PDF version
for printing

Impact
Statement

ฉ 2007 Plant Management Network.
Accepted for publication 26 September 2007. Published 19 November 2007.

Root Zone Composition Effects on Putting Green Soil Water

E. L. McCoy, P. Kunkel, G. W. Prettyman, and K. R. McCoy, School of Environment and Natural Resources, OARDC, The Ohio State University, Wooster 44691

Corresponding author: Ed McCoy. mccoy.13@osu.edu

McCoy, E. L., Kunkel, P., Prettyman, G. W., and McCoy, K. R. 2007. Root zone composition effects on putting green soil water. Online. Applied Turfgrass Science doi:10.1094/ATS-2007-1119-02-RS.

Abstract

A field study was conducted to examine sand texture and root zone amendment effects on soil water fate, focusing particularly on turfgrass water use. Six root zone treatments were used, each with a depth of 0.3 m overlying a 0.1-m thick gravel layer. Two were 100% sand where the sands were relatively finer and coarser, two used the same sands blended to contain 90% sand and 10% sphagnum peat (by volume), and the final two root zones were blended by adding 10% soil to the sand:peat mixes. While supporting ‘Penncross’ creeping bentgrass (Agrostis palustris Huds.), a water balance accounting was conducted during the growing seasons of 2000 and 2001. This involved daily measurement of rainfall, irrigation, drainage volume, and soil water contents with turfgrass evapotranspiration (ET) determined as the change in water content within the root zone plus rain or irrigation depths and minus the drainage depths. Turfgrass rooting was measured in the first week of October each year, and water retention curve and saturated hydraulic conductivity measurements were conducted in the fall of 2001. Overall, the results of the field experiment showed that increasing amendment amounts to either sand yielded significantly greater water retention, reduced saturated hydraulic conductivity, and greater root zone water contents; however, the sand and amendment treatments had little consistent effect on turfgrass response as judged by actual ET and rooting measurements.

Introduction

Peat and soil are commonly used amendments in high sand root zone mixes for putting greens. Extensive laboratory research has shown that whereas sand texture plays a dominant role in the hydraulic properties of a root zone, measurably increased water retention and frequently reduced saturated hydraulic conductivity occur from the addition of modest quantities of peat, soil, or both (4,5,11,18,19,22,23). In a sense, differential peat and soil amendments to a particular sand serve to fine tune the balance between water retention and transmission properties of a root zone.

Following the extensive study by Waddington et al. (21) field-scale research has generally confirmed but also extended the lab-based physical property implications (2,3,6,7,14,16). Some of these additional field study observations include: the role of organic matter accumulation in the surface layer of established greens that increases water retention and reduces water transmission, often to a much greater extent than amendment addition (3,7); and the fact that naturally sloped greens can lead to spatially changing water contents due to subsurface lateral flow (6,16).

The objective of this study was to further refine the field-scale knowledge base on putting green soils by examining sand texture and root zone amendment effects on soil water fate, focusing particularly on turfgrass water use.

Examining Root Zone Composition Effects on Soil Water Fate

Six root zone treatments were used each with a depth of 0.3 m overlying a 0.1 m thick gravel layer. Two of the root zones were 100% sand where the sands were relatively finer and coarser as based on USGA guidelines (20). Two root zones were blended to contain 90% sand and 10% sphagnum peat (by volume) using the finer and coarser sands, and the final two root zones were blended by adding 10% soil to the sand:peat root zone again using the coarser and finer sands. The particle size distribution and organic matter contents of the root zones are shown in Table 1. The finer sand had a Fineness Modulus value of 1.9 and a D₉₀/D₁₀ uniformity coefficient value of 4. The coarser sand had a Fineness Modulus of 2.2 and a D₉₀/D₁₀ uniformity coefficient value of 5. The gravel was selected based on USGA recommendations (20) for a two-tier profile and using the particle size distribution of the finer sand:peat:soil root zone. This assured that the same gravel material was acceptable for all remaining root zones. Water retention measurements and coefficient values of a water retention curve were determined for these construction mixtures following the procedures of McCoy and Stehouwer (13). Saturated hydraulic conductivity was measured using the constant head technique (13). Each root zone treatment was replicated 3 times for a total of 18 experimental greens. The treatments were arranged in a randomized complete block design.

