Sports turf is an important area of turfgrass application. Unlike golf courses, most of the sports turf fields require more stable playing surface to support the activities of players and facilities, and to provide protection against sports injuries. The quality of playing surface is a function of turfgrass and soil media (Canaway and Baker, 1993). Surface quality is usually expressed as friction, traction, stiffness, and resilience when the interaction between the surface and player is the main concern. It also can be evaluated from the ball bounce resilience and rolling resistance or ball speed when information about the behavior of sport facilities on a playing surface is needed (Bell et al., 1985; Baker et al., 1988; McClements and Baker, 1994). Of all the qualities of playing surface, perhaps the safety of the players is the most important consideration when constructing and evaluating a sports field. Many sports injuries are related to varying degree of surface stability (Valiant, 1988; Powell and Schootman, 1993; Waddington and McNitt, 1995). To assess the risk requires knowledge from many disciplines including the sports ground mechanics, which is not well understood because of very limited research. The objective of this study was to evaluate the relative importance of sand particle size, particle-size distribution, particle shape and roundness, plant roots and root-zone water content in the stability of a sand-based sport field surface.

The study was established on an existing sand-based sport turf area at the Horticultural Research Station. The root zones were excaved to form 5 X 10 ft plots 6 inches deep. Five treatments, Hallet mason sand, Hallet concrete sand, Sidlley Proangle sand, Bunker white sand, and Hallet mason sand + 15% soil (v/v) were filled in the plots and compacted with a Whacker vibrating compactor for ten passes. The experiment design is a randomized complete block, with three replications.

Sand angle at repose, particle-size distribution, particle shape/roundness (Li, 2001) was analyzed before loading materials to the plots (Table 1). On June 25 2001, half of each plot was planted with washed sod and half seeded with Kentucky bluegrass 'Unique'. The turf has been mowed once a week at 5 cm. The turf received 159 kg ha^-1 N, 35 kg ha^-1 P, and 96 kg ha^-1 K in three months period after sodding. Water content was measured by TDR each time surface stability was evaluated. Playing surface parameters were evaluated both before and after turf establishment. The playing surface was evaluated by measuring the penetration with a penetrometer, surface hardness with a B&K 2500 vibration analyzer (Rogers and Waddington, 1990) and traction with a cleated torque wrench device (Canaway and Bell, 1986). Bulk density also was measured to determine the compaction status. Two month into the turf establishment grass root and thatch dry mass was quantified in addition to other measurements.

Adding soil to Hallet mason sand did not increase penetration resistance. Penetration resistance increased after sodding for all materials except silica sand (Table 2).

On 27 July 2001 (32 days after sodding) there was a noticeable increase in G_maxA. Thereafter, G_maxA decreased with no explanation (Table 2).

Hallet mason sand had significantly greater traction than Hallet concrete or silica sand. Adding soil to mason sand did not increase traction (Table 3).

For each sand material there was an optimum water content that maximized penetration resistance. Sand source (angularity) played a more important role in maximizing penetration resistance than adding 15% soil to mason sand.

Two months after sodding traction, measured by the torque wrench method, did not correlate with many of the plant mass or sand property measurements in this study (Table 3).

Table 1. Particle size analysis of the sand sources used in the study.

Table 2. Mechanical properties of various sand sources after establishment with washed Kentucky bluegrass sod.

G_max A-Maximum deceleration. Average of five readings measured at five different locations, one drop per location.

Table 3. Correlation coefficients between sand properties, plant mass, and playing surface parameters measured on 25 Aug. 2001, two months after sodding.

Ang.-Angle at repose at non compacted condition.

Cu -Coefficient of uniformity.

I_g -Gradation index.

I_r - Index of particle surface roughness.

Penet. Res. -Resistance of penetrometer penetration.

G_max A-Maximum deceleration. Average of five readings measured at five different locations, one drop per location.

G_max B-Maximum deceleration. Average of five readings measured at same location, five drops per location.

At this point it appears that multiple parameters, as opposed to a single type of measurement, will be required to accurately assess the stability of sand-based sport fields.