Calculation of PEC regional

In this section, the following parameters are derived:

- regional exposure concentrations in all environmental compartments

Regional computations are done by means of multimedia fate models based on the fugacity concept. Recently, models have been described by Mackay et al., (1992) and by Van de Meent, (1993) (SimpleBox).

These models are box models, consisting of a number of compartments (see Figure 11) w h i c h a r e c o n s i d e r e d homogeneous and well mixed. A chemical released into the model is distributed between the compartments according to the

properties of both the chemical and the model environment. Several types of fate processes are distinguished in the regional assessment, as drawn in Figure 11:

• Emission, direct and indirect (via STP) to the compartments air, water, industrial soil, and agricultural soil;

• Degradation, biotic and abiotic degradation processes in all compartments;

• Diffusive transport, as e.g. gas absorption and volatilisation. Diffusive mass transfer between two compartments goes both ways, the net flow may be either way, depending on the concentration in both compartments;

• Advective transport, as e.g. deposition, run-off, erosion. In the case of advective transport, a chemical is carried from one compartment into another by a carrier that physically flows from one compartment into the other. Therefore, advective transport is strictly one-way.

Chemical input to the model is regarded as continuous and equivalent to continuous diffuse emission. The results from the model are steady-state concentrations, which can be regarded as estimates of long term average exposure levels. The fact that a steady-state between the compartments is calculated, does not imply that the compartment to which the emission takes place is of no importance.

Figure 11 The relevant emission and distribution routes

In a Mackay-type level III model, the distribution and absolute concentrations may highly depend upon the compartment of entry.

Advective import and export (defined as inflow from outside the model or outflow from the model environment) can be very important for the outcome of both regional and local model calculations. Therefore, the concentration of a chemical at the "border" of the region must be taken into account. This is defined as the background concentration of a chemical. The background concentration in a local model can be obtained from the outcome of the regional model. For chemicals with many relatively small point sources, this background concentration may represent a significant addition to the concentration from a local source.

The background concentration in the regional model has to be calculated using a similar box model of a larger scale, e.g. with the size of the European continent. In this continental model, however, it is assumed that no inflow of air and water across the boundaries occurs.

Furthermore it is assumed that all chemical releases enter into this continental environment.

The resulting steady-state concentrations are then used as transboundary or background concentrations in the regional model. The continental and regional computations should thus be done in sequence. Figure 1 visualises the relationship between the concentrations calculated for the different model scales. For both the regional and continental scale, the total emission amounts (through diffuse and point sources, summed over all stages of the life-cycle) are used.

For the PECregional calculation, in contrast to PEClocal, an average percentage connection rate to STPs should be included in the calculation. This leads to a more realistic estimation of the likely background concentration on a regional scale. For the purposes of the generic regional model, a STP connection rate of 70% (the EU average according to Appendix XII) will be assumed.

The results from the regional model should be interpreted with caution. The environmental concentrations are averages for the entire regional compartments (which were assumed well mixed). Locally, concentrations may be much higher than these average values. Furthermore, there is a considerable degree of uncertainty due to the uncertainty in the determination of input parameters (e.g. degradation rates, partitioning coefficients).

Model parameters for PECregional

When calculating the PECregional it is important which modelling parameters are chosen and what fraction of the total emission is used as emission for the region. There are two different possibilities:

• Calculation of a PECregional on the basis of a standardised regional environment with agreed model parameters;

• Calculation of a PECregional on the basis of country specific model parameters.

A standardised regional environment should be used for the first approach in the calculation of PECregional. When more specific information is available on the location of production/emission sites, this information can be applied to refine the regional assessment.

The second approach may sometimes result in a better estimation of the concentrations for a specific country. However, depending on the information on production site location, it will lead to a number of different PEC values which makes a risk characterisation at EU level more complicated.

Calculations are performed for a densely populated area of 200 x 200 km with 20 million inhabitants. Unless specific information on use or emission per capita is available, it is assumed that 10 % of the European production and use takes place within this area, i.e. 10%

of the estimated emission is used as input for the region. The model parameters proposed for this standard region are given in Table 10. It should be noted that it is extremely difficult to select typical or representative values for a standard European region. Therefore, the rationale behind the values of Table 10 is limited. Nevertheless, these values present a starting point for the regional scale assessments. Characterisation of the environmental compartments for the regional model should be done according to the values in Table 3.

