A Longitudinal Study of Playground Surfaces to Evaluate Accessibility - Final Report

Qualitative Analysis by Surface Type

The observational data collected through the visual inspections of the sites and discussions with the playground owners provided invaluable lessons learned from the longitudinal study and may provide some explanation to the overall effectiveness of various types of surfaces.

Poured in Place Rubber

Nine were surfaced entirely with poured in place rubber (PIP), while five sites were combined with engineered wood fiber (EWF).  The surface cost for PIP ranged from $6.59/sq. ft. to $19.80/sq. ft.  The wide range of cost per sq. ft. can be attributed to the fact that PIP is often sold on a sliding scale, the more material purchased, the cheaper the unit cost.  The cost for PIP has also been dramatically affected over the last five years due to volatility in the petroleum market.

Within the first 12 months of installation, the sites surfaced with PIP did not have any recorded locations where the surface samples exceeded the accessibility standards for slope, cross slope, changes in level or openings.  The mean for the surface firmness and stability was well under .50 inches.  From the “looks” of the surface locations, they appeared to be very accessible within a 12 month period from installation.  However, a major concern was discovered at one of the test locations where the playground owner opted to also have the surface tested for impact attenuation and compliance with ASTM F1292‒99/04.  Various locations on site were tested using the TRIAX to record GMAX and HIC.  The maximum values allowable by the standard are 200 for GMAX and 1,000 for HIC.  Drop heights from composite equipment up to 8 ft. high passed the field test.  But it was the PIP surface at two swing bays that was found in non-compliance with HIC scores well over the 1,000 HIC allowable under the standard.  The playground owner used the terms of the warranty and purchase order as a binding agreement requiring the manufacturer, at its own expense, to return to the site and repair the surface installation.  Approximately 2,000 sq. ft. at the swing bays was resurfaced to add more depth to the PIP.  When the surface area was retested with the TRIAX, the HIC ranged from 650‒750 at the swings, well under the 1,000 maximum allowable by the standard.  Had the playground owner not discovered the non-compliant surface area until after the warranty had expired, it would have cost the agency in excess of $35,000 to correct the surface area serving four swings.  During the course of the longitudinal study, at least two additional playgrounds surfaced with PIP were found in non-compliance with ASTM F1292‒99/04.  In each case, the playground owners required the installers to return to the site to make corrective actions.

Between 24‒36 months, locations surfaced with PIP began to show signs of cracking and instances where the top layer of surface had worn off under the swings, slides, or other equipment with rapid motion.  At one such site, the top granular layer of PIP began flaking off in 1‒4 inch sections throughout the surface area, not just high traffic or rapid motion areas.  As it turns out, the surface material at this site was not installed per the manufacturer’s recommendations.  The installation occurred in the late fall when the temperature was 40 degrees Fahrenheit and falling.  The manufacturer installation instructions show the preferred atmospheric temperature for installation to be 40 degrees Fahrenheit and rising.  The error in installation is attributed to the contractor and pressure to stay on schedule as the construction season came to a close for the winter.  The playground owner insisted upon corrective measures.  Another contractor was brought in to apply a top binding coat.  This cost was absorbed by the sales representative.

As previously noted, two sites were installed as one layer comprised of bonded large particle rubber shreds.  This type of installation is a break from the traditional product known as PIP and consisting of two layers, a wear layer with larger rubber particles and a custom top layer with granular particles.  These two sites were assessed with locations throughout the playground where the large particles had separated from the bonded layer in chunks.  This was notable in areas such as the swings and teeter totters.  Particles were also separated from the bonded layer by the turning movement of the wheel on the Rotational Penetrometer during the site assessments.  Under the terms of the product warranty, the playground owner required corrective action where the damaged sections were cut out and patched with new material.

Tiles

There were two sites surfaced completely with tiles (TIL), while eight additional sites were combined with engineered wood fiber (EWF).  The 2 ft. x 2 ft. bonded rubber interlocking sections are marketed as easier to install and more flexible should they need to be reconfigured to accommodate new playground equipment. The cost for TIL ranged from $8.96/sq. ft. to $21/sq. ft.  TIL was the most expensive of the five types of surfaces identified for study.  This can be attributed to the number of small surface area installations where the use of TIL was less than 2,000 sq. ft.  These installations limited the use of the TIL to connect the accessible route from the playground perimeter to the transfer system of the elevated composite structure.  The remainder of the larger play area was surfaced with a loose fill material. 

