• David Brooke Newton David Brooke Newton

The potential of porous pavement to meet the key stormwater management objectives of peak discharge control, pollutant removal and runoff volume reduction has been recognised for several decades. However, concerns over maintenance and the structural inferiority of porous pavements have led to interest in pavement systems that utilize both porous and impervious pavements. In such systems the porous pavement may act as a treatment device for impervious area runoff. This study examines the extent to which such combined pavement systems are capable of reducing the impacts of urbanisation on downstream hydrology and water quality. To achieve this objective, experimental and numerical investigations were undertaken to quantify the hydrologic, hydraulic and pollutant removal characteristics of modular, lattice pavement, constructed on an impermeable membrane. This type of construction eliminates the potential for differential settlement associated with variations in sub-grade moisture content and avoids the migration of dissolved pollutants to groundwater. Using design scenarios, the numerical models of component processes are combined to investigate the performance of pervious/impervious pavement systems for effective stormwater treatment. A plot-scale field experiment was undertaken to quantify evaporative water loss from this type of pavement. The potential to increase evaporation by incorporating extended detention within the pavement structure was also investigated. Conceptual and numerical models of evaporative water loss from coarse granular media were developed and successfully applied to the experimental data. It is shown that, even without infiltration, this type of porous pavement can substantially reduce runoff volume. However, under subtropical conditions, increasing the amount of water available within the pavement has only a small effect on evaporation. The hydraulic characteristics of porous pavement treating runoff from an impervious area were investigated in an experimental pavement flume. A numerical model was developed to simulate the coupled surface and subsurface flow interactions through the experimental porous pavement. With increasing discharge, surface water runs further onto the porous pavement surface, while the infiltration rate increases sharply towards the leading edge of surface flow. It was found that combined pavement systems can substantially reduce peak stormwater discharges, although the relationships between attenuation, rainfall intensity, rainfall volume and pavement detention time are complex. In general terms, the attenuation provided by porous pavement increases as rainfall intensity becomes larger. However, storm volume has an overriding effect on this relationship. For very short or very long detention times, storms greater than the pavement voids volume receive little attenuation. An intermediate detention time, of the order of 1 to 6 hours, is likely to result in the best average attenuation over the widest range of rainfall intensities and storm volumes. The results of water quality experiments in the experimental pavement flume demonstrate that the experimental pavement can remove inorganic particulate contaminants, down to approximately 5 microns. The particulate removal performance can be improved by increasing the detention time within the porous pavement. However, little benefit is obtained by increasing detention time beyond about 1 hour. An exponential decay model with only one calibration parameter, adapted from filtration theory for wastewater treatment, is used to predict particulate removal efficiency. This model overcomes some of the deficiencies of the k-C* model, widely employed in the analysis of stormwater treatment. The new model predicts the variation in total concentration and particle size distribution through the pavement, as well as incorporating the effects of varying hydraulic conditions on particulate removal.