王一淳 李德颖

摘要：泥炭/砂混合物是一种广泛应用于高尔夫果岭坪床，运动场草坪，园艺盆栽，及滤水系统中的基质。 然而，对于泥炭/砂混合基质水力特性的直接测量耗时较长。试验的目的为：（1）测试符合美国高尔夫协会（USGA）推荐标准的砂与高比例泥炭混合基质的饱和导水率（Ksat）是否为层流并遵循达西法则。 （2）研究泥炭类别和有机物含量对之前建立的预测饱和导水率的多重线性回归模型的准确性的影响。试验采用3种泥炭类别即木本泥炭藓（Peat，Inc.Minnesota，USA），苔藓泥炭（Sun Gro Horticulture，Maryland，USA）和苇苔泥炭 （Dakota Peat，North Dakota，USA），并分别按照泥炭占0%，0.2%，0.4%，0.8%，1.6%，4%，8%和10% 的重量配比与纯砂混合。试验评估的模型如下：

关键词：高尔夫球场；果岭；根层基质；导水率；土壤分析

中图分类号：G 849.3；S 151.9 文献标识码：A 文章编号：1009-5500（2013）05-0066-06

收稿日期：2013-08-15； 修回日期：2013-10-09

作者简介：王一淳（1983-），女，黑龙江省佳木斯市人。

E-mail：Yichun.Wang@Live.cn

Estimating water conductivity of sand-based root zone

materials from particle size distribution：

effects of peat amendments

WANG Yi-chun，LI De-ying

（Department of Plant Sciences，North Dakota State University，Fargo，ND 58108，USA）

Abstract：Sand and peat mixtures are widely used in constructed root zones of golf course putting greens and sports fields，containerized horticulture，and water filtering systems.Direct measurement of hydraulic properties often is time consuming.One of the objectives of this study was to test if saturated water flow is laminar and obeys Darcy's law when a large amount of peat is mixed with sand that conform to the USGA specifications.Another objective was to evaluate a previously developed multiple linear regression （MLR） model for predictingsaturated water conductivity （Ksat） as affected by peat types and organic matter （OM） content.Woody sphagnum peat （Peat，Inc.Minnesota，USA），sphagnum peat moss （Sun Gro Horticulture，Maryland，USA），and reed sedge peat （Dakota Peat，North Dakota，USA） were mixed with sand at 0%，0.2%，0.4%，0.8%，1.6%，4%，8%，and 10% （w/w） in the final mixtures.The model tested was Log10（Ksat）=5.340 7-0.528 6ρb-1.2846CP-0.044 2c+0.0612φ5-0.6095φ95，with ρb as bulk density （g/cm³），CP as capillary porosity （%），c as silt content （%），and φ5，φ10，φ16，φ84，φ95 values from the particle size distribution curve representing grain size in phi （φ） unit.Briefly，φx= -log （2，d），with x representing the percentage of sand mass smaller than d in size in a traditional particle size distribution curve.Results showed that Darcy's law prevailed at hydraulic pressure gradients up to 3.Results also showed that，with exception of less humified sphagnum peat moss at >4%，the model provided fair predictions of Ksat（R² =0.74） for OM content up to 10%.

Key words：golf course；putting greens；root zone medium；water conductivity；soil analysis

INTRODUCTION

Sand and peat mixtures are widely used in constructed root zones of golf course putting greens and sports fields （Li et al，2000），containerized horticultural （Heiskanen and Rikala，1998），and water filtering systems （Tao et al，2009）.Water holding and water conductivity of such mixtures are very important properties in their application because these properties dictate irrigation，drainage，solute movement，and soil aeration.The United States Golf Association （USGA） recommends tests of particle size distribution，water retention，capillary porosity （CP） at 30 cm water suction，OM，and Ksat，and evaluations of shape/roundness of sand particles （USGA Green Section Staff，1993）.Direct measurement of hydraulic properties is time consuming.According to a survey conducted by the proficiency test program，the confidence interval for particle size analysis is +/-10 to +/- 35%，and that for saturated water conductivity is +/- 20% using the USGA specified procedures （Miller and Amacher，2001）.Inconsistency of these test results between and within the laboratories may cause inconvenience in bidding and contract during the construction and management of a golf course.Predicting saturated water conductivity from some basic and more accurate analysis is an incentive driven alternate approach.