For one agency which manages more than 50 playgrounds, their selection of TIL was based on the agency’s perceptions that installation by its own personnel would help to drive down the overall cost of the playground project, stretching more dollars when budgets are tight.  They also believed that the TIL would create an accessible route with less maintenance requirements than the EWF predominately used on their other playgrounds.  During the initial site visits, the TIL had started to shift on at least two playgrounds where the parks maintenance staff had installed the surface system as opposed to installation by a contractor certified by the manufacturer.  The playground owner attributed the construction error to the learning curve involved with installation of the new surface and reported each new site was improved based on the experience maintenance staff was gaining.  The agency’s third playground with TIL was bordered by a landscaped paver retaining wall.  Improper drainage from the landscape in the retaining wall was causing a build-up of silt on and under the tiles.  Within the first month of installation, at least a dozen tiles at the border were pulled up to remove the silt build-up, the section was thoroughly cleaned, dried and the TIL were re-adhered to the concrete sub-base.  Maintenance staff was on site making the repairs when the accessibility assessment was conducted.  The assessment team brought another area to their attention where the four tiles bordering the concrete walk at the entry to the playground were raised more than a quarter of an inch and adversely affecting the accessible route into the play area.  Maintenance staff was able to remove the four tiles, shave the underside and re-install so that they were flush level with the concrete walk all while the assessment team was concluding the field testing. 

As noted in Table 3, the Mean SDS for TILs continued to increase each year of the study.  The mean for the firmness and stability of the tiles tested in the sample was also under .50 inches, similar to the PIP.  Throughout the period of the study, there were reoccurring instances where the TIL were assessed with punctures holes ranging from .50 inches to more than 2 inches in diameter and where the seams had started to shift or buckle creating openings and changes in level along the accessible route.  It was unclear whether the puncture holes were products of intentional vandalism or unintentional damage from users stepping on rocks and other foreign objects with enough force to penetrate the surface.  One playground owner went so far as to install signage at a site “No high heels” with the image of a woman’s shoe.  The maintenance staff was also able to replace the TIL with puncture holes following the site assessments.

Deficiencies were identified at two playground sites surfaced with a combination TIL and EWF.  The intent of the playground design was to use the TIL as the primary accessible route to points of aggress/egress and fill the remaining use zone with EWF.  The loose fill particles of EWF were scattered throughout the play area, across the tiles, concrete walkway and in the grass.  Some of the particles had started to lodge in the TIL seams causing separation at the seams.  There were even instances where the particles had lodged so deep in the seams that the adhesive had degraded and the TIL had separated from the concrete subsurface.

During Years 3‒5, locations were identified where the TIL had cracked in the center unable to support repeated weighted foot traffic.  These instances occurred where either the subsurface or structural integrity of the surface product was compromised.

Engineered Wood Fiber

There were six sites surfaced entirely with engineered wood fiber (EWF).  In addition, there were five sites surfaced with a combination PIP and EWF, and eight sites surfaced with a combination TIL and EWF.  The EWF ranged in cost from $ .74/sq. ft. to $2.50/sq. ft.  One of the emerging playground surfacing trends is to install a unitary surface, such as PIP or TIL, as the primary accessible route to accessible equipment and fill the remainder of the equipment use zones with a less costly loose fill surface material, EWF or shredded rubber (SHR). While this has been thought to be a cost effective solution to playground surfacing, the loose fill material has caused more maintenance issues when combined with a unitary surface.

The playground sites in the sample with EWF experienced the greatest frequency of high SDS and mean for firmness and stability.  Every playground installed with EWF was observed with undulation across the horizon of the surface area.  The undulating surface material created changes in level, running and cross slopes exceeding the maximum allowable standards resulting in non-compliant accessible routes to play components.  There was no observational difference in the issue of undulating surface between sites installed by maintenance personnel compared to sites installed by contractors.  Review of installation data and discussions with staff indicated the loose fill surface installations did not follow the same procedures noted in the installation instruction by the surface manufacturer or in ASTM F1951‒99 lab reports.  EWF surface installations were mostly infilled, raked and leveled.  A minimum amount of surface compaction was conducted, if any.  This is a serious departure from the installation procedure used on the lab test samples for ASTM F1951‒99, where the surface material is installed in 3‒6 inch layers, watered, raked, compacted and installed with another layer following the same procedure and finally compacted with either a drum roller or mechanical tamper. 

Some EWF marketing literature reports that the surface material will naturally settle and compact over time and with visitor use.  Observations at new installations with heavy visitor use indicate the high traffic may actually create even greater peaks and valleys in the undulating surface.  At the sites where the surface material has had the opportunity to naturally settle, several occurrences were noted where there were changes in level greater than .50 inches at the point of entry to the playground from the sidewalk or at transitions with unitary surfaces. 

