Print Email Download Reference This Send to Kindle Reddit This
submit to reddit

Suds And Synthetic Sports Pitches Biology Essay

The review of literature will cover present and previous knowledge, commencing with broad research of SUDs and synthetic sports pitches, allowing the review to then focus on the specific analysis of pervious pavements and drainage behaviour of synthetic turf pitches.

The philosophy of sustainable urban drainage is to mimic, as closely as possible, natural drainage within a developed site. The concept is to replicate the natural drainage of the site, prior to development. The sustainable aspect of SUDs enforces the requirement of mitigating the impact that urban development has on flooding and ensuring pollution to bodies of water is minimised.

The objective of SUDs is to minimise the adverse effect development has on quality and quantity of runoff, whilst maximising amenity and biodiversity of a site (CIRIA, 2007a). The three phase concept is illustrated by figure 2.1, symbolising effective design by considering all three components with equal standing. An optimal solution will maximise achieved benefits in all three categories in relation to site constraints.

The optimal solution is known as the “SUDs triangle”, which is the focus of a reduced whole life cost, part of the sustainable approach (Wilson, 2009). This emphasises the multidisciplinary approach that sustainable urban drainage requires: Quantity is managed through source control, quality through the management train and amenity/biodiversity through landscape and ecological design (Wilson, 2009).

At present, the Construction Industry Research and Information Association (CIRIA) provides guidance documents and publications to aid the design and implementation of SUDs. CIRIA emphasise the driver of sustainable drainage being the crucial benefits it enables over conventional piped drainage (CIRIA, 2007a) (CIRIA, 2007b) (CIRIA, 2001):

Control of run-off rates, volumes & reducing the risk of downstream flooding

Reducing pollutant concentrations, protecting downstream water bodies

Encouraging natural groundwater recharge

Enhancing amenity and the aesthetic value of developed areas

Providing habitats for wildlife and opportunity for biodiversity enhancements

Traditionally, the development of an area has resulted in reduced surface permeability, reducing the natural process of free drainage (CIRIA, 2007a). Figure 2.2 represents the change in hydrological process due to impervious surfaces.

Traditionally, a developed area suffers from reduced evapotranspiration and subsurface flow processes, which become replaced by an increased level of surface flow. Storm events provide differing hydrographs given pre- and post-development as illustrated by figure 2.3.

The sustainable philosophy of SUDs in the mimicking of nature can be seen within Figure 3. It is the pre-development drainage nature that is to be maintained, post development. A SUDs approach provides the ability to reduce the magnitude of peak surface run-off and surface run-off volume, whilst delaying peak run-off. SUDs are applied through various techniques and processes that increase infiltration/ground recharge, slowing the drainage process and localising treatment where possible. This approach is applied within the management train model.

The SUDs management train is a tool used to aid drainage design in order to mimic natural drainage as closely as possible (CIRIA, 2007a). A hierarchy of techniques is implemented in the management of surface run-off, treatment and ground recharge:

Prevention – Through efficient design, runoff and pollution can be prevented.

Source control – Local control of water, at or near to the source

Site control – Management of water over an area of a site.

Regional control – Management of runoff from a site or several sites.

The hierarchy is headed by prevention; it being the preferred option for management of surface water. These techniques are illustrated in implementation in figure 2.4.

Figure 2.4 illustrates the philosophy to manage water locally (not end of pipe solutions) with the use of sub catchments in conveyance to the natural water course (CIRIA, 2001). It emphasises that source control and prevention are preferred tools, whilst site and regional control are only considered if site constraints are insurmountable.

From Figure 2.4, it can be seen the four processes of water management are: attenuation, treatment, infiltration and evapotranspiration. These formulate the key tools in the parameter control of designing SUDs. It is these parameters including conveyance that ensures water is adequately managed and conveyed to allow discharge to the natural surroundings without causing excessive problems to the environment and users (CIRIA, 2001). The key processes and are underpinned by the minimal use of impermeable surfaces and a reasonable and responsible manner in returning water into the natural water cycle (CIRIA, 2001). The management of runoff is a dual operation, one of quantity and one of quality control.