Many models for estimation of hydraulic properties do not include OM content as a predictor （Hazen，1893）.When OM is considered，very often it is treated as clay-sized particles （Arya and Paris，1981；Carman，1956；Childs and Collis-George，1950；Fair and Hatch，1933；Millington and Quirk，1959）.However，peat and other organic materials used in sand root zone mixtures are fibrous rather than layer-silicates.Predicting water conductivity from basic soil properties using multivariate analyses and MLR have been attempted （Brakensiek，et al.，1984；Puckett，et al.，1985；Campbell，1985；Saxton et al.，1986；Vereecken，et al.，1990；Jabro，1992；Sperry and Peirce，1995）.Model evaluations also have been conducted by many authors （Tietje and Hennings，1996；Zhang et al.，2000）.There have been no thorough model comparisons for sand-predominant soils that are used for sports fields.Li et al.（2008） developed a step-wise MLR model to predict Ksatof sand-based root zone materials from known physical properties including bulk density，capillary porosity，clay content，and particle size distribution.The 292 samples were collected from commercial laboratories representing 200 locations from over 40 states in America and two provinces of Canada，with peat content ranging from 0 to 1.2 % （w/w） （Li et al.，2008）.The model is：

Log10（Ksat）=5.340 7-0.5286ρb-1.2846CP-0.044 2c+0.0612φ5-0.6095φ10+0.085φ95[1]

where ρb is bulk density （g/cm³），CP is capillary porosity （%），c is silt content，and φ5，φ10，φ95 values from the particle size distribution curve for grain size in phi（φ） unit.Notice that OM is not included in this model.This may be because of the low content （0% to 1.2%） in common sand/peat mixtures of golf course putting green root zones，or collinearity between those predicting variables.

Organic matter tends to accumulate as the sand-based root zones age （McClellan et al.，2007；Wang et al.，2013）.Horticultural container mixes and water filtering systems use OM in high percentages.Therefore，a robust model is needed for predicting water conductivity under these conditions.Taylor et al.（1997） reported that water infiltration rate was as high as 1.03 m/h for fine sand with up to 2.98% reed sedge peat by weight.One of the objectives of this study was to evaluate adequacy of the MLR model developed by Li et al.（2008） for predicting water conductivity based on particle size analysis of sand-based root zone materials with different peats and a wide range of OM content.Another objective was to test if the saturated water flow is laminar and obeys Darcy's law in a porous medium of sand that conform to the USGA specifications in mixture with a wide range of peat type and ratio.

MATERIALS AND METHODS

Samples and Physical Properties

Woody sphagnum peat （Peat，Inc.Minnesota，USA），sphagnum peat moss （Sun Gro Horticulture，Maryland，USA），and reed sedge peat （Dakota Peat，Minnesota，USA） were thoroughly mixed with sand that has a particle size distribution conform to the USGA specifications.Weight percentages of peat in the final mixture were 0%，0.2%，0.4%，0.8%，1.6%，4%，8%，and 10%.The mixtures were packed into brass cylinders （6 cm diam.5.4 cm i.d.） using a compactor equipped with a 1.36 kg hammer.Compaction was kept consistent by 5 drops of the hammer from a height of 305 mm （USGA Green Section Staff，1993）.There are three replicates for each mixture forming a total of 66 mixture samples.

Organic matter content was tested by the loss on ignition method.Particle size distribution for sand fractions was analyzed with the dry sieve method，and the clay fraction was analyzed with the pipette method （USGA Green Section Staff，1993）.Total porosity was calculated from：

X=1-ρbρsS-ρbρpP

where ρb is bulk density；ρs is the sand particle density；ρp is the peat particle density；s and p are sand and peat content in the mixture by weight，respectively.Particle density of sand was measured with pycnometers.Capillary porosity was calculated from the total porosity minus the volumetric water content of the samples at -30 cm suction head.Sand grain shape and roundness was visually evaluated and assigned a descriptive category.Saturated water conductivity was measured by a constant head method following Klute （1986）.

Particle size distribution curves were developed based on the cumulative percentage weight versus particle diameters in φ units.The φ units are related to nominal diameter d （mm） via：

Statistics

Regression variables in the MLR model to predict Ksat include OM content，CP，clay content，silt content，and φ5，φ10，φ16，φ50，φ84，φ 95 values of particle size distribution curve.Saturated water conductivity is of lognormal distribution，with the logarithmic mean and the logarithmic standard deviation.Therefore，Ksat data were transformed using log10 function before statistical analysis.

Stepwise regression with forward，backward，and stepwise methods was used in the Procreg procedure of SAS （9.1） package （SAS Institute Inc.，Cary，NC，USA）.The stepwise regression process was also compared with the Robustreg procedure for outlier and leverage identification.

