This article was downloaded by: [Moraine Comm College]On: 11 November 2014, At: 01:16Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: MortimerHouse, 37-41 Mortimer Street, London W1T 3JH, UK

South African Journal of Plant and SoilPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/tjps20

The effect of soil compaction on the water retentioncharacteristics of soils in forest plantationsC W Smith a , M A Johnston b & S A Lorentz ca Institute for Commercial Forestry Research , P.O. Box 100281, Scottsville , 3209b School of Applied Environmental Sciences , Private Bag X01, Scottsville , 3209c School of Bioresource Engineering and Environmental Hydrology , Private Bag X01,Scottsville , 3209Published online: 15 Jan 2013.

To cite this article: C W Smith , M A Johnston & S A Lorentz (2001) The effect of soil compaction on the waterretention characteristics of soils in forest plantations, South African Journal of Plant and Soil, 18:3, 87-97, DOI:10.1080/02571862.2001.10634410

To link to this article: http://dx.doi.org/10.1080/02571862.2001.10634410

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose ofthe Content. Any opinions and views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be reliedupon and should be independently verified with primary sources of information. Taylor and Francis shallnot be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and otherliabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to orarising out of the use of the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

S. Afr. 1. Plant Soil 2001, 18(3)

The effect of soil compaction on the water retention characteristics Qf soils in forest plantations

C WSmith* Institute for Commercial Forestry Research, P.O. Box 100281, Scottsville 3209

M A Johnston School of Applied Environmental Sciences, Private Bag X01, Scottsville 3209

S A Lorentz School of Bioresource Engineering and Environmental Hydrology, Private Bag X01, Scottsville 3209.

Accepted 16 May 200 1

A study was carried out to evaluate the relative effects of soil compaction on the water retention characteristics of a range of soils in forest plantations in the summer rainfall regions of South Africa. In' all cases compaction resulted in the 'flattening' of the S-shaped water retentivity curve expressed on either a mass or volumetric basis. This had the implicit effect ofiowering the water release index (a log-linear plot of matric potential against water content). A clear relationship between available water capacity (AWC) and bulk density and soil type was not discernible since changes in water retentivity curves following compaction are deperident upon the complex relationship between compressive processes, soil properties and pore geometry. AWC,responded to compac­tion in three ways: (i) AWC was reduced with increasing compaction (most soils); (ii)increasing compaction resulted in increasing AWC (some sandy and clayey soils) and (iii) increasing compaction resulted in increasing AWC up to a point after which it declined. In general, field capacity (FC) increased with increasing compaction when water content was expressed on a volumetric basis but no clear trends were apparent on a mass basis. Effects of compaction on AWC depend upon the designated matric potential for FC which is often arbitrarily defined since it may vary for different s.oils and crops. Changes in pore geometry are better reflected by the expression of the water content on a mass basis without consideration of volume effects. From a practical point of view, however, changes in available water are better expressed on a volumetric basis.

Die rela~iewe effek van kompaksie op die waterretensie eienskappe van 'n reeks gronde in bosbouplantasies van die somerreenvalgebied in Suid Afrika is bestudeer. In aile gevalle het kompaksie die'afplatting' van die S-vormige waterretensiekromme tot gevolg gehad, hetsy die kromme op 'n massa- of volumebasis uitgedruk is. Dit impliseer 'n afname in die watervrylatings-indeks (d.i. die gradient van die lineere gedeelte van 'n grafiese voorstelling: matrikspotensiaal versus waterinhoud op 'n log-lineere skaal). 'n Duidelike verwantskap tussen beskikbare waterkapasiteit en brutodigtheid of grondtipe kon nie ge'identifiseer word nie omdat veranderinge in die waterretensiekurwes na kompaksie afhanklik is van die komplekse verwantskap tussen kompakterende prosesse, grondeienskappe en die vorm van grondporiee. Kompaksie het die beskikbare waterkapasitiet (BWK) op drie maniere beTnvloed: Toenemende kompaksie het (i) BWK verminder in meeste gronde; .(ii) BWK vergroot in spmmige sande~ige en kleierige gronde en (iii) BWK verhoog tot op 'n punt waarna dit verlaag is in sommige