Table 1. Particle size distribution and organic matter contents of the root zone mixtures and gravel used in experimental greens construction.

 * not determined.

This water balance and turfgrass water use study required a complete accounting of all water inputs and outputs from the root zones. For this reason, the greens soil profile was constructed within 1.8-m diameter non-weighing lysimeters where drainage from individual greens was directed to an adjacent pit. Rainfall was recorded from a rain gage adjacent to the site, and in 2001, reference evapotranspiration (ET) was measured using three ETgages (ETgage Co., Loveland, CO). Irrigation inputs to each green were recorded by placing collection tins on each green prior to an irrigation event. Soil water content was measured using a TRASE-BE (Soilmoisture Equipment Corp., Santa Barbara, CA) connected to buriable and horizontally placed TDR probes located at 0.076-, 0.15-, and 0.23-m depth (one probe at each depth per plot). A National Climate Data Center AB COOP weather station was located 0.6 km from the experimental site and included Class A Pan evaporation measurements. The greens were seeded to ‘Penncross’ creeping bentgrass (Agrostis palustris Huds.) in the spring of 1998 and maintained at a mowing height of 4 mm.

During the growing season of 2000 and 2001, measurements of rainfall, drainage volume, soil water contents, and ETgage evaporation (year 2001 only) were collected daily. Turfgrass ET was determined using a water balance approach as the change in water content within the root zone plus rain or irrigation depths and minus the drainage depths. Keely and Koski (9) criticized the use of TDR probes in ET studies, yet they also observed roots of the 6.3-cm height-of-cut turfgrass to extend below the maximum TDR measurement depth in a native, sandy clay soil. In this study, however, the close-mown turfgrass and the presence of gravel under the root zones should serve to limit rooting to within the TDR measurement region. During periods where it was intended to maintain the turfgrass in a well watered condition, irrigation was applied as needed (judged by prior rainfall and drainage measurements), no more than 3 times per week, and at average amounts of 7.5 mm.

The first week in October in 2000 and 2001, turfgrass roots were recovered from 32-mm diameter cores sampled from 25- to 75-mm and 155- to 205-mm depth increments within each green. These shallower and deeper sampling depths were deemed adequate to evaluate treatment effects on turfgrass rooting. Also, all disruptive sampling, including those for turf rooting, were not performed till the end of the season to preserve the hydraulic integrity of the experimental green surface. The root samples were dried at 75ฐC and weighed. At the conclusion of the overall study in 2001, 55-mm diameter by 30-mm undisturbed soil cores were collected to a depth of 76 mm, water retention measurements were conducted and the results fit to a water retention curve, all as described by Prettyman and McCoy (16). We did not collect undisturbed cores for water retention below 76 mm because we did not anticipate water retention properties deeper in the root zone to be greatly different than the construction mixes (7), and because of the likely disturbance to samples collected from such depth. Also, field measurements of saturated hydraulic conductivity were conducted at this time in 2001 using a Guelph permeameter (Soilmoisture Equipment Corp., Santa Barbara, CA) and well depths of 60 and 120 mm. Calculation of saturated conductivity from the Guelph permeameter measurements were by the shallow well pump-in method [equation 12 in Amoozegar and Warrick (1)].

Sand Texture, Root Zone Amendments, & Turfgrass Water Use

The root zones used in this study principally contained medium-coarse sand with the finer sand root zones containing roughly equal amounts of these two fractions and the coarser sand root zones containing about 1.8 times coarse sand than medium (Table 1). Also, total sand contents ranged from 92 to 98% and organic matter content ranged from 0.5 to 2% by weight. Consequently, the composition of these root zones would be characteristic of that commonly employed in putting green construction.