Table 10 Proposed model parameters for regional model

Parameter Value in regional model

area of the regional system 4⋅10⁴ km²

area fraction of water 0.03

area fraction of natural soil 0.60

area fraction of agricultural soil 0.27

area fraction of industrial/urban soil 0.10

mixing depth of natural soil 0.05 m

mixing depth of agricultural soil 0.2 m

mixing depth of industrial/urban soil 0.05 m

atmospheric mixing height 1000 m

depth of water 3 m

depth of sediment 0.03 m

fraction of the sediment compartment that is aerobic 0.10

average annual precipitation 700 mm.yr^-1

wind speed 3 m.s^-1

residence time of air 0.7 d

residence time of water 40 d

fraction of rain water infiltrating soil 0.25 fraction of rain water running off soil 0.25 EU average connection percentage to STP 70%

The area fractions for water and for natural, agricultural and industrial/urban soils, are average values obtained from ECETOC (1994b), supplemented with data received from Sweden and Finland. Data for Norway and Austria are obtained from the FAO (Agrostat data-base). The residence time for air (defined as the time between air entering and leaving the region) of 0.7 days is derived from the wind speed of 3 m/s and the area of the region. The residence time of water of 40 days is selected as a reasonable average for the European situation. The flow of water through the system is the sum of the amount of rain (run off and directly into surface water), effluent discharges, and inflow of rivers.

Given the average annual rainfall of 700 mm and a runoff fraction of 0.25, the resulting flow of water through the model environment necessary to obtain this residence time is 6.9⋅10⁷ m³.d^-1.

The amount of waste water discharged, is the product of the amount of waste water discharged per inhabitant equivalent and the number of inhabitants of the system. Using a flow per capita of 200 l.d^-1 (equivalent to the value used in the SimpleTreat model) and a population of 20 million, this results in an additional water flow through the model environment of 4.0⋅10⁶ m³.d^-1. Therefore, the remaining inflow, caused by inflowing riverwater, is 6.5⋅10⁷ m³.d^-1.

In addition to the environmental characteristics of the region, selected intermedia mass transfer coefficients are required in the multimedia fugacity model to ensure comparability of the outcome with other models. These transfer coefficients are summarised in Table 11.

Table 11 Intermedia mass transfer coefficients

Parameter Value

air-water interface: air side partial mass transfer coefficient 1.39 ⋅ 10^-3 m.s^-1 air-water interface: water side partial mass transfer coefficient 1.39 ⋅ 10^-5 m.s^-1

aerosol deposition rate 0.001 m.s^-1

air-soil interface: air side partial mass transfer coefficient 1.39 ⋅ 10^-3 m.s^-1 air-soil interface: soilair side partial mass transfer coefficient 5.56 ⋅ 10^-6 m.s^-1 air-soil interface: soilwater side partial mass transfer coefficient 5.56 ⋅ 10^-10 m.s^-1 sediment-water interface: water side partial mass transfer coefficient 2.78 ⋅ 10^-6 m.s^-1 sediment-water interface: pore water side partial mass transfer coefficient 2.78 ⋅ 10^-8 m.s^-1

net sedimentation rate 3 mm.yr^-1

Model parameters for the continental concentration

The continental box has the size of all EU countries together (Norway included) and similar percentages for water and natural, agricultural and industrial/urban soils as given in Table 10.

These values are summarised in Table 12. All other parameters are similar to the ones given in the preceding tables. Emission estimation to this continental box should be based on the EU-wide production volume of the chemical. The resulting concentrations in water and air must be used as background concentrations (i.e. concentrations in water or air that enter the system) in the regional model. It is assumed that no inflow of the substance into the continental system takes place.

Table 12 Parameters for continental model

Parameter Value in continental model

area of the continental system 3.56⋅ 10⁶ km²

area fraction of water 0.03

area fraction of natural soil 0.60 area fraction of agricultural soil 0.27 area fraction of industrial/urban soil 0.10