Run-off quantity is controlled and managed through infiltration, attenuation, conveyance and harvesting for reuse (CIRIA, 2007a). Run-off quality is controlled and managed through sedimentation, filtration (soil/aggregate matrix, vegetation or geotextile), biodegradation, precipitation (removal of soluble metals by chemical reaction), vegetation uptake, nitrification and photolysis (CIRIA, 2007a). The control processes are determined by the SUDs components implemented.

The SUDs approach for the management of precipitation and surface flow, allows for greater variance in available components, in comparison to conventional drainage (CIRIA, 2001). Components are selected and integrated to form chains in line with the management train concept. The typical components available provide different management characteristics, taken from CIRIA (2007a):

Filter Strips: Sloping areas that are vegetated or grass covered; providing treatment and infiltration to runoff of adjacent impermeable areas.

Swales: Broad, shallow channels that are vegetated or grass covered; providing conveyance, attenuation and infiltration (ground conditions permitting).

Infiltration Basins: Surface depressions designed for attenuation and infiltration.

Wet Ponds: Designed to hold a permanent body of water for quality treatment. Temporary attenuation is available and provide amenity and biodiversity.

Extended Detention Basins: Basins are normally dry surface depressions. They provide storage for large storm events and quality treatment.

Wetlands (constructed): A shallow wetland area improves water quality through pollutant removal (e.g. vegetation, biodegradation) and enhances biodiversity.

Green Roofs: There are many varieties of green roofs, however, all maintain the similar principle of a vegetated roof that attenuates and slows conveyance.

Pervious Paving: Allows infiltration of surface water into the ground or conveyance in a drainage system. Water can be attenuated in a storage layer beneath, reused or allowed to infiltrate.

The Flood and Water Management Act 2010 was published 8th April 2010. Amendments have created the 2010 Act, which encourage the uptake of sustainable drainage systems by removing the automatic right to connect to sewers (DEFRA, 2009). Secondly, the legislation included the provision for unitary and county councils to adopt SUDs for new developments and re-developments, further increasing the need for comprehensive understanding of SUDs and the applications available.

November 2010 provided the revision of The Code for Sustainable Homes part of the Management of Surface Water Run-off and Developments. Water quality criteria provide award for ensuring that rainfall, up to 5mm, produces zero discharge from a site (CLG, 2010). This further emphasises the need to integrate SUDs to housing developments and surrounding areas in order to meet such requirements.

SUDs is a philosophy that encourages a sustainable ethos to design and construction of drainage, resulting in improved water management of three key indicators; quantity, quality and amenity/biodiversity.

The concept is to provide permeable surfaces and mimic natural processes of drainage. The tools of implementation are varied; however, they must be intelligently integrated forming a management train to provide successful drainage solutions.

Effective design is determined by the optimal understanding of the SUDs triangle. Site constraints must be considered and balanced within the design of sustainable drainage. The SUDs trend could be greatly increased given legislation changes and the dissemination of benefits over conventional piped drainage.

Synthetic turf pitches are designed to the Sports and Play Construction Association (SAPCA) code of practice. SAPCA is the UK trade association for the sports facility construction industry. Synthetic sports pitches have a common construction with variance dictated by site conditions and preference of use.

The synthetic carpet is the upper most layer; providing surface characteristics. The shock-pad is the subsequent layer, providing performance requirements. The next layer can vary from macadam (bound base) to a graded stone base (unbound base). The base provides tolerance control for the upper layers. The final layer of the foundation is the sub-base that provides stability, load transfer and is required to be free draining. Figure 2.5 is a representation of a basic bound construction.

The four components will now be reviewed in greater detail, with specific analysis to the manufacture or specification typically used.

Carpet type is determined primarily by the use of the pitch. Shock absorbency and drainage requirements vary depending on site, sport and performance requirements/constraints. There are six major types of synthetic turf: non-filled, water-based, sand-filled, sand-dressed, rubber/sand filled long pile turf (3rd generation or 3G) and needle-punched (SAPCA, 2009). Water-based and 3rd Generation surfaces are of most interest to the thesis, therefore, form the focus of the detailed research:

Water-based synthetic turf is predominantly used for Hockey pitches and utilises water (bound in fibre macro-structure) to change the surface properties. The presence of water reduces surface friction and is applied via an irrigation system. An approximation of 3mm depth of water precipitation is sufficient to meet the International Hockey Federation (FIH) requirements (SAPCA, 2009).