. gronde. Met waterinhoud uitgedruk op 'n volumebasis toon veldwaterkapasiteit oor die algemeen 'n toename met toenemende k6mpaksie, maar wanneer waterinhoud op 'n massabasis uitgedruk word, is 'n duidelike ten­dens nie waarneeri,baar nie. Veldwaterkapasiteit is gewoonli~ 'n arbitrer gekose syfer omdat dit kan v~rieer tus­sen grondtipes en verskillende gewasse. Die effekvan kompaksie op BWK is egter afhanklik van hierdie gekose waardevir veldwatE:!rkapasiteit. Veranderinge in die porievolume a.g.v. kompaksie word duideliker gereflekteer deur die waterinhoud op 'n massabasis uit te druk sonder inagneming van die volumetriese effek. Van 'n prak­tieseoogpunt bly dit egter die beste om veranderinge in beskikbare water uit tedruk op 'n volumebasis.-

Keywords: Available water capacity; soil compaction; water retention; water retention curves.

* To whom correspondence should be addressed (E-mail: colin@icfr.unp.ac.za) ..

87

Introduction

Since water availability is a principal determinant of produc­tivity in forest plantations throughout the summer rainfall regions of southern Africa (Grey, Schonau & Herbert, 1987), any impact of forest management on soil physical properties may have consequences for tree growth. In particular, man­agement operations such as timber harvesting or soil tillage/ land preparation, may considerably affect the level of com­paction as expressed by Dulk density (e.g. Grey & Jacobs, 1987; Moffat, 1991). A review of the effect of soil physical

properties on soil water retention by Rawls, Gish & Braken­siek (1991) revealed that over 20 a!lthors have considered bulk. density alone or in combination as a predictor variable for estimating soil wat~r retention parameters. Furthermore, only a few authors (e.g. Gupta & .Larson, 1979; Gupta, Sharma & DeFranchi, 1989) have considered the relationship between water retention and compaction in terms of changes in pore size distribution ~nd pore geometry. Most studies are dominated by a descriptive, empirical approach in estimating the effects of bulk density on· water retention characteristics

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

88

(e.g. Arya and Paris, 1981; De Jong, Campbell & Nicholai­chuk, 1983; Rajkai et at., 1996). For simulation and hydro­logical modelling purposes the water retention parameters may be derived from a knowledge of soil texture, organic matter content and bulk density (e.g. Schulze, Hutson & Cass, 1985; Hutson, 1986). However, these authors obtained the information by multiple regression analysis and the effect of bulk density on water retention parameters was not measured per se. The effect of bulk density on water retention could only be expressed by multiplying water content on a mass basis by the bulk density to obtain volumetric water content.

Defining the effects of compaction on available water capacity is difficult since 'field capacity' (FC) is arbitrarily defined, varying with soil and crop characteristics. In addi­tion, FC is not expressed consistently in the literature either on a volumetric or a mass ba~is. In a study of the effects of soil compaction on the water retention of nine Kwa~

Zulu-Natal soils of various textures, Hill and'Sumner (1967) grouped water retention curve (WRC) responses to compac­tion into three categories based on texture: (I) sands,{{.i.i). sandy loams/sandy clay loams, and (iii) clay loams and clays: For sands and clays, compaction .increased water retention at the same matric potential. The magnitude of this effect decreased with decreasing matric potentials for sandy soils but increased with decreasing matric potential for clays. For sandy loams and sandy clay loams increasing bulk density resulted in decreased water retention at constant matric poten­tial with the reverse occurring at very low matric potentials. A problem with interpreting the data of Hill and Sumner, (1967) is that their work considered only water retention at matric potentials higher than -10 kPa and thus the generally reported 'flattening' of the WRC with increasing compaction (Gupta

,et ai., 1989; Katou, Miyaji & Kubota, 1987) was not observed. Also, hysteresis was not studied and, as Croney and Coleman (1954) have shown, the effect of compaction is generally to narrow the hysteresis loop.