Water retention curve coefficients (Table 2) and saturated hydraulic conductivity values (Table 4) demonstrated that the various construction mixtures had significantly different soil water properties. These differences were principally noted for the amendment treatments where adding peat or peat and soil resulted in increased saturated water contents and the quantity of water retained at larger soil water suction values [via the intermediate water content, θ_i, and curve shape α₂ and n₂ values (13)]. The amendment treatments also reduced saturated hydraulic conductivity values (Table 4). Sand texture of the construction mixes, however, had little influence on water retention and saturated hydraulic conductivity results. This was expected due to the narrow range of sand textures used in this study. These results are in agreement with McCoy and McCoy (12), who observed that a field estimate of available water for these same root zones showed progressive increases with added peat and soil but was unaffected by sand texture.

Table 2. Mean water retention curve coefficients of the construction mixtures. The terms θ_s, θ_r, and θ_i are the saturated, residual and intermediate water contents; and the a and n terms are curve shape parameters (13).

 * Significant at the 0.05 level of probability.

**Significant at the 0.01 level of probability.

 † Not significant.

 ‡ Least significant difference (P = 0.05) of the sand ื amendment interaction.

Water retention curve coefficients from undisturbed root zone samples collected in the fall of 2001 (Table 3) again showed treatment responses primarily due to the presence and quantity of added amendments. This response was similar to that of the construction mixes, where added peat or peat and soil increased water retention (via θ_i values). Yet the 2001 results were unlike those of the construction mixtures in that the shape of the water retention curves at larger suctions (via α₂ and n₂ values) were not affected by amendment. This suggests that about 3.5 years after establishment there was little effect of the type of amendment (i.e., either peat or soil) used, and only the increasing amount of amendment from these various mixtures retained an effect. Also, as with the construction mixtures, sand texture had little effect on water retention of the 2001 samples. Finally, the field-measured saturated hydraulic conductivity values of 2001 showed a similar treatment response as the construction mixtures (Table 4). Whereas the amendment treatments expectedly reduced hydraulic conductivity values, sand texture did not significantly affect the results of these field measurements. The significant sand by amendment interaction term in the hydraulic conductivity analysis (Table 4) may be considered an artifact due to an error in blending where the finer sand plus peat and soil root zone contained lower than expected organic matter contents (Table 1).

Table 3. Mean water retention curve coefficients of undisturbed root zone samples collected in the fall of 2001. The terms θ_s, θ_r, and θ_i are the saturated, residual and intermediate water contents; and the a and n terms are curve shape parameters (13).

 * Significant at the 0.05 level of probability.

**Significant at the 0.01 level of probability.

 † Not significant.

 ‡ Least significant difference (P = 0.05) of the sand ื amendment interaction.

Table 4. Mean saturated hydraulic conductivity of construction mixtures and from field measurements conducted in the fall of 2001.

 * Significant at the 0.05 level of probability.

 † Not significant.

 ‡ Least significant difference (P = 0.05) of the sand ื amendment interaction.

Nearly 4 years after seeding, Baker et al. (3) observed infiltration rates of about 100 mm/h with marginal effects of sand type, amendment, and amendment incorporation rate. This contrasts with our generally greater saturated hydraulic conductivity values that were more strongly influenced by amendment incorporation. The difference, however, principally lies in the measurement protocol where infiltration rates of the previous study were influenced by organic matter accumulation at the surface (ranging from 9 to 11% by weight) whereas our Guelph permeameter measurements were from within the root zone, below the surface layer. Also, from using infiltration rate measurements, a 5% by volume soil replacement for the same volume addition of peat in a root zone resulted in lower infiltration rates both of the original root zone and after the greens matured (7). The rate of decline as the greens aged up to 7 years, however, was independent of root zone composition.

