Methodology

Our study analyses the entire system by which resource inputs to the industrial economy are converted into outputs that must be absorbed and processed by the environment. For analysing a given chemical, this systems approach can be divided into three steps:

1. Identification of materials in which the chemical is embodied, and the pathways by which they flow through the industrial economy.
2. Estimation of emissions and deposition to air, water, and soil for each material at each stage of its life cycle.
3. Construction of a basin-wide pollution model for assessment of proposed emission reduction policies, environmental impacts, or other relevant issues related to the chemical in question.

Fig. 1 The Rhine Basin. Place-names in boxes signify locations of monitoring stations of the International Commission for the Protection of the Rhine

In step 1, it is essential not to miss any important source of pollution. In this regard, it should be noted that many chemicals enter the industrial economy inadvertently as trace impurities of high-volume raw materials such as fossil fuels and iron and non-ferrous ores. Moreover, a full accounting should be made of all stages of the material's life cycle. These include not only the stage of production, but also the later stages of use and disposal.

Overlooking important sources of emissions can be costly. For example, Tschinkel (1989) has noted that billions of dollars have been spent in the US on the construction of secondary sewage treatment plants. Many of the benefits gained from this technology, however, have been nullified because discharges of untreated storm waters containing toxic urban street dust continue to flow into lakes, rivers, and estuaries. Such an omission may not have happened had planners been more aware of the significance of street dust as a major source of toxic materials.

In step 2, emissions and deposition are estimated quantitatively. Emissions may be classified broadly into two categories: point source and diffuse.. Point sources include electric power plants, industry, incinerators, sewage treatment plants, and others. Their emissions are typically highly concentrated and confined to a specific location, usually within an urban area. For each type of point source, emission factors, generally expressed as weight of pollutant per unit weight of material consumed or produced, are assigned for emissions to air, water, and land. Emission factors may change over time, decreasing as cleaner technologies are implemented. Total emissions are calculated as the product of the emission factor and the weight of material consumed or produced.

Particularly in the case of atmospheric pollution, it is important to make a distinction between emissions and deposition (or immissions, as it is called in other languages). Via the mechanism of long-range atmospheric transport, emissions may be deposited hundreds or even thousands of kilometres from their sources. Thus, some emissions generated in the basin are transported and deposited outside the basin, and some emissions from outside the basin are transported into it. A long-range atmospheric deposition model has been developed at IIASA (Alcamo et al., 1992) for estimating deposition in the basin.

In contrast to point-source emissions, diffuse emissions are generally less concentrated, more dispersed spatially, and dependent on land use, which can be broadly categorized as forests, agricultural lands (both arable and grassland), and urban areas. The only inputs to forested lands are assumed to be atmospheric deposition via long-range transport. Chemical inputs to agricultural soils include not only longrange atmospheric deposition, but also agrochemicals, manure, and sewage sludge. Diffuse emissions from these two land uses are determined using a runoff export model (Jolankai et al., 1991).

Transport of pollutants to surface and ground waters is much greater from agricultural lands than from forested areas. Transport occurs via storm runoff, erosion, and vertical seepage. The relevant parameters to be determined are the rates of applications of particular chemicals, expressed as weight per hectare, and the partition coefficient, which determines the fraction of chemical that is mobilized and transported and the fraction that remains bound in the soil. Even if only a small percentage of the chemicals is mobilized, total emissions can be significant because of the enormous chemical inputs and the large spatial coverage of agricultural lands.

Another important source of diffuse emissions is transport of pollutants from paved urban areas to surface waters. This occurs by the build-up of toxic materials in street dust during dry periods, and the washing out of the dust during storm events. The pathways by which the transport may occur are shown in figure 2. There are three main sources of toxic materials in urban dust: corrosion of building materials (particularly for heavy metals such as zinc, used in galvanizing and surface materials), exhausts and lyre wear from automobiles and other road vehicles (important for lead and zinc), and local and longrange atmospheric deposition (a dominant source of cadmium).