3rd Generation is more common in football and rugby use due to its protection characteristic. Two-thirds of pile height is infill of sand and rubber. Rubber granules provide shock absorbent characteristic.

Tufted carpet is the most common form of manufacture. The fibre (normally woven polypropylene) is tufted into a “primary” backing cloth. These individual tufts are then anchored by application of a latex based secondary backing material, as described by SAPCA (2009) and ESTO (2010). Needle punching is an alternate form of manufacture, primarily for sand-filled. Knitted and woven conclude the manufacturing forms both with differing qualities, costs of production and less common. Punched holes aid drainage, which are spaced at manufacturer’s discretion; however, on average they are centred at 150mm spacing. Carpets are commonly manufactured in widths of 4m. The strips are joined on site and butted together with backing tape to secure jointing.

The shock pad can be integrated to the carpet or laid separate. It consists of rubber crumb/shred bound with a resin binder to define playing characteristics and provide user comfort. It can be prefabricated in rolls or mixed and laid on site if non-integrated. It is laid perpendicular to the carpet if prefabricated.

Commonly the base layer is a porous layer of open graded bituminous macadam. It provides stability to the carpet, frost heave resistance and spreads surface loads. It also aids in maintenance with replacement/repair of carpet.

The requirements of a sub-base to synthetic sports pitches are (SAPCA, 2009):

Should be capable of supporting and transmitting loads to sub-grade without causing deformation during construction.

Should be capable of transmitting operation and maintenance activity loads to the sub grade without long term deformation or adverse affect to performance requirements.

Allow the free drainage of ground and rain water to the natural ground or away with drainage system.

The minimum depth of a compacted sub base layer is 300mm, compacted in layers not exceeding 150mm unless conditions dictate otherwise (SAPCA, 2009).

The main components of a synthetic sports pitch have been reviewed; carpet, shock-pad, base and sub-base. However, the general drainage requirements of synthetic sports pitches is required for comprehensive understanding in relation to the thesis area of research.

The drainage requirements of a synthetic turf pitch are (SAPCA, 2009):

Water retained within construction must not result in reduction to load-bearing capacity or in any frost damage

Water removed from surface must be of sufficient rate to prevent surface ponding

Meet EA and local Water Authority requirements

Basic design involves a perimeter drain to prevent run off from surrounding areas. Silt/inspection chambers should be positioned on the perimeter drain at change in direction. Lateral drains are centred 5-15m depending of site conditions. Drains usually consist of perforated plastic pipes, with geotextile as a safeguard to silting. Figure 2.6 represents the typical layout of a synthetic turf pitch drainage scheme.

Synthetic turf pitch components can be categorised by their drainage characterises. Permeability is the key drainage characteristic to all four key components of a synthetic turf pitch, as all components are free-draining. However, storage is a characteristic shared only by the carpet and sub-base. Therefore, focus of these two components will be of most interest in this thesis.

Synthetic sports pitches are typically treated as enclosed drainage units, allowing isolation, creating ease of testing without interference.

The drainage characteristics of the carpet would appear to be most significant in determining a synthetic turf sports pitch’s drainage ability. This is assumed because of the latex backing applied to the carpet and the non-existence of drainage hole spacing specification.

An understanding of SUDs and synthetic turf sports pitches has been achieved as a general encompassment of the two ideologies. However, the Literature Review will now focus on the specific concept of pervious pavements to allow the evaluation of combining the concept with a synthetic turf sports pitch.

The principle of a pervious pavement is to allow the infiltration of rainwater through the surface to the underlying construction layers, allowing some attenuation (temporary storage) prior to infiltration to the natural ground or reuse by collection (CIRIA, 2004) (CIRIA, 2007a). Pervious pavements are commonly paved or loose surfaced car parks, driveways and low traffic roads. Pervious pavement systems have several benefits (CIRIA, 2004) (CIRIA, 2007a):

Effective removal of pollutants mitigating risk of contamination to surrounding environment

Reduction of surface ponding and peak outflow, through infiltration

Reduction of surface run-off volume providing mitigation of flood risk.