RESULTS AND DISCUSSION

From the 1∶1 line of measured Ksat values plotted against the predicted values using model（Arga and Paris，1984），it can be seen that the previous linear regression model （Li et al.，2008） is not adequate in predicting the Ksat values of the samples that had a wide range of peat types and content due to a low R² of 0.56 （Fig 1a）.Residues after the model fit were correlated with selected variables as：

Log10 （Ksat） =0.4081+0.6403CP-0.0694OM（2）

This means that the variability may change with OM types and content.A close observation of data points revealed that all outliers are sphagnum peat moss mixtures with OM at more than 4% （Fig.1a）.The predicted and measured 1∶1 line improved to R² of 0.74 （Fig.1b） after the removal of 6 sphagnum peat moss data points.Sphagnum peat moss is the least decomposed OM of three peats，which has the highest OM content and lowest bulk density （BK） （Table 1）.After the removal of those sphagnum peat moss data points，the model provided fairly adequate prediction of Ksat values of 60 samples that had OM content up to 10%.

Organic matter content is not included in the MLR model by Li et al.（2008）.This is probably because OM is correlated with BK，CP of the root zone mixtures.When combining the data from this study with that from Li et al.（2008） （Table 2），a new step

Fig.1 Comparison of the predicted saturated hydraulic conductivity based on a multiple linear model （Li et al.，2008） with measured value.a） Outlier data points of sphagnum peat moss indicated in shade.b） Outlier data points of sphagnum peat moss are removed.

wise MLR model was developed as follows：

Log10 （Ksat） =4.961-0.807ρb-1.178CP-0.037OM-0.010 9φ5-0.201 9φ84+0.180φ95 （3）

Again，this new model is not adequate in predicting Ksat，with an R² of 0.46，which is relatively low.However，it was shown that with the inclusion of OM，clay content was dropped as one of the predictors.Further analysis showed a significant correlation among OM，CP，BK，and Clay （Table 3），indicating that CP and BK may be more powerful predictors in the model.

Table 1 Physical properties of peats and sand used in the study

Table 2 Descriptive statistics for the data set from 66 sand/peat mixture samples

Table 3 Pearson correlation coefficients

（n=348，P<0.0001）

The above MLR model showed a negative correlation between Ksat and OM content.This result was consistent with the study by Nemes et al.（2005） that three of their studied Pedo-transfer models which include OM as one of the predictors negatively correlate Ksat with OM.However，other studies also found positive correlation between OM and Ksat contributing to the effect of OM on soil aggregates and pore distribution （Lado and Ben，2004）.However，the contribution of soil aggregates may be negligible in sand-based root zone materials due to low clay content.

Saturated water flow through peat has been reported to deviate from Darcy's law （Hemond and Goldman，1985）.In this study，flux density was linear with hydraulic pressure gradient from 0.5 to 3 （Fig.2），indicating that water flow was laminar and that Darcy's law prevailed for the sand/peat mixtures.Wallach et al.（1992） reported that the flux density of coarse scoria was linear with hydraulic gradients up to 1.5.Further study is needed to find the peat content in sand/peat mixtures at which Darcy's law fails.In conclusion，results showed that Darcys law prevailed at hydraulic pressure gradients up to 3.Results also showed that，with exception of less humified sphagnum peat moss at >4%，the model provided fair predictions of Ksat （R² = 0.74） for OM content up to 10%.

Fig.2 Saturated hydraulic conductivity of sand/peat mixtures tested at different hydraulic pressure gradients.Woody sphagnum peat，sphagnum peat moss，and reed sedge peat were mixed with sand at 0%，0.2%，0.4%，0.8%，1.6%，4%，8%，and 10% （w/w） in the final mixtures.

REFERENCES

Arya，L.M.and J.F.Paris.1981.A physicoempirical model to predict the soil moisture characteristic from particle-size distribution and bulk density data.Soil Sci.Soc.Am.J.45：1023-1030.

Brakensiek，D.L.，W.J.Rawls，and G.R.Stephenson.1984.Modifying SCS hydrologic soil groups and curve numbers for rangeland soils.Annual meeting ASAE Pacific Northwest Region.Kennewick，WA.

Campbell，G.S.1985.Soil physics with basic.Transport models for soil-plant systems.Development in Soil Science 14.Elsevier Science Publishing Company Inc.New York，NY 10017 USA.

Carman，P.G.1956.Flow of gases through porous media.Academic Press Inc.New York，NY.

