Normally, rain falling on impermeable surfaces, such as a road surface, quickly enters the drainage system, arriving at the outfall to river or stream very soon after falling from the sky. It is estimated (Interpave, 2005) that in a fully forested, lowland catchment only 5% of rainfall will flow across the ground surface, the remainder will be delayed by the vegetation to such an extent that it will soak into the ground. For agricultural land with less vegetation 30% may flow across the surface. However for an urban environment with piped stormwater drainage systems 95% is carried to the surface water bodies. For this reason, in an urban environment the water arrives much more quickly than if it had taken a natural route (movement over vegetated surfaces and by percolation through the ground). Thus the flow pattern in the river or stream rises and falls much more quickly, and reaches higher maximum flow values and lower minimum flows, than it would in a non-built-up area (see Fig. 13.11, explained in more detail below). This “peaky” flow leads to increased frequency of flooding and to reduced irrigation flows in times of drought as there is less water soaked into the ground to provide dry weather seepage supplies to surface water bodies. For these reasons, if a rapidly filling, but slowly emptying store can be provided in the pavement then this undesirable effect will be reduced.
In the example illustrated in Fig. 13.11, three rainfall events occur within a two — day period. It is assumed that this particular pavement can hold 20 mm of rainfall (i. e. 201 per m2). The first storm (0-8 h), which peaks at 5mm/hour, almost causes the pavement to fill with water. The pavement has drained back to half full when
|
Time (hours) Fig. 13.11 Sample rainfall, storage and outflow hydrograph |
the next storm (18-22 h) arrives. This storm, with peak loading of 8mm/hour does cause the pavement to fill for a short time, whereas by the time the third storm arrives (40-44 h) the pavement has drained further and the 5 mm/h peak intensity storm is just handled by the system. Note how the maximum outflow (egress) is about 1.5 l/m2 when the pavement storage is full (21-23 h) very much less than the peak rainfall of 8 l/m2 which would otherwise arrive into the drainage system and be fed to a stream or river.
This benefit is well illustrated by the performance of car park drainage pavements six years after initial construction reported by Brattebo & Booth (2002). Two of their pavements had grassed unbound surfaces (sand and gravel) held in a plastic grid arrangement while another two used concrete block surfacing with about 40% and 10% open area. Figure 13.12 shows rainfall and run off for a grass-sand and a reference asphalt pavement. Even for the heavy rainfall event (121 mm in 72 h) only 3% of this ran-off the surface of the permeable pavement whereas run-off from the asphalt pavement closely follows rainfall. The other three permeable pavements gave even less run-off.