Dual space usage

Low maintenance (periodic brushing of surface is common practice in UK)

Pervious pavement systems have typical limitations (CIRIA, 2000) (CIRIA, 2004):

Not all uses of land are suitable (industrial areas are not appropriate as risk of widespread pollution is too great)

Require a flat/level area.

Technical understanding is a limitation, which is slowly diminishing with more study and analysis. Long Term performance and large scale in-situ testing are two areas which are aiding the ability of implementation.

Pervious pavements can be catagorised in to two distinct groups:

Pourous: infiltration is across the entire surface eg pourous asphalt, gravel etc.

Permeable: impervious material forms the surface, however, surface voids are formed to allow infiltration. (CIRIA, 2004) e.g. block paving or syntheitc turf. Therefore, synthetic turf pictches would be classified as permeable pavements.

Pervious pavements are characterised by a free draining sub-base to allow flow of water (CIRIA, 2001). The surface characteristic is to infiltrate water to the sub-base faster than the rate of rainfall to prevent ponding (CIRIA, 2001). The permeable sub-base then stores the runoff volume and slowly discharges. The discharge process can take several forms; full, partial or zero ground infiltration with/without the integration of outlet pipes (CIRIA, 2001). Figure 2.7 represents a pervious pavement that allows full infiltration into the ground.

Perforated pipes laid in the sub base can aid in capacity of the system by conveyance and also reduce infiltration into the sub grade if there is risk of destabilisation through increased moisture levels as shown in Figure 2.8.

Figure 2.9 represents zero Infiltration, which involves a lined (impermeable) system. This system would be implemented for sensitive soils, high water tables or risk of contaminant mobilisation (CIRIA, 2007a).

A geotextile aids in filtration of pollutants. The sub base and geotextile work in conjunction to treat runoff water by filtration, absorption, biodegradation and sedimentation (CIRIA, 2007a). The three construction types allow for variation given site constraints and hydraulic design required for use and location.

There are four key criteria to hydraulic design of a pervious pavement (CIRIA, 2007a):

Infiltration rate of pavement surface

Storage volume required

Outfall capacity from pavement system

Extreme event performance ability

Capacity of storage is determined by void volume of the sub-base and depth of layer. Storage volume may be calculated based on the volume and porosity of the underlying storage layer (either aggregate or geocellular plastic systems) as described in CIRIA (2002). Hydraulic design must be considered in the context of structural design, determine use and performance characteristics.

Structural design of a pervious pavement is based upon the guidance of CIRIA. The sub-base is the main structural element of a pavement. A sub base depth is a function of hydraulic design and of the sub-grade performance.

Attenuation of flow is dependant on porosity (function of void volume), therefore, attenuation is determined by sub-base design (CIRIA, 2000). The sub-base is defined by CIRIA (2007a ) to be of rough and angular particles to maximise inter-particle friction. The grading is a compromise between three characteristics: Stiffness, permeability and storage capacity (CIRIA, 2007a). Specification of grading was discovered to be of some variance with a common principle, free draining. Typical grading provided by CIRIA (2007a) is represented by figure 2.10, being a 4/20mm grade.

This is in agreement with Interpave (2006) and Tobermore (2011), who also suggest 4/20mm grading. However, the suitable grading of sub-base is contradicted by alternate sources. Previous work by Danes (2010) and Emm (2009) utilised a Type 1X mix for sub-base. Type 1X (figure 2.11) is referred to as “a permeable sub-base” and “free-draining”, as specified in TRRL Report PA/SCR243 by Luck (1994) (DMRB, 2001).

Luck, 1994: states Type 1X is a granular material which can be considered for the use in lower layers of a road foundation platform to facilitate “under drainage”. Type 1X is defined as crushed rock, slag or concrete and meets BS sieve passing percentages of Figure 2.11 (Luck, 1994).

In agreement, Gridforce Ltd specify to use The Department of Transport ‘Specification for Highways Works Road Pavements’ clause 805 Type 3 (DOT 3), which “is effectively the former MOT Type 1X” (Gridforce, 2010). The concept of MOT Type 1X and MOT Type 3 being of equivalent grade is further concluded by suppliers. MOT Type 3/Type 1X (also known as DOT Type 3) is described as a 40mm, pure crushed granite or limestone; screened to reduce fines but not zero fines (Mainland Aggregates, 2010).