Childs，E.C.and N.Collis-George.1950.The permeability of porous materials.Proceedings of the Royal Society of London.Series A，Mathematical and Physical Sciences 201：392-405.

Fair，G.M.and L.P.Hatch.1933.Fundamental factors governing the streamline flow of water through sand.J.American Water Works Association 25：1511-1565.

Hazen A.1893.Some physical properties of sands and graves.24th Annual report of the State Board of Health of Massachusetts.Wright & Potter Printing Co.，State Printers，18 Post Office Square.

Heiskanen，J.，and R.Rikala.1998.Influence of different nursery container media on rooting of Scots pine and silver birch seedling after transplanting.New Forests 16：27-42.

Hemond，H.F.，and J.C.Goldman.1985.On non-Darcian water flow in peat.J.Eco.73：579-584.

Jabro，J.D.1992.Estimation of saturated hydraulic conductivity of soils from particle size distribution and bulk density data.Trans.ASAE.35：557-560.

Klute，A.and C.Dirksen.1986.Hydraulic conductivity and diffusivity：Laboratory methods.687-734.in：A.Klute （ed） Methods of soil analysis.Part 1.Agronomy 9.ASA and ASSSA.Madison，WI.

Lado M，Paz A，Ben-Hur M .2004.Organic matter and aggregate-size interactions in saturated hydraulic conductivity.Soil Sci.Soc.Am.J.68，234-242.

Li，D.，D.Kuehl，and W.Fang.2008.Estimating water conductivity of sand-based root zone materials from particle size distribution：evaluation of models.Acta Hort.783：113-145.

Li，D.，Y.K.Joo，N.E.Christians，and D.D.Minner.2000.Inorganic soil amendment effects on sand-based sports turf media.Crop Sci.40：1121-1125.

McClellan，Ty.A.，R.C.Shearman，R.E.Gaussoin，M.Mamo，C.S.Wortmann，G.L.Horst，and D.B.Marx.2007.Nutrient and chemical characterization of aging golf course putting greens：establishment and rootzone mixture treatment effects.Crop Sci.47：193-199.

Miller，R.O.，and J.K.Amacher.2001.Laboratory performance of root zone test methods.Abstract ASA-CSSA-SSSA Annual Meeting，October 21-25，2001.Charlotte，NC.

Millington，R.J.and J.P.Quirk.1959.Permeability of porous media.Nature 183：387-388.

Nemes，A.，W.J.Rawls，and Y.A.Pachepsky.2005.Influence of organic matter on the estimation of saturated hydraulic conductivity，Soil Sci.Soc.Am.J.69 （4）：1330 1337.

Puckett，W.E.，J.H.Dane，and B.F.Hajek.1985.Physical and mineralogical data to determine soil hydraulicproperties.Soil Sci.Soc.Am.J.49：831-836.

Saxton，K.E.，W.J.Rawls，J.S.Romberger，and R.I.Papendick.1986.Estimating generalized soil-water characteristics from texture.Soil Sci.Soc.Am.J.50：1031-1036.

Sperry，J.M.and J.J.Peirce.1995.A model for estimating the hydraulic conductivity of granular material based on grain shape，grain size，and porosity.Ground Water 33：892-898.

Tao，J.，K.M.Mancl.，and O.H.Tuovinen.2009.Treatment of sanitary sewer overflow with fixed media bioreactors.Applied Engineering in Agriculture 25：39-43.

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

Tietje，O.and V.Hennings.1996.Accuracy of the saturated hydraulic conductivity prediction by pedo-transfer functions compared to the variability within FAO textural classes.Geoderma.69：71-84.

USGA Green Section Staff.1993.USGA recommendations for a method of putting green construction.USGA Green Sect.Rec.31（2）：1-3.

Vereecken，H.，J.Mass，and J.Feyen.1990.Estimating unsaturated hydraulic conductivity from easily measured soil properties.Soil Sci.149：1-12.

Wallach，R.，F.F.da Silva，and Y.Chen.1992.Hydraulic characteristics of tuff （scoria） used as a container medium.J.Amer.Soc.Hort.Sci.117：415-421.

Wang，Y.，W.Fang，D.Li，and R.Gaussoin.2013.Soil chemical properties of golf greens affected by age，initial construction and establishment.Inter.Turfgrass Soc.Res.J.12：539-544.

Zhang，J.，K.Nakayama，G.R.Yu，and T.Miyazaki.2000.Scaling of saturated hydraulic conductivity：A comparison of models.Soil Sci.165：718-727.