the basis that SS3 particles are >5µm in radius (10µm diameter), thus too large.
- The above formulae are subject to an overestimation of the sea salt contribution to PM by a factor of 4.3 as described here. This mostly affects PM in cleaner (especially coastal) environments with background aerosol, rather than the PM peaks in polluted regions, which are probably of most concern to users.
See also: Morcrette, J.-J., and Coauthors, 2009: Aerosol analysis and forecast in the European Centre for Medium-Range Weather Forecasts Integrated Forecast System Forward modeling, J. Geo. Res., 114, D06206, doi:10.1029/2008JD011235
For CAMS NRT forecasts from January 24th 2017 (model cycle 43r1) and the CAMS Reanalysis (based on model cycle 42r1):
The formulae used are:
PM10 = RHO * ( 1 * SS1 / 4.3 + SS2 / 4.3 + 1 * DD1 + 1 * DD2 + 0.4 * DD3 + 1 * OM1 + 1 * OM2 + 1 * SU1 + 1 * BC1 + 1 * BC2 )
PM2.5 = RHO * ( 1 * SS1 / 4.3 + 0.5 * SS2 / 4.3 + 1 * DD1 + 1 * DD2 + 0.7 * OM1 + 0.7 * OM2 + O.7 * SU1 + 1 * BC1 + 1 * BC2 )
Where
RHO: air density (around 1.2 KG/M3 at altitude 0), which can be computed from pressure P and temperature T as RHO = P/R*T, R is the specific gas constant for dry air, 287.058 J/(kg·K).
SS1: seal salt (bin 1)
SS2: seal salt (bin 2)
DD1: desert dust (bin 1)
DD2: desert dust (bin 2)
DD3: desert dust (bin 3)
OM1: organic matter (bin 1)
OM2: organic matter (bin 2)
SU1: sulphur (bin 1)
BC1: black carbon (bin 1)
BC2: black carbon (bin 2)
Sea salt bins: [0.03-0.5], [0.5- 5], [5- 20 microns] radius
Desert dust bins: [0.03-0.55], [0.55-0.9], [0.9-20 microns] radius
Notes:
- SS3 does not contribute to PM10 and PM2.5, on the basis that SS3 particles are >5µm in radius (10µm diameter), thus too large. This may need revisiting in future as a result of correcting for the 80% humidity issue, which means the particles now look smaller than before.
- DD3 (large dust particles) does not contribute PM2.5.
For CAMS NRT forecasts from July 9th 2019 (model cycle 46r1) with additional Nitrate and Ammonium aerosols
The formulae used are:
PM10 = RHO * ( 1 * SS1 / 4.3 + SS2 / 4.3 + 1 * DD1 + 1 * DD2 + 0.4 * DD3 + 1 * OM1 + 1 * OM2 + 1 * SU1 + 1 * BC1 + 1 * BC2 + 1 * NI1 + 1 * NI2 + 1 * NH3 )
PM2.5 = RHO * ( 1 * SS1 / 4.3 + 0.5 * SS2 / 4.3 + 1 * DD1 + 1 * DD2 + 0.7 * OM1 + 0.7 * OM2 + O.7 * SU1 + 1 * BC1 + 1 * BC2 + 0.7 * NI1 + 0.25 * NI2 + 0.7 * NH3 )
RHO: air density (around 1.2 KG/M3 at altitude 0), which can be computed from pressure P and temperature T as RHO = P/R*T, R is the specific gas constant for dry air, 287.058 J/(kg·K).
SS1: seal salt (bin 1)
SS2: seal salt (bin 2)
DD1: desert dust (bin 1)
DD2: desert dust (bin 2)
DD3: desert dust (bin 3)
OM1: organic matter (bin 1)
OM2: organic matter (bin 2)
SU1: sulphur (bin 1)
BC1: black carbon (bin 1)
BC2: black carbon (bin 2)
NI1: Nitrates fine (bin 1)
NI2: Nitrates coarse (bin 2)
NH3: Ammonium
Sea salt bins: [0.03-0.5], [0.5- 5], [5- 20 microns] radius
Desert dust bins: [0.03-0.55], [0.55-0.9], [0.9-20 mic