The varying sub-base grade envelopes have been combined to show comparison. MOT Type 1 (DOT Type 1) and MOT Type 3 (DOT Type 3) are taken from MCHW (2009). The 4/20mm grade envelope is provided by Interpave (2006). Type 1X is provided by Luck, 1994.

Figure 2.12 represents the Type 3 (DOT3) containing reduced fines compared to Type 1 (DOT1), hence the ability to be free draining. The 4/20mm grade is not a well graded material and has, by definition; very few particles passing 4mm. 4/20mm will be free draining but will suffer from compaction difficulty compared to Type 3 due to its poor grading.

Figure 2.13 is a comparison of Type 1X and Type 3 grade envelopes. They show greater comparison but with Type 1X being of slightly narrower envelope and with reduced fines (not zero). If a sub-base is specified as Type 3 or 1X, the characteristics will be of greater similarity, as opposed to 4/20mm. However, all grades considered for sub-base structural design have wide envelopes, allowing variance of aggregate performance in a single specified grading.

Full breakdown of Type 1X gradings and related aggregates are provided in Appendix B.

The limitations of pervious pavements are complimented by synthetic sports pitches, which are large open flat areas and at low risk to pollution.

The sub-base grading has concluded to be suitable to Type 1X yet variation in specification is likely, yet the main principles remain unchanged. The construction is similar to that of a synthetic sports pitch, with permeable sub-base but with variance in permeable pavement surface, appropriate to use.

Hydraulic and structural designs are basic and simplistic. Pervious pavements follow simple principles yet there is a suggestion there is little understanding of field testing and long term performance.

Published academic research will now be reviewed with focus on permeable pavements. This is due to the lack of literature on drainage of synthetic sports pitches.

Scholz & Grabowieki (2007) provide a general review of permeable pavement systems in relation to literature and industry direction. Common applications are driveways, low traffic roads, pavements and car parks; hence past research has focussed on these themes (not synthetic turf pitches). Published researched has focussed on three themes on the topic of permeable pavements: hydraulic performance, pollutant removal and long term performance. Two additional topics that will be touched upon will be value and aggregate drainage.

Research has focussed on the ability of permeable pavements to reduce the effects of storm events through hydraulic performance. Pratt et al. (1995) concluded that pervious pavements showed reduction in outflow volume, rate and pollutant concentrations. The other key conclusion was that antecedent factors were influential on variability of performance.

In agreement, Schluter & Jefferies (2001) and Macdonald & Jefferies (2001) illustrated the effectiveness of pervious pavements in reducing peak outflow rate and total volume of runoff by the monitoring of two sites in Scotland (CIRIA, 2007). Macdonald & Jefferies (2001) observed a lag time between rainfall (centre of gravity) and peak outflow, of the range 29-600 minutes, much greater than the impervious surface adjacent that averaged 9.3 minutes (CIRIA, 2007). This was slight variance from the research of Schluter & Jefferies (2001) that observed a range between 43-143 minutes (CIRIA, 2007).

Overall, variation in outflow percentage was between 2.5% – 79.5% from both studies. This was dependant on duration of storm event, total rainfall and antecedent precipitation according to Schluter & Jefferies (2001). Macdonald & Jefferies (2001) produced a variance of 21.45 – 72.8% in outflow percentage. The difference between the two is attributed to evaporation or retention within the pervious pavement construction (CIRIA, 2007).

Figure 2.14 represents an example hydrograph from the pervious pavement study by Schluter & Jefferies (2001).

The peak flow was shown to reduce at a mean value of 76.8% by Macdonald & Jefferies (2001). The pervious pavement could withstand an average of 7.4mm of rainfall before runoff was induced, compared to a mean value of 0.76mm on a conventional surface (CIRIA, 2007).