The influence of bulk density on FC, taken as the water content at a matric potential of -5 kPa for a range of British soils, was studied by Archer and Smith (1972) who found that volumetric water content at -5 kPa increased with increasing compaction until a critical bulk density was,reached and then declined rapidly. Archer 'and Smith (1972) suggested that since wilting point is predominantly controlled by texture, available water capacity (A WC) varies in a manner similar to Fe. Thus a number of optimum bulk densities were identi­fied, corresponding to maximum A WC providing that aera­tion was not limiting at these bulk densities.

Reeve, Smith & Thomasson (1973) showed the relation­ship between bulk density and FC, and therefore A WC, to be more complex by considering a wider range of soil textures than those studied by Archer and Smith (1972). Volumetric water content at FC, which for all soils was taken as -5 kPa, was found to decrease with increasing compaction for all soils except' loams and clays where FC increased with increasing bulk density. Data presented by Bennie and Burger (1979) and Gupta and Larson (1979) showed that the way in which FC is modified by compaction is dependent upon the matric potential chosen for Fe. Bennie and Burger (1979) found that volumetric water content for a fine sandy loam increased with increasing 'compaction at -10 kPa matric

S. Afr. Tydskr. Plant Grond 2001,18(3)

potential but was relatively constant at -5 kPa. The purpose of this work was to gain an insight into the

effect of soil compaction on the water retention characteris­tics, in particular the available water capacity, of South Afri­can forestry soils. The information from this work will form part of a framework for understanding the resilience and sen­sitivity of forestry soils following timber harvesting opera­tions.

Materials and Methods

The soils were selected to represent a range of textures and organic carbon contents occurring in the main timber growing regions of South Africa. Soil and site information are pre­sented in Table I.

Samples were prepared following similar procedures to those of Hill and Sumner (1967) and Mirreh and Ketcheson (1972). Air-dry sieved « 2 mm) soil was used to prepare the soil cores. The mass of soil required to obtain the desired bulk density when compressed into an aluminium cylinder (75.0 mm in diameter and 50.0 mm high) was measured out. The samples were then equilibrated in a plastic bag for two days at water contents approximating the critical water content for maximum compressibility (CWCcmax ; Smith, Johnston & Lorentz, 1997). These were then compressed into the cylinder using an Instron Universal Testing Machine to bulk densities corresponding to low, medium and high levels of compaction. The lower bulk densities corresponded closely to uncom­pacted field bulk density. For the higher bulk densities, half of the sample was compressed in two increments. The first half was compressed into the cylinder and then the second half was compressed on top of that. Triplicate cylinders at three or four levels of compaction were established for each soil.

Water retention determination

Fu II water retention cu~ves, between saturation and -1500 kPa, were determined for ten soils and between saturation and -160 kPa for the remaining soils. For the soils on which, the fUll water retentivity curve was determined, water retention was determined at matric potentials of 0, -1.0, -2.5, -5.0, -7.5, -10.0, -33, -100 and -1500 kPa. The same incremental matric potentials were used for the seven soils on which water reten­tion curves were determined only to -100 kPa. All the soil cores were saturated with water for several days to ensure as complete a saturation of the soil as possible. As the original mass of air-dry soil was known, together with the dimensions of the core and the hygroscopic water content, it was possible to ascertain when the soil was close to saturation by weigh­ing.

The apparatus used varied for the different'matric poten.tial ranges. For high matric potentials the soils were equilibrated on a tension table apparatus consisting o(diatoinaceous earth over coarse sand (Smith & Thomasson, 1974). Soil water contents were recorded at matric potentials of -1.0, -2.5, -5.0, -7.5 and -'10.0 kPa, allowing a minimum of 48 hours for equi­libration at each matric potential. Water retained at matric potentials of -33, -100 and -1500 kPa was measured using pressure plate apparatus.