The field component of this study examined the influence of contrasting root zones on putting green soil water and turfgrass response. Root zone water contents ranged from 14 to 24% by volume across all treatments and measurement depths from day 235 to 258 in 2000 and day 183 to 212 in 2001 (Table 5). During both of these periods, frequent rainfall or irrigation created well watered conditions with regard to turfgrass requirements. These measurements showed that although amendments significantly influenced root zone water contents, sand texture did not. Also, as expected, larger water contents were observed at increasing sampling depths. The pattern of amendment effects was also mostly as expected where increasing amendment amounts resulted in larger water content values at all sampling depths. An exception to this pattern is the finer sand plus peat and soil treatment combination where water contents were lower than the finer sand plus peat alone. This is explained by the unexpectedly low organic matter values of this root zone (Table 1) suggesting that some mistake in blending may have produced the observed water content result. This also likely produced the significant sand by amendment interaction term seen in Table 5.

Table 5. Mean root zone water contents from day 235 to 258 in 2000 (18 sampling dates) and from day 183 to 212 in 2001 (21 sampling dates).

 * Significant at the 0.05 level of probability.

**Significant at the 0.01 level of probability.

 † Not significant.

 ‡ Least significant difference (P = 0.05) of the sand ื amendment ื depth interaction.

The water content results presented here were in general agreement with previous field studies using similar sand textures, amendments, and amendment rates (2,3,14). A significant sand texture effect, yielding greater water contents in finer textured sands, was observed by Baker and Richards (2) and Baker et al. (3). Yet these two previous studies employed sands having a greater contrast in particle sizes than those in the present study. So whereas sands classified as medium-coarse or medium showed water content differences in well watered conditions (3), those principally falling within the medium-coarse range, as in our case, did not.

In separately comparing soil and peat as amendments, Baker et al. (3) observed little difference in water content throughout the root zone but did show a stronger mixing ratio effect with amendment incorporation between 10 and 30% by volume. Murphy et al. (14), on the other hand, showed greater surface water contents for sphagnum than loam when incorporated at 20% by volume, but not when incorporated at 5% by volume. Although individual amendment properties can clearly have an effect on water contents within root zones, it may be useful to generalize that peat or soil have similar effects on the water content distribution within root zones when combined at modest levels (i.e., 10% volume addition).

During these same periods, when the turfgrass was maintained in a well-watered condition, actual evapotranspiration (Table 6) averaged across all treatments equaled 3.8 cm for 2000 (25 days) and 8.6 cm for 2001 (32 days). These values correspond to weekly average actual ET rates of 10.6 mm/week for 2000 and 18.8 mm/week for 2001. And in comparison with class A pan ET measurements for the same period yielded actual ET at 37% of class A pan ET for year 2000 and 42% of pan ET for year 2001. Also, in 2001 actual ET was 60% of the ETgage measurements.

Table 6. Mean cumulative turfgrass ET from day 233 to 258 in 2000 (17 sampling dates) and from day 183 to 215 in 2001 (22 sampling dates). For this same period in year 2000, the Pan ET value was 10.2 cm and in year 2001 was 20.7 cm. Mean ฑ standard error ETgage measurements in 2001 were 14.4 ฑ 0.5.

 * Significant at the 0.05 level of probability.

**Significant at the 0.01 level of probability.

 † Not significant.

 ‡ Least significant difference (P = 0.05) of the sand ื amendment interaction.

The correspondence between our actual and reference ET was low in comparison with other research using close-mown turfgrass. Using a mid-summer conversion value of 1.3 for the ratio of class A pan ET to modified Penman ET (15) our evaporation rates were estimated to be 50% of the reference turfgrass value. This compares with 65% determined by Lodge and Baker (10) and 98% by Sass and Horgan (17). One essential difference between the present study and those cited above were that we estimated ET via a water balance equation using water content changes as determined by TDR sensors whereas the other studies employed a more direct weighing lysimeter approach to track daily water gains or losses. Even though the TDR sensor measures the average water content within a cylindrical volume of soil surrounding the probe, our 0.076-m measurement may have missed some water content changes occurring at the surface. A second difference was that the previously reported studies employed daily and light irrigation that, as discussed by Sass and Horgan (17), may have principally wetted the canopy and thatch leading to subsequent high rates of water loss. By irrigating less often a greater portion of the added water enters the soil, is taken up by root in support of turfgrass transpiration, and leaves the soil-plant system at a lower rate.