Large areas where the loose material had been displaced under heavy use areas with motion such as at swings, slides, sliding poles, climbers, spinners, and teeter totters were observed at all of the sample sites with EWF.  A kick-out area at a swing could be as large as 3 ft. x 8 ft. with a depth of more than 5 inches.  The accessibility standards require the minimum 30 x 48 inch clear floor space for transfer to/from the accessible play components have a level surface with less than a 2.08 percent cross slope in all directions.  The displaced surface material at locations such as the bottom of slides, a swing, or ground level play component rendered the accessible route to the play component non-compliant with the accessibility standards. Maintenance issues at sites began to emerge where the product was filled at the kick-out area rather than the raked level, compacted and then filled and compacted.  Where the kick-out areas had been filled, again it would eventually be displaced.  This time it created higher undulating mounds at the front and back of the kick-out area and greater cross slopes within the required clear floor space. 

At locations where the EWF was paired with a unitary surface, deficiencies were identified at the transition between the two surface materials.  The EWF had settled by 1‒5 inches creating a change in level of excessive running slope up to 16 percent at the transition.  This was most prevalent at sites installed with PIP as the primary access route. At locations where TIL was intended as the primary accessible route and EWF was used as secondary safety surfacing, the EWF particles began contaminating the TIL seams.

To the layman, the terms EWF and woodchips are often, incorrectly, interchanged.  The difference between EWF and wood chips is one where the EWF goes through several additional processes following the output from what would come from a typical landscape chipper.  Unlike woodchips out of the chipping equipment, EWF is then shredded again, stamped/flattened and made pliable to the extent that the particles will weave together to create a traversable, impact attenuating surface.  In addition, there is an ASTM standard specification for EWF, further distancing the material from any product made on site or purchased from a nursery or home improvement store.  The ASTM standard for EWF requires the particles be small enough to pass through a series of three sieves, ¾ inch, 3/8 inch and No. 16 (0.0469 inch).  The sample is considered compliant if there is no more than 1 percent residue is left on any individual sieve.  Large wood particle chips, chunks and shredded twigs were found at all of the EWF sample sites.  The observable quantity of large wood particles raised into question whether a test sample from any of the sites would comply with the ASTM standard specification for EWF and specifically the sieve test.  In addition to the large particles, there were instances where vegetation and mold were found growing in the surface material.

A conference call was conducted with all of the playground owners prior to the release of the first year finding report in 2011.  At this time, acquisition, installation and maintenance of EWF was thoroughly discussed.  Owners cited the ability to buy directly from mills with ASTM-IPEMA lab certificates as a process to purchase the product at lesser cost than buying directly from the manufacturer. They also stated that no installation instructions accompanied the surfaces from the manufacturer or the mill,.  Thus, none of the owners were aware of any need to install the surface materials in layers by watering and compaction. 

Sites visited in 2012 showed a marked improvement in SDS where playground owners participated in the conference call and gained greater information on the compaction requirement for installation and maintenance.  One site utilizing PIP as the primary access route and EWF as the secondary access route was assessed with less than 1 percent slope at the transition between the two surface materials.

Hybrid Surface Systems

There were a total of four sites with three different hybrid (HYB) surface systems evaluated in the study.  All three systems have been purposefully designed and marketed to provide an impact attenuating and accessible surface to accommodate both safety and accessibility.  One site used an outdoor carpet (HYB-A) over engineered carpet padding infilled with silicone sand.  Two sites used a system where the base consisted of 2 ft. x 2 ft. pillows filled with shredded rubber and covered by 5 ft. wide rubber top mats (HYB-B), resembling melted spaghetti, affixed at the seams similar to how carpet is seamed together.  The last site used an artificial turf grass system (HYB-C), similar to that used on football fields. The HYB surface systems ranged in cost from $7.50/sq. ft. to $12.65/sq. ft.

As tested within 12 months of installation, all three HYB surface systems were observed to have minimal deficiencies, comparable to the SDS with PIP.  The means for firmness, stability and the standard deviation were also comparable to the other unitary surfaces, PIP and TIL. 

The outdoor carpet (HYB-A) was installed by a professional contractor approved by the manufacturer.  The top layer outdoor carpet is laid over an engineered carpet pad. Silicone sand is filled in over the top carpet layer to contribute to its ability for impact attenuation.  The outdoor carpet installation comes in a variety of primary colors.  The surface system was purposefully selected by the owner as a pilot site to observe how the product would perform over time.  If it was deemed successful, the playground owner intended to use the product at other sites.  After 12 months of installation, the top layer carpet began to shrink and separate at the seams creating openings and changes in level greater than ½ inch.  Under the terms of the product warranty, the manufacturer returned to the site to make spot repairs where carpet remnants were inlaid at the widest gaps between the seams.  Through the fifth year of installation, the carpet layers have continued to shrink and separate at the seams.  Field testing for ASTM 1292‒99/04 was also conducted annually at the request of the playground owner.  Between years two and five, the HIC value at an 8 ft drop height went from 688 to 922.  By year five, one location was within compliance for ASTM 1292‒99/04 by less than 80 HIC.  The manufacturer has since gone out of business and the owner has been left with a product warranty that is now useless.