A WC was determined for each soil as the difference in water content between -10 kPa and -1500 kPa, and readily

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

S. Afr. J. Plant Soil 2001, 18(3) 89

Table 1 Soil identification and site characteristics of the soils studied

Soil classification Soil Organic

Site Co-ordinates Soil form and depth Soil Clay Silt carbon Parent

number Location (Latitude/Longitude) family· Soil group# (mm) texture (%) (%) (%) material

IA Hightlats 30° 15' 30"/30° 13' 18" Cartref 21 00 Typic Haplaquept 0-400 SaLm 13 17 1.42 Natal Group-

sandstone

IE 400-600 SaLm II 15 0.32

2A Hightlats 30° 13' 48"/30° 12' 24" Lusiki 2120 Aquic Kanhaplustalf 0-200 Lm 24 34 2.37 Dwyka tillite

3A Wartburg 29° 28' 12"/30° 38' 06" Inanda 1200 Ustic Kandihumult 0-200 SuCILm 30 15 2.15 Natal Group

sandstone

4A Howick 29° 27' 30"/30° 12' 36" Kranskop I 100 Humic Haplustox 0-200 SiCI 51 44 5.77 Dolerite

4B 200-450 CI 46 29 2.64

4B2 450-1200 CI 65 23 1.09

5A Pietermaritzburg 29° 40' 12"/30° 24' 12" Inanda Typic Kanhaplustult 0-350 SaCILm 40 44 3.58 Ecca Shale

6A Ifafa 30° 26' 12'"130° 36' 48" Nomanci 2200 Pachic Haplumbrept .0-600 SaClLm 27 22 3.83 Pelitic

gneiss/schist

7A Umkomaas 30° 12' 42"/30° 46' 36" Cartref 2200 Typic Haplaquept 0-300 Lm 16 34 0.95 Dwyka tillite

8A K wambonambi 28° 33' 00"/32° II' 24" Fernwood 1110 Aquic Ustipsamment 0-500 LmSa 8 6 0.30 Recent

aeo Iian sands

9A Kwambonambi 28° 31' 30"/32° 12'42" Hutton 2100 Typic Kandiustult 0-200 SaLm 12 9 0.43 Berea

sandstone

lOA Piet Retief 27° 12' 24"/30°57' 30" Nomanci 1200 Lithic Ustochrept 0-150 SaCILm 24 17 2.36 Biotite

granite

\lA Paulpietersburg 27° 13' 48"/30° 59' 00" Hutton 1200 Typic Kandiustult 0-150 SaCILm 26 10 1.49 Biotite

granite

12A Amsterdam 26° 38' 18"/30° 43' 42" Inanda 1200 Typic Kanhaplustult 0-200 SaCILm 28 16 2.42 Leucocratic

granite

13A Amsterdam 26° 32' 06"/30° 41' 54" Kranskop I 100 Humic Xanthic 0-250 CI 66 28 4.13 Gabbro,

Haplustox norite ,

14A Lothair 26° 22': 12"/30° 27' 18" Kranskop I 100 Typic Haplustox 0-300 SaCI 44 15 4.23 Granitic

I gneiss

15A Warburton 26" 12i 12"/30· 27' 12" Clovelly 1100 Lithic Haplustox 0-300 SaCILm 35 14 1.37 Ecca

sandstone

16A Barberton 25° 43', 18"/30° 50' 18" Hutton 1200 Rhodic Kanhaplustult 0-200 SaCILm 32 8 1.21 Hornblende I I

biotite ,

• Soil Classification Working Group (1991); # Soil Survey Stalf (1990)

Sa - Sand; Lm - Loam; CI - Clay; Si - Silt; SaCILm - Sandy clay loam; SaCI - Sandy clay; LmSa - Loamy sand; SiCI - Silty clay; SaLm - Sandy loam

available water (RA W) between -10 kPa and -100 kPa. The use of -I 0 kPa as the upper limit of available water was cho­sen for convenience. Cassel and Nielson (1986) have sug­gested -10 kPa as an appropriate estimation of FC in the lack of a practical alternative. Porosity was calculated assuming a particle density of 2.65 Mg m·3. Reeve et ai, (1973) have noted that for soils with less than 5% organic carbon the vari­ation in particle density is usually less than 0.1 Mg m·].

As the WRC tends to be S-shaped when plotted on a semi-log scale (Van Genuchten, 1980) a useful indicator of change in compacted soils is the water release index which is the slope at the mid-point of the WRC and is inversely related to pore size uniformity. It may be viewed as the maximum volume of water, expressed in mm m", released for a ten-foid

decrease in matric potential (O'Sullivan & Ball, 1993). This was estimated graphically from the WRC for all the soils in this study.