Gilbert & Clausen (2006) investigated hydraulic performance of permeable pavements on driveways. Reduced water runoff was exhibited by paver and stone permeable surfaces (asphalt>paver>stone). The mass export of measured pollutants was relative to the difference in runoff volume rather than concentrations. Asphalt to paver was 72% reduction of runoff and asphalt to crushed stone was 98% reduction. Gilbert & Clausen (2006) observed paver runoff to be 40% of rainfall, which was similar to that of Pratt et al. (1995) findings, which were 37%-47% of rainfall. Booth & Leavitt (1999) observed less than 1% runoff from precipitation for turfstone, similar to Gilbert & Clausen (2006) crushstone (i.e. loose granular stone).

Abbot & Comino-Mateos (2003) findings agreed with Pratt et al. (1995) as their In-situ, full scale testing of a car park (over 13 month period), showed a permeable pavements ability to reduce peak flow, extend outflow duration compared to rain events. Abbot & Comino-Mateos (2003) suggested that the limitation to widespread adoption is lack of in-situ full-scale technical data. The testing acknowledged was that of Macdonald & Jefferies (2001), Schluter & Jefferies (2001) full scale and lab scale by Bond et al. (1999) and Pratt et al. (1995).

Abbot & Comino-Mateos (2003) undertook monitoring 150m downstream of the test site, using a BS3680 Triangular- notch thin-plate weir and ultrasonic probe. Rainfall was monitored with a 0.2mm rain gauge tipping bucket. Lag time from storm event on catchment to rise in outflow varied from 5 minutes to 2 hours. Peak flow delays ranged from 5 minutes to >9 hours (average 2 hours). The amount of water of water draining each storm event ranged from 4-47% (22.5% average).

Scholz & Grabowieki (2007) agreed with the two full scale findings of 2001, concluding that pervious pavements are more effective than conventional asphalts at reducing peak flow, ponding and providing longer discharge periods, supported by Booth & Leavitt (1999), Abbot & Comino-Mateos (2003) and Pagotto et al. (2000).

The following year, 2008, Straet et al. (2008) compared “greened” porous pavements to open jointed paving with lab testing. The parameters varied were rainfall intensity, slope and initial moisture. Rain simulation was discussed and highlights the influence it could have on results. Results showed pavers producing less run off than greened porous pavements. Straet et al. (2008) observed the influence of initial saturation and drying time of different systems. It was evident that the plastic geo-cells and open celled paving grids were underperforming the block paving in infiltration. This is in line with the concept of vegetation slowing flow/infiltration. The recovery time was longer for the “greened” paving systems, with block pavers drying out quicker. The moisture characteristics were emphasised to be determined by the porosity of the sub-base (Straet et al. 2008). Schluter & Jefferies (2002) concluded that porosity of sub-base had little effect on volumes of flow discharged but had a significant impact on the peak outflow rate.

Permeable pavements with aggregate sub-bases can provide good water quality treatment (CIRIA, 2007a). The attenuation of sub-bases enhances the settlement and biodegradation of pollutants. Reduced peak flow causes less of a short-term shock pollutant load to the receiving waters and allows increased dilution (CIRIA, 2007a).

Three mechanisms reduce the concentrations of pollutants discharged (Day et al., 1981; Pratt et al., 1995 and CIRIA, 2007a):

Pollutants are retained within the pervious construction, physically trapped or adsorbed on material

The volume of water discharged is reduced hence the mass of pollutant being conveyed is itself reduced at any given concentration

Hydrocarbons and other organic materials trapped in the upper layers of the construction are degraded by micro-organisms (present in the ground)

Day et al. (1981) compared pollutant concentrations in runoff from various surfaces, permeable and impervious. Day et al. (1981) concluded the permeable surfaces provided a general reduction in pollutants, attributed to the reduced load being discharged due to lower runoff volumes (CIRIA, 2007a). Gilbert et al. 2006 partially differed in view. His findings showed paver driveways producing lower pollutant concentrations in runoff than asphalt and crush. However, crush produced the lowest overall runoff. This is likely due to nature of crush not trapping pollutants as effectively as the paver.

Macdonald & Jefferies (2001) monitored an unsealed base car park adjacent to an impervious car park, whilst Schluter & Jefferies, 2001 monitored a sealed base porous car park. Macdonald & Jefferies (2001) reported some increased runoff pollutants, attributed to decay of organic matter from plant debris. However, Schluter & Jefferies (2001) study showed a reduction to heavy metals and general pollutants (CIRIA, 2007a).

