Pollutants build up on highways and are washed off during a rainfall event and are usually discharged via an outfall to a watercourse. CIRIA report 142 (1994) guidance indicates that pollutant build-up on highways is a linear process which is proportional to traffic flow and time. This guidance has since been adopted in the Design Manual for Roads and Bridges. However, recent studies suggest that these assumptions may not be ideal. This article explores the pollutant build-up and runoff mechanisms on highways and suggests possible alternative assumptions which may be implemented to enhance the original CIRIA methodology.
1. SOURCES OF POLLUTANT BUILD-UP
Vehicles have been cited as the major contributors to chronic highway pollutant build-up in non-urban areas. Tyre, brake and engine wear are a source of zinc, cadmium, iron, chromium, copper and aluminium. Engine leakage (and un-combusted fuel) is the major source of hydrocarbons. Highway installations such as safety barriers have also been cited as a source of pollutants in highway runoff. Atmospheric deposition can contribute a significant amount to the pollutant load on highways; this can occur as dust falls during dry periods. This type of deposition has been shown to have a strong seasonal and industry-related component. A number of highway maintenance practices also contribute to the pollutant load in highway runoff. De-icing salt is predominantly sodium chloride; however, this usually has associated chemicals including iron, nickel, lead, zinc, chromium and cyanide. Vegetation control on road-side verges contributes to organic loading and other more dangerous pollutants if herbicides have been used. Construction works on highways are generally associated with the production of dissolved and particulate solids. In addition the use of heavy plant can introduce hydrocarbons in highway runoff. It should be noted that temporally these pollutant build-up mechanisms can be chronic, acute and seasonal.
2. FACTORS AFFECTING POLLUTANT BUILD-UP
Conventionally pollutant loading has been thought to be exclusively related to traffic flow. Several authors have since reported findings which show weak or insignificant correlation between traffic flow and pollutant build-up. Recently Kayhanian et al.1 in a comprehensive four-year study based in California have unequivocally concluded that ‘No simple linear relationship exists between highway runoff pollutants. . .and AADT’ (annual average daily traffic), including determinands which are known to be associated with vehicle deposition.
Local land use (including industry), seasonal variations, road surface material, embankments and cuttings have all been identified as important factors in influencing pollutant buildup. This has led to a categorisation of sites under investigation according to these factors in a number of studies.
3. RECOMMENDATIONS
3.1. Pollutant loading – relationship with traffic flow
CIRIA report 1422 identifies traffic flow as the major factor in pollutant loading build-up by directly relating traffic flow to pollutant build-up rates, the effect of which is capped for traffic flows greater than 30 000 vehicles per day (annual average).
The relationship between traffic flow and pollutant loading is at most a tenuous one. If pollutant loading rates are dependent on a number of unrelated factors then a method is required to account for these different factors. Multiple linear regression models have been suggested3 to understand the factors which influence pollutant build-up. This method allows the identification of a number of ‘casual’ variables which can affect pollutant build-up. Another approach would be to consider land use categories which have been shown to be major factor in pollutant build-up; again these could be integrated with traffic flow data or more radically traffic flow could be neglected altogether. If these approaches appear too great a departure from the CIRIA methodology then a more subtle approach could be implemented by introducing a background value which accounts for ‘other’ non-traffic flow-related factors.
3.2. Pollutant build-up – temporal variation
The CIRIA guidance indicates a linear build-up of pollutants on highways with time, by implementing pollutant loading rates (kg/ha per year). This offers a simple and effective way of understanding the temporal build-up of pollutants. However, this assumption leads to the fact that as the antecedent dry period increases so does the pollutant build-up, linearly and indefinitely. A number of models have therefore been proposed which account for the removal of pollutants by processes other than rainfall (e.g. wind). The implementation of such models have two key results.4
Pollutant build up is limited on roads because of removal processes other than rainfall (referred to as ‘loading capacity’).
Varying the antecedent dry period can have a completely different impact on pollutant load compared with the linear build-up model assumed in CIRIA 142.2
3.3. Rainfall depth
CIRIA report 1422 indicates that the use of the Wallingford Procedure,5 which considers the 1 year 24 h storm in order to determine the site-specific rainfall. This in turn is used in a mass balance calculation to calculate the water quality downstream of a highway outfall. The values of rainfall obtained from the Wallingford Procedure are relatively large in comparison with the minimum rainfall required for pollutant wash off; and may not represent the worse case (least dilution) scenario. For a simple assessment the use of a single value of minimum rainfall required for pollutant wash off may be advantageous in order to develop the worse case scenario. Such a value would represent the depth which exceeds the road infiltration capacity to produce runoff and wash off the pollutants. This value has been shown to be in the range of 0·1–2·2 mm.6–8 Values of infiltration capacity suggest that much smaller values of rainfall than those provided by the Wallingford Procedure could be implemented to represent the worse case (least dilution) scenario.
4. SUMMARY
The issue of pollutant loading and runoff from roads is increasingly becoming an important issue since the adoption of the Water Framework Directive (WFD).9 Currently traffic flows exceed 70 000 vehicles per day on UK highways with flows greater than 100 000 the norm for major motorways.10 The effect of such high traffic flow on pollutant loading is unquantified. Given that there are thought to be more than 12 000 outfalls on the motorway network (Fig. 1), a need therefore clearly exists to re-evaluate pollutant loading and its effects in the UK.