Results and Discussion

Effect of compaction on the WRC

To illustrate the general effects of compaction on water reten­tion, WRCs ranging from saturation to -1500 kPa are pre­sented for six soils representing a range of textures (Figure I, a - d). For clarity, water content is expressed on both a mass and a volumetric basis. Soil water retention parameters for all the soils in this study are given in Table 2 and 3.

In general, for all soils, there ,was a decrease in volumetric

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

90

(a)

- 0.70 ~

~

E M

E 0.60 --I: CI) 0.50 -I: 0 0 ... 0040 CI) -ctI

~ 0.30 (,)

'i: -CI)

E 0.20 ::::s 0 > 0.10

0

0.80

0.70 >--\ -";-C)

.:.:: C) 0.60

.:.:: -- 0.50 I: CI) -I: 0 0040 0 ... CI)

0.30 -ctI

~ tn 0.20 tn ctI :E

0.10

0.00

0

-1

iii I Ii i

S. Afr. Tydskr. Plant Grond 200 I, 18(3)

i " -10 -100

Matric Potential (kPa) Bulk Density (Mg m·3

)

1-0-0.906 -0-1.111 -6-1.251 1

-1000 -10000

i i ,: ",:i,! ,:,::,",,:!,

l \ \ : : :: I

,J ' I

1/ II I ! / :11

.J 1 i

! I:

. \ [, i

. ::

!

'I:

: .:: I iii

i j iii , . - :

! l", I", :: : !i

-10 -100

Matric Potential (kPa) Bulk Density (Mg m-3

)

1-0-0.906 ~1,111 ---6-1.251 I

-1000 -10000

Figure 1 (a) Water retentivity curves for selected forestry soils at a range of bulk densities expressed on a volumetric and mass water content basis; Kranskop silty clay - 4A. .

and mass water content with increasing compaction at higher matric potentials. Below a certain matric potential increasing compaction resulted in an increase in volumetric and mass

water content. This effect is manifest in the 'flattening' of the WRC with increasing compaction (Figure I a - d). The reduc­tion in the number of large pores and the resultant increase in

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

S. Afr. J. Plant Soil 2001, 18(3)'

(b)

,;;- 0.60

'E M

E 0.50 --I: CI) 0.40 -I: 0 0 ... 0.30 CI) -ctI

~ 0.20 (,)

"i: -CI) 0.10 E

::::s 0 > 0.00

0

o

-1

-1

-10 -100

-Matri~ Potential (kPa) . Bulk Density (Mg m-3

)

I~U51 ~1.504 -6-1.650 I

-10 -100

Matric Poten.tial (kPa) Bulk Density (Mg m-3

)

91

-1000 -10000

-1000 -10000

Figure 1 (b) Water retentivity curves for selected forestry soils at a- range of bulk densities expressed 011 a volumetric and mass water content basis; Fernwood loamy sand - 8A. - - .

rriesopores and micropores largely accounts ,for this effect I

(Hill & Sumner, 1967). Another point of interest in the flat-tening of the WRC's - with: increasing compaction' is the

'crossover' of the WRC's. The range of matric potentials at which this 'crossover' occurred depended upon soil type but was commonly higher when water content was expressed on

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

92

(C) r 0.60 , E

(')

E 0.50 -- >-C Q)

0040 -C 0 0 ... 0.30 Q) -C'tI 3: 0.20 CJ 'i: -Q)

O~ 10 E

1111)1

1·1111

::::s '0 > 0.00

0 -1

o -1

-10

, : ,

: :

: i!

i ,:

l:

: :

: :

-100

Matric Potential (kPa) Bulk Density (Mg· m-3

)

S. Afr. Tydskr. Plant Grond 2001, 18(3)

-1000

i I : :, :

., .

i i

-10000

1~1.242 -0-1.511 ~1.7651

-10 -100

Matric Potential (kPa) Bulk Density (Mg m-3

)