Scholz & Grabowieki (2007) agreed that benefits of permeable paving was more than reduced run off and natural recharge to the ground but pollution prevention such as Pratt et al. (1999b). Pollutants such as hydrocarbons and heavy metals were highlighted to be treated through biodegradation and infiltration (Brattebo & Booth, 2003). With variation in findings, which can be justifiably explained, the consensus is pollutant removal can be effective.

Abbot & Comino-Mateos (2003) studied a 12,400m2 car park area over a 13 month period. One interesting aspect to arise was two of the storm events produced >100% outflow, assumed to be water held in storage from prior storm events. This occurrence was also observed by Bond et al. (1999) (Abbot & Comino-Mateos, 2003).

Infiltration tests were done with a TRL Infiltrometer, BS DD 229:1996. The infiltration rate decreased over 3 years, becoming 83% of the initial construction (Abbot & Comino-Mateos, 2003). Gilbert et al. (2006) produced single ring infiltration results that showed that areas of wear (therefore compacted crush material) had lower infiltration rates. Gilbert et al. (2006) linked in the concept of antecedent conditions to be a cause to outliers of the infiltration test data for the paver driveways. Gilbert et al. (2006) concluded that asphalt had zero infiltration compared to both paver and crush having “moderately rapid” rates as expected (9.7-11.8cm/h means). Gilbert et al. (2006) also found Infiltration rates reduced (small reduction) over the study duration (22 months) due to fines clogging openings in paver, similar to Abbot & Comino-Mateos (2003).

Pratt et al. (1995) assessed sediment degrade of permeable paving systems. Sediment and sediment associated pollution will tend to be trapped in the upperparts of the sub-base (Pratt et al. 1995). The accumulation affects the infiltration rate, reducing it (Pratt et al. 1995). This is in agreement with Gilbert et al. (2006). The time of failure is difficult to predict given the numerous factors that contribute to this process. Reconstruction of the surface and upper section of sub-base is estimated to be possibly required between 15-20 years. Therefore, the opportunity exists, to coincide reconstruction of the upper sub-base with replacement of the synthetic turf (maintenance requirement).

Booth & Leavitt (1999) tested four types of in-situ full scale pervious pavements (3x6m) with impervious asphalt as a fifth area. The parking bays were constructed in 1996. Brattebo & Booth (2003) tested the same bays after 6 years of parking use. The four pervious paving systems infiltrated all rainfall even during intense storms. Quality remained consistent with areas of increased, decreased or the same concentration levels. The maintained performance is partially attributed to the ground conditions and the low intensity nature of rainfall given the geographical location, North West USA (Brattebo & Booth, 2003). Scholz & Grabowieki (2007) agreed with the key findings of no major signs of wear, runoff was negligible and infiltrated outflow had reduced levels of pollutants than direct run off from the asphalt.

Pratt (1999a) assesses innovation. Pratt (1999a) assumes 30% void space in a sub-base to the pervious pavement. Plastic cellular void products can achieve 95% void spacing but not always as viable an option but are less affected by silt (Pratt, 1999a). Another innovation is provided by reuse of water through storage and retreatment of permeable pavements reservoir. However, all research to date has been in terms of house driveways or car parks.

Scholz & Grabowieki (2007) raised the innovation of heating and cooling systems applied to the sub base (geothermal systems) region of pervious pavement systems. Focus of this research was the effect on increase growth of microorganisms during the artificial temperature fluctuations. Scholz & Grabowieki (2007) highlight long term performance of pavements, in particular relation to pollutants is still requiring further study. However, due to the nature of MPS sports pitches, it is unlikely that pollutants will be of great interest. Scholz & Grabowieki (2007) concluded recycling pervious pavements was encouraging and particularly combined with geothermal systems being incorporated with upper sub-base treatment, put forward by Pratt (1999b).

The final area of published research is that of the general approach of SUDs. Bastien et al. (2010) assessed the holistic approach of SUDs management and treatment train. The barriers of implementations are highlighted be as land take, construction costs uncertainty regarding maintenance and adoption (Bastien et al., 2010). Conversely, the key drivers to implementation were highlighted as improved environmental amenity and increased amenity (Bastien et al., 2010). Ideally, the opportunity for the combination of SUDs and MPS sports pitches appears well matched to meet the issues raised by Bastien et al. (2010).

