Highway stormwater runoff

This paper reviews international highway runoff characterization studies from each continent to make a comparison of the results and to underline homogeneity in the models and the methods of monitoring and analysis of highway stormwater. To reach this objectives, the paper is organized as: review and organization of runoff characterization paper from North America, Europe, Asia, Australia and Africa (annual average daily traffic, drainage area, number of lanes, pavement type, impervious area, land use, hydrologic characteristic, sampling methods, QA/QC information, analytical technique, holding time, detection limits and water quality parameters); comparison of the results; description of importance of first flush (Gnecco et al., 2006); examination of toxicity of highway runoff (Kayhanian et al., 2008), and particle bound pollutants; individualization of surrogate water quality parameters; description of modeling and prediction of HWY runoff. Finally, the paper describes: indications presents in environmental regulations; potential use of universal method of collecting, analyzing, and interoperating runoff data; potential application of regional and international runoff quality data to address water quality standards and regulatory compliances. Discussion of results evidences that there is a need for an homogeneous approach in collecting data to the aim of compare them with the existing data. In fact, a high variability of the many water quality constituents is observed in highway stormwater runoff (table 1). This variability makes difficult the prediction of stormwater runoff constituents from site to site, or between runoff events at a single site, which is necessary for the planning and assessment of highway stormwater runoff data. To enable accurate prediction of the various constituents, a comprehensive understanding of the constituents of highway stormwater runoff, and more importantly, how these constituents vary both temporally and spatially, is required(Thomson, McBean, Snodgrass, 1995). Finally, this comparison of results also shown that is not necessary to implement new data campaign where conditions are quite similar other experiences, to avoid useless costs.