1~1.242 -0-1.511 ~1.7651

-1000 -10000

Figure 1 (c) Water retentivity curves for selected forestry soils at a range of bulk dersities expressed on a volumetric and mass water content basis; Hutton sandy clay loam - 16A.

a volumetric basis rather than on· a mass basis (Table 2 and 3 and Figure 1 a - d). The range of matric potentials where this feature occurs largely affects whether increasing compaction

increases or decreases A We. Expressed on a volumetric basis, this 'crossover' occurs at water contents corresponding to matric ,potentials of between -2.0 and -10.0 kPa whereas on

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

S. Afr. J. Plant Soil 200 I, .18(3)

(d)

c?' 0.70

E M

E 0.60 --C Q) 0.50 -c 0 0 ... 0.40 Q) -C'tI 3: 0.30 CJ

"i: -Q) 0.20 E

::::s 0 > 0.10

0

0.60

- 0.50 ";'

C) ~

C) 0040 ~ --c

0.30 i Q) -C I

0 I

0 0.20 I ...

Q)

! -C'tI 3: 0.10 [ en en C'tI I

:! 0.00

0

-1

•

-1

-10 -100

Matric Potential (kPa) Bulk Density (Mg m'3)

1~1.284 -<>-1.399 ~1.5891

i:

11

.

-1000

I:, II 111,i II : 'I: : : ~::: : i . ~ ~ i

: : , I

-10000

-10 -100 -1000 -10000

Matric Potential (kPa) Bulk Density (Mg m·3)

1~1.284 -<>-1.399 ~1.5891

93

Figure 1 (d) Water retentivity curves for selected forestry soils at a range of bulk densities expressed on a volumetric and mass water content basis; Cartref sandy loam - I A.-

a mass basis it occurs a~ a much wider range of matric poten­tials varying between -4 kPa and -\500 kPa, depending on soil type (Table 2 and 3).

This 'crossover' range of matric potentials is of particular interest since, on a volumetric basis and with a number of exceptions on a mass basis, it 'occurs in the range of matric

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

94 S. Afr. Tydskr. Plant Grond 2001,18(3)

Table 2 Parameters of the water retentivity curve down to -1500 kPa for selected forestry soils at a range of bulk densities

Bulk \llm at -10 kPa \11 m at -I OOkPa \llm at -1500 kl'u Crossover

density Porosity 8, 8m 8,. 8m 8,. 8m RA W· A WCh WRle \llv (volume) \1101 (maSs)

Soil torm

Soil texture

Site No. (Mg m·3) (m3 m·3) (m3 m·J ) (kg kg· l) (m3 n1"3) (kg kg-I) (m3 m·3) (kg kg-I) (mmm- I) (mm m· l) (mm n1"l) (-kPa) (-kPa)

C1"

Sandy loam

(IA)

Lli

Loam

(2A)

1.284 0.516

1.399 0.472

1.589 0.400

1.092 0.588

1.304 0.508

1.602 0.395

0.210

0.248

0.311

0.334

0.299

0.349

0.164

0.177

0.196

0.306

0.229

0.218

0.152 0.118

0.168 0.120

0.179 0.113

0.171 0.157

0.225 0.172

0.281 0.176

0.116 0.090

0.127 0.091

0.143 0.090

0.139 0.127

0.196 0.150

0.241 0.150

58

80

132

163

74

68

94

121

168

195

103

108

0.530

0.445

0.220

0.310

0.280

0.115

2-5 4-6

4-10 20-200

la 1.215 0.542 0.273

0.286

0.225

0.196

0.204

0.168 0.138 0.129 0.106 105

76

86

144

91

101

0.385

0.380

0.120

2--6 20-70

Sandy clay' loam 1.461 0.449 0.210 0.143 . 0.195 0.134

(3A)

Kp

Silty clay

(4A)

Cf

Loam

(7A)

Fw

Lflamy sand

(8A)