With regard to sand texture and amendment effects on actual ET, the differences were mostly small for 2000 and nonexistent for 2001 (Table 6). The exception being the coarser sand plus peat and soil treatment combination that yielded approximately 1 cm greater ET losses than all other treatments in 2000 (and also yielded the largest mean for 2001). This single response likely produced the observed significance of sand, amendment, and the interaction term for year 2000. Consider however that this same treatment showed only marginally greater water contents, ranging from 2 to 7% more, in the surface layer than other treatments and that this water content difference was consistent year to year. This, together with the year to year inconsistency of actual ET treatment responses serves to discount the practical significance of the large actual ET values for the coarse sand plus peat and soil treatment in year 2000. Accepting this implication therefore implies that the sand or amendment treatments of this experiment had little influence on actual ET values.

Aside from a significant and expected depth effect, turfgrass root dry weights showed varying treatment responses in 2000 and 2001 (Table 7). In year 2000, the significant sand and sand ื depth interaction appeared to result from the coarser sand root zones having larger root weights at shallow depths than the finer sand root zones. In year 2001, the significant amendment effect came from root mass reduction at both depths with increasing amendment rates. The amendment effects of 2001 could suggest that either wet soil conditions were limiting root growth in the higher amended root zones or that dry soil conditions in the un-amended root zones were forcing the turfgrass to invest more resources into root system expansion in order to tap limited water resources. The water content results (Table 5) suggest that the latter explanation is the better of the two. During the well watered period of 2001, amended root zone water contents ranged from 20 to 24% by volume leaving sufficient air-filled pore space for adequate soil aeration. Water contents of the un-amended root zones, however, were significantly drier ranging from 14 to 20% by volume.

Table 7. Mean turfgrass root dry weight from field sampling in the fall of 2000 and 2001.

 * Significant at the 0.05 level of probability.

**Significant at the 0.01 level of probability.

 † Not significant.

 ‡ Least significant difference (P = 0.05) of the sand ื amendment ื depth interaction.

When examining amendment and rate effects on creeping bentgrass rooting one year after establishment, Murphy et al. (14) observed the general relationship of lower root mass with root zones having greater water storage, even though these same root zones produced high turfgrass quality. From measurements conducted in December through September, Hannaford and Baker (8) observed deeper rooting in a medium-coarse sand than with a medium sand, and deeper rooting with peat amendment than with soil amendment. In both of these treatment contrasts, the greater rooting depth was associated with root zones containing a greater volume of macropores and lower water-holding capacity. Thus, our implication that drier soils lead to more expansive rooting is in qualitative agreement with previous explanations of rooting within high sand content putting greens (8,14). Although the maximum rooting depth of cool season grasses varies throughout the season, there is no evidence to suggest that these changes would yield substantially different treatment responses than those observed at a single sampling time.

Overall, the results of the field experiment showed that increasing amendment amounts to either sand yielded significantly greater water retention, reduced saturated hydraulic conductive, and greater root zone water contents, but the sand and amendment treatments had little consistent effect on turfgrass response as judged by actual ET and rooting measurements. Finally, these root zone treatments failed to yield any visual differences in turfgrass quality.

Acknowledgments

This research was supported by funds received from the US Golf Association, the Golf Course Superintendents Association of America, and the Ohio Turfgrass Foundation.

Literature Cited

1. Amoozegar, A., and Warrick, A. W. 1986. Hydraulic conductivity of saturated soils: Field methods. Pages 735-770 in: Methods of Soil Analysis, Part 1. 2nd Edn. Agron. Monogr. 9. A. Klute, ed. ASA and SSSA, Madison, WI.

2. Baker, S. W., and Richards, C. W. 1993. Rootzone composition and the performance of golf greens. III. Soil physical properties. J. Sports Turf Res. Inst. 69:38-48.

3. Baker, S. W., Mooney, S. J., and Cook, A. 1999. The effects of sand type and rootzone amendments on golf green performance. I. Soil properties. J. Turfgrass Sci. 79:2-17.

4. Brown, K. W., and Duble, R. L. 1975. Physical characteristics of soil mixtures used for golf green construction. Agron. J. 67:647-652.