Duffy et al. (2008) presented the capital costs of traditional drainage being more than double the capital costs of SUDs and that average maintenance costs would be 20-25% greater for traditional drainage. Whole life cycle costing of traditional drainage is, therefore; approximately double that of SUDs (Duffy et al. 2008). Heal et al. (2009) evaluated the maintenance of Hopwood and concurred the findings of Duffy et al. (2008). Hopwood service area had annual maintenance costs of £2,500 for SUDs, compared to £4,000 for conventional drainage structures. There is little research on costs, however, from the brief research reviewed, SUDs appears to have cost benefits.

The aggregate sub-base is critical to the drainage behaviour of a synthetic sports pitch. Therefore, focussed review of aggregate drainage is required for comprehensive understanding.

Flow of water through a granular material is driven by hydraulic pressures. For Laminar flow in a saturated granular material, Darcy’s law states (Jones & Jones 1989):

v = ki

v: discharge velocity, k: coefficient of permeability , i: hydraulic gradient

Darcy’s Law is only valid at low hydraulic gradients (Jones & Jones 1989), therefore, is valid in the context of a permeable paving sub-base. Permeability is a function of 3 properties (Jones & Jones 1989):

Grading (degradation considered)

Particle Size and texture

Packing (function of compaction)

Permeability will be relatively high in Type 1X given the nature of the grading. Jones & Jones (1989a) state if the likely overall porosity is 10-20% of an aggregate, this provides a yield capacity of 3-10%. Therefore, 3-10% volume of the sub-base can fill and subsequently drain due to percolation into the sub-base from storm event.

A more open grading or lowering the water table will increase the effective porosity (yield capacity) of a sub base (Jones & Jones, 1989). The most significant drainage processes in an aggregate are vertical and horizontal flow (Jones & Jones, 1989). Jones & Jones (1989) highlight the impact hydraulic boundary conditions can have by reducing free draining nature if confined by impermeable boundaries.

Research has observed and proven the benefits of permeable pavements through reduction of runoff (as a source control) and pollutant reduction. However, further full scale testing and long term performance is required to further reduce the barrier of adoption through lack of full understanding.

Research has not combined the concept of permeable paving to uses other than conventional driveways, car park and low traffic roads. Research on the influence of snow and freezing conditions have yet to be published in relation to permeable pavement performance.

The elements of synthetic sports pitch share similarity to a car park but constructed with a difference surface. Published research suggests the strong opportunity to combine the two concepts of permeable pavement and synthetic turf pitch with positive results.

Published research in the field of permeable paving can be categorised by the SUDs objectives quality, quantity and amenity. Initial research focussed on quantity; then quality become of emphasis, however, the more recent interest of long term performance due to the difficult nature of quantifying amenity.

From the literature review it can be concluded that the concept of permeable paving can be incorporated within a synthetic turf sports pitch. There has been increased research of SUDs, in particular in permeable paving, in the past two decades. However, none of the permeable paving research reviewed has much knowledge in relation to synthetic turf sports pitches. At present there is little understanding of synthetic turf sport pitches ability to attenuate flow and general drainage behaviour. It is this concept that will form the focus of this thesis.

The limitations of permeable pavements are complimented by synthetic turf sports pitches. Artificial pitches are level and flat, with low weight/low speed use. The activities that take place on a sports pitch create negligible pollution. Present sports pitch design requires a sub-base layer of free draining properties, therefore, permeable paving and synthetic turf sports pitches share a common concept.

The thesis will focus on the gap in research knowledge, through the combination to permeable pavements and synthetic sports pitch. The two concepts exist successfully in isolation; therefore, it is the aim of the thesis to investigate the effectiveness of combining the two theories. It is a concept that would meet the growing demand of social and urban development responsibility.

Print Email Download Reference This Send to Kindle Reddit This

Share This Essay

To share this essay on Reddit, Facebook, Twitter, or Google+ just click on the buttons below:

Request Removal

If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal:

Request the removal of this essay.


More from UK Essays

Doing your resits? We can help!