The rubber top mat system (HYB-B) was installed at one site entirely by the manufacturer.  Another playground owner chose to have its site installed by a combined crew of the manufacturer and park maintenance personnel.  There were no observable differences from the two installations within the first 12 months.  It was not until years three and four that the site installed entirely by the manufacturer started to show instances of separation at the seams.  The playground owner attributed this to the park maintenance personnel’s lack of knowledge for maintaining the surface system since they were not involved in the installation and did not have any other playground sites utilizing this surface system.  During year four, the playground owner requested training from the manufacturer on the specific methods to mend seams and patch sections.  By year five, the site showed improved attention to maintenance.  The site installed by the combined crew did not show any issues of surface deficiencies.  At this site, the originally planned surface was PIP.  Because of construction changes on site, the playground owner grew concerned at the increased cost for the PIP and decided to try the HYB-B as a pilot installation.  The park maintenance crew worked alongside the manufacturer to learn the process to install the surface base and top mat.  The playground owner was purposeful in this decision in order to avoid long term costs to bring the manufacturer to the site for repairs or instances where equipment might be added or moved.  The playground owner reports that maintenance personnel have grown in their knowledge working with the product and recently installed another playground utilizing the same surface system.

The artificial grass surface system (HYB-C) was installed by the manufacturer.  The two locations each have a different base.  The first site has an engineered carpet pad base and the top turf layer is filled in with rubber granules.  The second site has a base comprised of packed shredded rubber.  Neither site showed any measurable deficiencies during the first year of installation.  During year two, the second site began to show signs of traffic patterns.  These areas were not measurable with the digital level, however they should continue to be assessed over the life of the playground.  Heavy pedestrian traffic can cause the artificial grass to fall flat.  Both playground owners have planned for the manufacturer suggested maintenance.  The first playground owner purchased mechanical equipment to rake and fluff the grass.  The other playground owner brings a contractor back seasonally to do the same thing.  One issue this playground owner had not planned for was the build-up of static electricity.  During the second year site visit, the assessment team was unable to touch any of the metal playground support posts in a section of the playground without receiving a static charge.  The playground owner met the assessment team on site and reported the problem emerged at the end of the first season of use.  The playground owner contacted the manufacturer and was given a solution to apply to the surface to minimize the static electricity.  The playground owner reports at least three applications of the solution were required at the beginning of the season in order for the static electricity to dissipate.  The playground owner also reports disappointment with the manufacturer as this information was not shared with the owner during the time the system was being considered for purchase.

The playground owners for both HYB-C sites had included the requirement for third party testing for ASTM 1292‒99/04 as part of their purchase agreements.  As such, both sites were tested for impact attenuation by independent consultants following installation.  While both sites were found in compliance for ASTM 1292‒99/04, both playground owners have reported concerns for the HIC values at 8 ft drop heights and plan to closely monitor those locations.

Shredded Rubber

Shredded Rubber (SHR) is most often the byproduct of recycled tires shredded into particles, crumbs or nuggets, sized 3/8 to 7/8 inches.  The surface material used as a play safety surface has gained some popularity as a means to reuse tires.  During pilot site visits and testing of the study protocol, the research team found sites with SHR containing tramp metal.  Some studies have shown mixed findings on the health effects and environmental impact for use (California Office of Environmental Health Hazard Assessment, 2007).  There is a lack of research data on the ability of the SHR to perform as an accessible surface when installed in public park playgrounds.  A loose fill surface, similar to EWF, SHR may experience many of the same issues such as undulation of the surface area and displacement resulting in inaccessible routes for play areas.  Observational research outside of this study has shown the undulation and instability of the particles to be difficult for people with mobility impairments to traverse.  Shredded rubber manufacturers and the member association did not respond to the research team requests to participate in the NCA longitudinal study.  As such, public park playground installations with shredded rubber surfacing are not represented in this study.  The inclusion of sites surfaced with shredded rubber would have been beneficial to the study in order for researchers to compare this loose fill material with the only other loose fill material in the study, namely EWF.  A greater examination of the tramp metal in the particles may help to determine if there was an effect on the use of assistive devices such as wheelchairs, walkers, canes or crutches.  The firmness and stability of the product material could have been studied much more thoroughly at real sites and analyzed with controlled laboratory findings such as that for ASTM F1951‒99.  Comparison of the two loose fill materials, EWF and SHR, may provide playground owners with a better understanding of how compaction during installation can affect the undulation and usability of the surface in the accessible routes.  Comparison of the two materials may also give more guidance on the specific needs for maintenance of loose fill particles used for the accessible route.