in the framework of the EU Integrated Project

Global and regional Earth-system Monitoring using satellite and in-situ data (GEMS)


The forecast system runs operationally in order to predict air pollutants concentrations (O3, NO, NO2, CO, SO2, PM10 and Rn). A 3-day air quality forecast is issued on a daily basis, using a modeling system that consists of the Fifth Generation PSU/NCAR Mesoscale Model (MM5 version 3.7) and the photochemical air quality model Comprehensive Air quality Model with extensions (CAMx version 4.40). Daily mean and maximum pollutants concentrations are mapped for Europe, the Balkan Peninsula and Athens for 4 vertical levels: 1) Surface, 2) 500m, 3) 1000m and 4) 3000m. The air quality forecast maps are available on the web around 07:00 UTC.

Figure 1 depicts the flow chart of the forecast system.

Fig.1: The forecast system flow chart.

Modeling Domains

MM5 is implemented on three nested grids that share the same Lambert Conformal Conic projection (Fig. 2). The first, coarse grid covers Europe and has spatial resolution equal to 30-km. The second and third grids have higher spatial resolution (10-km and 2-km) and focus on the Balkan Peninsula and the Greater Athens Area respectively. All grids have a vertical structure of 33 -levels, reaching at approximately 100 mbar height.

CAMx is also implemented on three nested grids, which have the same projection and spatial resolution with the corresponding MM5 grids. CAMx grids have 15 vertical layers extending up to approximately 7 km above ground level. The vertical layers are unevenly distributed with higher resolution at the near-surface layers. The first layer height is about 20 m.

Fig.2: Modeling domains of the forecast system.


The forecast system is driven by the meteorology provided by the MM5 model. MM5 is a limited area, non-hydrostatic, terrain following and sigma-coordinate model, designed to simulate and/or predict atmospheric circulation. The model consists of several pre- and post-processing programs which are referred to collectively as the MM5 modeling system.

The model develops initial and boundary conditions (ICBC) based on the global 12:00 UTC forecast of GFS/NCEP of 1-degree spatial resolution. The forecast is transferred to the local infrastructure via an anonymous ftp client server and becomes available around 16:00 UTC. Interpolation of the ICBC onto the grid of the forecast system is carried out by the REGRID pre-processing program of MM5.


CAMx is an Eulerian, photochemical air quality model that simulates the emission, dispersion, chemical reactions and removal of pollutants in the troposphere, and it is thus extensively used in order to assess gaseous and particulate air pollution.

The MM5toCAMx interface is implemented for linking the meteorological driver with the air quality model. The Carbon Bond (CB-IV) mechanism is implemented for solving chemical kinetics. The TUV radiative transfer and photolysis model is used as a CAMx preprocessor to provide the air quality model with a multi-dimensional lookup table of photolytic rates.

The anthropogenic emission data, used as CAMx input data, are gridded emission rates of gaseous pollutants (NOx, SO2, NMVOCs, CH4, NH3, CO) and particulate matter (PM10). Two emission inventories were compiled, one for Greece (10-km resolution) and one for the Greater Athens Area (2-km resolution). The development of the emission inventories was based on the methodologies of the EMEP/CORINAIR emissions inventory guidebook. All anthropogenic emission sources (industry, central heating, road transport, waste etc.) were taken into account in the inventories. The emission data for all European countries other that Greece were provided by The Netherlands Organization in 0.125ox0.063o resolution. The temporal disaggregation of the emissions was based on temporal profiles (seasonal, weekly and diurnal) developed by the Institute of Energy Economics and the Rational Use of Energy (IER) of the University of Stuttgart under the EUROTRAC project (subproject GENEMIS). Example maps of the anthropogenic emissions are presented in Figure 3.





Fig. 3: a) Hourly NOx emissions in Europe (September, Monday, 09:00 UTC) and b) Annual anthropogenic PM10 emissions in Greece.

A biogenic emission model was developed and integrated in the forecast system. The model is implemented for the calculation of isoprene, monoterpenes and OVOCs emissions using the methodology described in Guenther et al., (JGR, 1995). The model processes different input data to estimate gridded biogenic NMVOCs fluxes on an hourly basis as: a) land use data, b) land-use-specific foliar biomass densities and emission potentials and c) the MM5 forecasted temperature and solar radiation data. Figure 4 shows calculated biogenic NMVOCs emissions for the Greater Athens Area.

Fig. 4: Isoprene (left) and monoterpenes (right) hourly emission rates in the Greater Athens Area on the 2nd July 2007.


 ..:: Return ::..