5. Davis, W. B., Paul, J. L., Madison, J. H., and George, L. Y. 1970. Evaluating sands and amendments used for high trafficked turfgrass. Ext. Serv. Publ. No. AXTn113. Univ. of Calif., Davis.

6. Frank, K. W., Leach, B. E., Crum, J. R., Rieke, P. E., Leinauer, B. R., Nikolai, T. A., and Calhoun, R. N. 2005. Effect of rootzone material and depth on moisture retention in undulating USGA putting greens. USGA Turfgrass Environ. Res. 4:1-9.

7. Gaussion, R., Shearman, R., Wit, L., McClellan, T., and Lewis, J. 2006. Soil physical and chemical characteristics of aging golf greens. USGA Turfgrass Environ. Res. 5:1-11.

8. Hannaford, J., and Baker, S. W. 2000. The effect of rootzone composition and compaction on root development in sand-dominated golf green profiles. J. Turfgrass Sci. 76:24-36.

9. Keeley, S. J., and Koski, A. J. 1997. Measuring turfgrass evapotranspiration rates by time-domain reflectometry: A field assessment. Int. Turfgrass Soc. Res. J. 8:1280-1290.

10. Lodge, T. A., and Baker, S. W. 1992. Soil moisture content and evapotranspiration rates of three types of golf green construction in response to different rates of irrigation. J. Sports Turf Res. Inst. 68:104-113.

11. McCoy, E. L. 1992. Quantitative physical assessment of organic materials used in sports turf rootzone mixes. Agron. J. 84:375-381.

12. McCoy, E., and McCoy, K. 2005. Putting green root zone amendments and irrigation water conservation. USGA Turfgrass Environ. Res. 4:1-9.

13. McCoy, E. L., and Stehouwer, R. C. 1998. Water and nutrient retention properties of internally porous inorganic amendments in high sand content root zones. J. Turfgrass Manage. 2:49-69.

14. Murphy, J. A., Samaranayake, H., Honig, J. A., Lawson, T. L., and Murphy, S. L. 2005. Creeping bentgrass establishment on amended-sand root zones in two microenvironments. Crop Sci. 45:1511-1520.

15. Nokes, S. E. 1995. Evapotranspiration. Pages 91-132 in: Environmental Hydrology. A. D. Ward and W. J. Elliot, eds. CRC Press, Boca Raton, FL.

16. Prettyman, G. W., and McCoy, E. L. 2003. Profile layering, root zone permeability and slope affect soil water contents during putting-green drainage. Crop Sci. 43:985-994.

17. Sass, J. F., and Horgan, B. P. 2006. Irrigation scheduling on sand based creeping bentgrass: Evaluating evapotranspiration estimation, capacitance sensors, and deficit irrigation in the Upper Midwest. Online. Applied Turfgrass Science doi:10.1094/ATS-2006-0330-01-RS.

18. Taylor, D. H., and Blake, G. R. 1979. Sand content of sand-soil-peat mixtures for turfgrass. Soil Sci. Soc. Am. J. 43:394-398.

19. Taylor, D. H., Williams, C. F., and Nelson, S. D. 1997. Water retention in root-zone soil mixtures of layered profiles used for sports turf. HortScience 32:82-85.

20. US Golf Association Green Section Staff. 1993. USGA recommendations for a method of putting green construction. USGA Green Sect. Rec. 32:1-3.

21. Waddington, D. V., Zimmerman, T. L., Shoop, G. J., Kardos, L. T., and Duich, J. M. 1974. Soil modification for turfgrass area: I. Physical properties of physically amended soils. Agric. Exp. Stn. Prog. Rep. 337. Penn. State Univ., University Park, PA.

22. Whitmyer, R. W., and Blake, G. R. 1989. Influence of silt and clay on the physical performance of sand-soil mixtures. Agron. J. 81:5-12.

23. Zhang, J., and Baker, S. W. 1999. Sand characteristics and their influence on the physical properties of rootzone mixes used for sports turf. J. Turfgrass Sci. 75:66-73.