*2.1. Definition*

Permeable pavements are pavements that simultaneously support the demands of mechanical stresses and rolling conditions, whose structure allows the percolation and temporary accumulation of water, reducing surface runoff without causing damage to their structure [17]. In this type of pavement, the structure is composed of a combination of layers, which are: permeable sub-base, permeable base, permeable bedding layer (when applicable) and permeable surface, dimensioned to withstand traffic loading, distribute stresses on the subgrade and allow the percolation of water. The base and sub-base of the pavement consist of open granulometry materials with aggregates that do not contain fines, or with a small amount of fines, resulting in a relatively large void ratio after compaction [18].

Permeable pavements can be modelled with various types of permeable surfaces, such as porous asphalt, pervious concrete, and permeable interlocking concrete [19]. They can be used as an alternative to conventional impervious hard surfaces, such as roads, car parks, footpaths and pedestrian areas [20].

As for the infiltration system, permeable pavements can be designed in three different ways: with total infiltration of the stormwater, partial infiltration or without infiltration, as shown in Figure 1.

**Figure 1.** Examples of permeable pavements systems: (**a**) with total infiltration; (**b**) with partial infiltration; (**c**) without infiltration. **Source:** Based on ABNT [17].

### *2.2. Permeability, Infiltration and Quality of Infiltrated Stormwater*

Several studies have demonstrated the benefits of using permeable pavements, such as reducing runoff, groundwater recharge, saving water through recycling and preventing pollution by improving the quality of the infiltrated stormwater [21–26]. The Ramsey-Washington Metro Watershed District [27] conducted a study that aimed to implement a permeable pavement system in a 650 m<sup>2</sup> parking lot in Oregon. The investment was aimed at infiltrating and storing precipitation, reducing runoff from stormwater, maximizing permeability of the area and improving water quality, retaining heavy metals and toxins. The cost for the implantation of permeable pavement was US\$102/m<sup>2</sup> and was designed to have 100% infiltration in precipitations of up to 51 mm. Thus, any precipitation up to this figure would not generate runoff. On the other hand, the implementation cost of a conventional pavement system would vary from US\$35/m<sup>2</sup> to US\$46/m2, and for this type of pavement the runoff would be 15,000 litres for a 25 mm precipitation.

Legret and Colandini [28] compared the pollution contained in the drainage of stormwater collected from a permeable pavement to the pollution contained in the drainage from a traditional pavement located in the city of Rezé, France. The retentions of suspended solids, lead, cadmium and zinc were, respectively, 59, 84, 77 and 73% higher in the permeable pavement.

Pratt et al. [29] studied the ability of a permeable pavement reservoir structure to retain and treat petroleum-derived pollutants through in situ microbial bio-degradation. The authors constructed a full-scale model permeable pavement in a laboratory. The pavement comprised pre-formed concrete blocks bedded on clean gravel, with vertical drainage provided through gravel-filled inlets between the blocks. A geotextile membrane separated the block bed from the underlying sub-base, comprising 600 mm depth of washed 20–50 mm granite. The entire structure rested on an additional geotextile underlay, supported by a stainless-steel mesh, allowing effluent to flow into a collection funnel located at the base of the tank. The model was subjected to prolonged low-level hydrocarbon contamination, representative of typical loadings to urban surfaces such as highways and car parks. Water quality was monitored by means of oil and grease concentration, chemical oxygen demand (COD), and pH. The retention efficiency of oil in the permeable pavement was 97.6%. The construction materials had a buffering effect, maintaining an effective pH of about 7.0, which is beneficial to microbial growth. With the benefits shown by the results, the study demonstrated that the structure can be used as an effective in situ aerobic bio-reactor. Also, the development of permeable pavements as pollution treatment devices offers a potential solution to the problem of uncontrolled discharge of contaminant loads associated with stormwater.

Pagotto et al. [7] compared the hydraulic behaviour and the quality of the stormwater drained by a section of a highway in the city of Nantes, France, first using a conventional pavement and finally after the replacement of the conventional pavement with a permeable pavement. Regarding the hydraulic behaviour, the permeable pavement system obtained excellent results. Response times (time elapsed between the beginning of the rain and the beginning of the runoff) were, on average, twice as long on this type of pavement. The delay caused the maximum flow rates to be reduced (6.2 litres/s in the conventional pavement and 5.5 litres/s in the permeable pavement) and the discharge time was higher (average discharge duration was 1.15 times greater for permeable pavements).

There was a grea<sup>t</sup> difference between the two types of pavements in the quality of the stormwater drained. The percentage of hydrocarbons decreased by 92% and the total suspended solids decreased by 81%. Regarding metals, the reduction ranged from 35% (copper) to 78% (lead). For all metals, the particulate forms are retained at a high rate (greater than 70%). However, metals in the dissolved form are retained with greater difficulty. These results explain the considerable level of retention of zinc, cadmium and lead (mainly present in particle form) by weight in percentage terms and the lower retention of copper (mainly present in dissolved form). The study also showed that in each rainfall event, on average, 0.28 kg of sediment was retained in the permeable pavement, against more than 4.1 kg in the conventional pavement [7].

James [30] has shown that traffic on highways is a major source of pollutants and that these are charged to rivers and streams when precipitation occurs. A survey by the Forth River Purification Board indicates that more than 14% of unsatisfactory river water is due to stormwater runoff in urban areas. The quality of the water drained by permeable and conventional pavements was compared and the results obtained are shown in Table 1. It is possible to perceive that the permeable pavement has grea<sup>t</sup> participation in the process of treatment of stormwater, being able to be a grea<sup>t</sup> facilitator in the development of sustainable drainage systems.


