1. Initial and Boundary Conditions

1.1 Initial conditions

There are two ways to initialise the concentrations at the start of a simulation. The first is to use data from a previous calculation by reading the data from a restart file. The other method is simply an interpolation of the boundary conditions specified for the first hour of the simulation. The boundary conditions used for the latter are described below.

Because normally LOTOS-EUROS model runs are performed over a whole year on an hour-by-hour basis, initial conditions have to be specified only on the first hour of January 1. The impact of the initial conditions will gradually disappear, and be no longer important after say 5 days of modelcalculations.

1.2 Boundary conditions

1.2.1 Logan in combination with the EMEP-method for ozone, aerosols and their precursors

Ozone is the gas where specification of accurate boundary conditions is most essential for a good model performance. This is due to the fact that ambient ozone levels in Europe are typically not much greater than the Northern hemispheric background ozone. In LOTOS-EUROS we use the 3‑D climatological dataset by Logan (1998), derived from ozone sonde data, or 3‑D datasets from global models (e.g TM3/5) for all boundaries (incl. top). By default we use the data set by Logan (1998) as global model results are not available for all years.

For a number of components, listed in Table 10.1 we follow the EMEP method (Simpson et al., 2003) based on measured data. In this method simple functions have been derived to match the observed distributions. The boundary conditions are adjusted as function of height, latitude and day of the year. The functions are used to set the boundary conditions, both at the lateral boundaries as at the model top. The annual cycle of each species is represented with a cosine-curve, using the annual mean near-surface concentration, C₀, the amplitude of the cycle DC, and the day of the year at which the maximum value occurs, d_max. Table 10.1 lists these parameters.

We first calculate the seasonal changes in ground-level boundary condition, C₀, through:

where n_y is the number of days per year, d_mm is the day number of mid-month (assumed to be the 15th), and d_max is day number at which C₀ maximises, as given in Table 10.1. Changes in the vertical are specified with a scale-height, H_z, also given in Table 10.1.

where C_i(h) is the concentration at height h (in km). For simplicity we set h to be the height of the centre of each model layer assuming a standard atmosphere. For some species a latitude factor, given in Table 10.2, is also applied. Values of C_i adjusted in this manner are constrained to be greater or equal to the minimum values, C_min, given in Table 10.1.

Ammonia boundary conditions are neglected. Sulphate is assumed to be fully neutralised by ammonium. Nitrate values are assumed to be included in those of nitric acid and are zero as well.

Table 10.1 Parameters used to set the boundary conditions

Parameter	C_mean	d_max	DC	Hz
	ppb	days	ppb	km	ppb	ppb
SO₂	0.15	15.0	0.05	∞	0.15	0.03
SO₄	0.15	180.0	0.00	1.6	0.05	0.03
NO	0.1	15.0	0.03	4.0	0.03	0.02
NO₂	0.1	15.0	0.03	4.0	0.05	0.04
PAN	0.20	120.0	0.15	∞	0.20	0.1
HNO₃	0.1	15.0	0.03	∞	0.05	0.05
CO	125.0	75.0	35.0	25.0	70.0	30.0
ETH	2.0	75.0	1.0	10.0	0.05	0.05
FORM	0.7	180.0	0.3	6.0	0.05	0.05
ACET	2.0	180.0	0.5	6.0	0.05	0.05

Table 10.2 Latitude factors applied to the prescribed boundary conditions

Component	Latitude (^oN)
	35	40	45	50	55	60	65	70
SO₂, SO₄, NO, NO₂	0.15	0.3	0.8	1.0	0.6	0.2	0.12	0.05
HNO3, FORM, ACET	1.0	1.0	0.85	0.7	0.55	0.4	0.3	0.2
PAN	0.33	0.5	0.8	1.0	0.75	0.5	0.3	0.1
CO	0.7	0.8	0.9	1.0	1.0	0.95	0.85	0.8

1.2.2 TM3/TM5 boundary conditions

For the meteorological year 1997 there is the option in LOTOS-EUROS to work with boundary conditions provided by the TM3 model. It is anticipated that in the future, boundary conditions for 1997 and for other meteorological years will become available provided by the TM5-model. The exchange between TM3 and LOTOS-EUROS is arranged by updating the boundary concentrations every 6 hours. So, the average concentrations of 28 species in the TM3 model over 6 hours are used.

The TM3 model is a global model with a vertical structure in which the height of the layers varies as a function of pressure. Since the vertical structure of LOTOS-EUROS does not match with the vertical structure of TM3 the concentrations of the TM3 species at the different levels must be redistributed over the adjacent levels of LOTOS-EUROS. In order to save time for each of the columns in the TM3 grid the vertical structure is fixed as a monthly average. In other words: the concentrations vary every six hours, but the vertical distribution of the levels varies only month by month.

The TM3 model has a 8°x10° horizontal resolution. The anthropogenic emissions are from the EDGAR/GEIA data base and they represent the emissions of the year 1997.

The methane concentrations in this TM3 model have the tendency to slightly underestimate the measured methane. For instance, comparing to Mace Head the monthly means of methane are about 50 ppb lower as compared with the measured methane in the summer, although the underestimation amounts to just 10-20 ppb in the winter.

For ozone the concentrations (on a monthly basis) compared quite well with the monitoring data at the western edge of the LOTOS-EUROS domain. For the south-eastern corner (Middle-East region) the TM3 model produced quite high ozone values. Due to lack of sufficient monitoring data it is hard to appreciate these values.

XXX missing: iets over top boundary conditions; iets over bc_steady ??