When storm sewers are separate from municipal sewers (path SSS in figure 2), the pollutants are transported directly to surface waters. Alternatively, storm sewers may be connected to municipal sewers that discharge to surface waters without treatment (path CSSW in figure 2), or they may be connected to municipal sewers in which the effluents are treated (path WWTP in figure 2). In sewage treatment plants with primary and secondary treatment, typically 50 per cent or more of input heavy metals are trapped in sewage sludges. Even when the storm sewers are connected to sewage treatment plants, however, some fraction of the polluted street dust may be transported to the river unabated if the volume of storm flow exceeds the flow capacity of the sewage treatment plant (path CSO in figure 2), which is often the case. Another important source of pollution, also indicated in the figure, is the atmospheric deposition on unpaved urban areas, with subsequent seepage to ground waters and transport to the river.

Fig. 2 Pathways by which pollutants in urban areas are transported to surface waters (Source: Behrendt and Boehme, 1992)

To calculate emissions from corrosion, it is necessary to determine rates of corrosion per unit surface of the corroded material and the total surface coverage of the material in question. For instance, rates of zinc corrosion are strongly linked to urban SO₂ concentrations and will decrease over time as SO₂ levels are lowered. The following equation (ECE, 1984) shows the empirical relationship between SO₂ concentration and the rate of zinc corrosion from galvanized steel:

Y = 0.45* [SO₂] + 0.7.

where Y = annual corrosion rate of galvanized steel (g/m²/yr), and [SO₂] = concentration of SO₂ in air (mg/m³)

Emissions from road traffic owing to tyre wear may be estimated by determining emission rates per vehicle km, and multiplying this rate by vehicle km per year and the number of vehicles per year.

Lead emissions from combustion of gasoline may be calculated by multiplying lead emitted per unit of gasoline burned and multiplying by annual gasoline consumption. The emissions are allocated spatially by apportioning them according to traffic density.

In urban areas, local atmospheric emissions are particularly important, since a significant fraction, typically around 10 per cent of the total emissions, are deposited within 10 to 20 kilometres of the source. Factors affecting the proportion of local to long-range emissions include smokestack height, velocity of the gases and particulate matter leaving the stack, meteorological conditions, and particle size of emitted pollutants. The IIASA study includes an analysis of trends in local emissions as affected by changes in the abovementioned factors (Hrehoruk et al., 1992).

Lastly, it is necessary to employ an urban hydrology model that estimates the fraction of street dust that flows to the river. Even though urban and suburban areas occupy only about 15 per cent of total surface area in the basin, their contribution to the total diffuse load of aqueous emissions is significantly higher. This is because of the prevalence of hard, impermeable surfaces in urban lands (typically around 33 per cent of total urban area), from which run-off and transport of pollutants can be as high as 90 per cent, compared to a maximum of about 25 per cent for agricultural lands (Ayres and Rod, 1986).

Completion of steps 1 and 2 for a given chemical provides a pollution model of the basin, including inputs and outputs for the chemical, its flows through the industrial economy over time, and its spatial allocation for each time period of interest. The model can be used for various purposes. For example, a historical analysis of pollution can provide information on changing trends in pollution sources. In the case of pollution in the River Rhine, the IIASA analysis indicates that since the mid-1970s diffuse sources of emissions of heavy metals have become increasingly important relative to point sources. Another useful application of the historical analysis is the possibility for estimating the cumulative build-up of toxic materials in soils and sediments. Currently, hardly any information exists on the rates of accumulation of toxic chemicals over wide spatial regions, or on the evaluation of resulting impacts to the environment and human health. (For a comprehensive discussion of cumulative chemical loading and potential environmental impacts see Stigliani, 1988, and Stigliani et al., 1991.)

The pollution model can also be used to test the effectiveness of proposed policy options for reducing emissions of toxic chemicals. Because the model is based on mass balance analysis, all material flows to air, water, and land within the basin must be taken into account. The model will thus expose options that would not reduce overall emissions in the basin, but, rather, would transfer them from one pollution pathway to another.

Continue