**Table 1.** Reduction of pollutants when using permeable pavements compared to conventional pavements.

**Source:** Based on James [30].

Gilbert and Clausen [31] evaluated the amount of stormwater drained in two types of sidewalks: one with typical asphalt surface and other covered with paver. Paver driveways were constructed with stone blocks (115 by 230 mm) interlocking concrete permeable pavement. Pavers were hand installed over 5 cm compacted and screeded coarse sand on top of 15 cm processed gravel. Drainage voids comprised 12% of the surface area and were filled with 3–6 mm peastone. The reduction in the runoff from asphalt to the paver was 72%. The mean infiltration was zero for the asphalt and 11.2 cm/h for the paver. However, the rate of infiltration of the paver pavement decreased with time due to pore obstruction by fine particles. The water drained by the paver sidewalk contained significantly less pollutants compared to the asphalt pavement. Considering the benefits in reducing the runoff and the high infiltration rates, the use of paver in the construction of sidewalks over the traditional asphalt material is more advantageous.

Hou et al. [32] evaluated the infiltration rate of three different types of permeable pavement systems compared to a conventional pavement system. For rainfall rates less than 59 mm/h, the runoff coefficient was zero for the permeable pavement, while the conventional pavement coefficient was 0.85. In addition to the better infiltration rate, it was also verified that the runoff start time after the rain event was higher for the permeable pavement (73 min later). Consequently, the discharge time of stormwater was also higher, which reduces the risk of flooding caused by heavy precipitation.

Eck et al. [33] evaluated the use of Permeable Friction Course (PFC) in the states of Texas and North Carolina in the USA. PFC is a layer of porous asphalt laid in thicknesses of 25 to 50 mm overlaying conventional impermeable pavement. PFC is a type of permeable pavement made of coarse and fine aggregates, asphalt binders, and stabilizing additives, but it does not encourage infiltration and reduces flow volume, such as the full depth permeable pavement. Instead, PFC layers remove rainfall from the road surface and allow it to flow through the porous layer to the roadside. With the use of PFC, the total suspended solids had a reduction of up to 96% when compared to conventional pavement, and good results were found for other parameters such as phosphorus (reduction of up to 78%), copper (69%), lead (above 90%) and zinc (90%). The performance of the Permeable Friction Course can be compared to that of a sand filter because the particulate substances are well filtered while the dissolved substances have little or no retention. Regarding the runoff, 29% to 47% of the total precipitation was retained.

### *2.3. Application of Stormwater Collected from Permeable Pavements for Non-Potable Uses in Buildings*

As seen in the previous section, permeable pavements have the ability to retain pollutants and improve the quality of stormwater. Some studies have evaluated the possibility of using this water for non-potable uses in buildings, such as toilet flushing, garden watering, car washing, among others. Pratt [34] performed a case study at a UK-based hostel whose building had 400 m<sup>2</sup> of roof area and 325 m<sup>2</sup> of parking area. Stormwater precipitated on both surfaces would be stored in the parking sub-base. The parking surface contained permeable blocks that allowed infiltration of stormwater into the sub-base. The water stored in the sub-base was connected to a tank in the hostel and used for toilet flushing. The water storage capacity on the pavement was approximately equal to 34 m3.

Antunes et al. [35] evaluated the possibility of using stormwater from permeable pavements in non-potable uses in residential, commercial and public buildings in the city of Florianópolis, Brazil. In the study, two models of porous asphalt concrete modified with rubber and Styrene-Butadiene-Styrene (SBS) polymer were assessed. The mean percentage of infiltration found for the models was 85%. In this way, the potential for potable water savings ranged from 1 to 18% in the residential sector, 2 to 57% in the public sector, and 6 to 69% in the commercial sector, depending on the tank size.

Hammes et al. [36] evaluated the performance of two permeable pavements in terms of quantity and quality of infiltrated stormwater, aiming at its use in activities that allow the use of non-potable water. The pavements structures are shown in Figure 2 (models A and B). The permeable pavements tested had a mean of 70 and 80% infiltration, respectively. The lower infiltration value for the model A was mainly due to the presence of the filter course. A positive influence of the pavements was observed in some parameters of water quality. However, the need for an additional treatment of the water to adapt it to the expected quality for use was verified. In addition, it was proposed to use the permeable pavement in a parking lot of the Federal University of Santa Catarina (Brazil) for stormwater infiltration, storage and subsequent use in toilets and urinals flushing. The potential for potable water saving would be at least 53%.

**Figure 2.** Permeable pavement models tested. **Source:** Hammes et al. [32].

Thives et al. [37] conducted a study to determine the infiltration capacity and the quality of stormwater infiltrated by permeable pavements with drainage asphalt concrete surface. The concentrations of phosphorus, iron, aluminium, zinc, nitrite, chromium, copper and pH increased after the infiltration in the pavements studied, while the ammonia concentration decreased. However, only phosphorus and aluminium concentrations exceeded the limits required for non-potable uses. It was also found that at least 84% of stormwater could be infiltrated and would be available for non-potable uses.

Thives et al. [38] carried out a study to estimate the potential for potable water savings in multifamily buildings using stormwater collected from paved streets in an area of the city centre of Florianópolis, southern Brazil. For a paved area equal to 9058 m<sup>2</sup> and a stormwater tank capacity of 1000 m3, the potential for potable water savings ranged from 17 to 33% according to the water demand for non-potable purposes.

Although pollutant removal rates vary according to climatic conditions and permeability parameters, the studies mentioned in this review demonstrate the efficiency of permeable pavements in reducing stormwater runoff, as well as improving water quality infiltrated through the pavement. However, the literature still lacks publications related to the real sustainability of permeable pavements, which should relate the benefits brought by these systems to the environmental impacts produced during all phases, from material extraction to the end of the pavement lifespan.