1.700 0.359 0.347

0.906 0.658

1.111

1.251

0.581

0.528

1.352 0.490

1.603 0.395

1.859 0.299

0.387

0.441

0.489

0.326

0.355

0.333

1.632 0.384 0.112

1.679 0.366

1.769 0.377

0.121

0.145

Hli 1.382 0.479 0.155

Sandy loam 1.541 0.419 0.114

(9A) 1.789 0.326 0.115

No 1.258 0.525 0.228

Sandy clay loam 1.5 15 0.428 0.284

(lOA) 1.757 0.337 0.325

Kp 0.871 0.671 0.319

Clay

(13A)

Kp

Sandy clay

(14A)

Hli

1.021 0.615

1.168 0.559

1.346 0.492

0.962 0.637

1.082 0.591

1.255 0.526

1.397 0.473

1.242 0.531

Sandy clay loam 1.5 II 0.430

(l6A) 1.765 0.334

0.375

0.435

0.456

0.241

0.282

0.316

0.366

0.273

0.263

0.292

0.427

0.397

0.390

0.242

0.222

0.179

0.686

0.072

0.082

0.112

0.074

0.064

0.182

0.188

0.185

0.366

0.367

0.373

0.339

0.251

0.260

0.252

0.262

0.188

0.174

0.165

0.261 0.154

0.326 0.360

0.391 0.352

0.440 0.35 I

0.173 :·e...I,,28

0.192 0.120

0.217 0.117

0.058 0.036

0.064 0.038

(Um 0.044

0.062 0.045

0.072 0.047

0.089 0.050

0.170 0.135

0.197 0.130

0.232 0.132

0.269 0.308

0.312 0.306

0.366 0.313

0.404 0.300

0.180 0.187

0.217 0'.200

0.25 I 0.200

0.294 0.210

0.166 0.13

0.222 0.147

0.256 0.145

0.246 0.145

0.278 0.307

0.354 0.318

0.405 0.324

0.126 .0.093

0.151 0.094

0.177 0.095

0.050 0.031

0.054 (UJ32

0.065 (Um

0.048 0.035

0.057 0.037

0.0720.040

0.164 .0.131

0.196 0.129

0.230 0.131

0.240 0.275

0.285 0.280

0.334 0.286 '

0.379 0.282

0.149 0.155

0.191 0.177

0.233 0.196

0.276 0.197

0.154 0.124

0.206 0.136

0.244 0.138

61

50

48

153

163

116

54

57

68

93

42

26

58

87

93

50

63

69

52

61

65

65

72

67

41

36

109

87

84

0.285

0.180

0.075

1-3 200-400

200

204

156

.O.185r 5-11 20-1500

62

67

80

0.195

0.170

0.185

0.216

0.171

107 0.345

57· ·0.275

43 0.220

64 0.380

88 0.190

95 0.145

79 0.505

90

101

77

92

91

83

90

78

57

48

0.400

0.205

0.070

,0.685

0.530

0.335

0.130

0.415

0.220

0.145

1-25 10-1000

8-40 20-50

2-4 4-100

2-10 2-1500

2-5 2-10

2-7 15-30

a Readily available water (-10 to -100 kPa); b Available water capacity (-10 to -1500 kPa); C Water release index (see text); 8v Volumetric water content

(mJ m·J); 8m Mass water content (kg kg-I); \llm Matric potential (kPa)

potentials most frequently designated as Fe. Since Fe desig­nations vary so frequently in the literature, this may explain the uncertain results regarding ~he effect of bulk density on Fe and therefore on A We. For example,.a clear interpreta­tion of the effects of compaction on the A we of Kwa­Zulu-Natal soils (Hill & Sumner, 1967) was difficult as the results were expressed only on a mass basis. However, the expression of results on a mass basis provides a better means with which to examine physical differences in compacted soils (Hill & Sumner, 1967) but is less useful for practical

interpretation. The overall effect of compaction on soil is the reduction in porosity and the conversion of larger pores to smaller pores, the relative importance of these two processes being dependent on. soil texture (Hill & Sumner, 1967; Katou et aI., 1987).

When considering the effects of compaction on water retention, the expression of water content on a volumetric or mass basis is more than just semantics. While differing expressions do not necessarily imply any physical differences in water-energy relations for a particular soil, they do convey

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

S. Afr. J. Plant Soi12001, 18(3) 95

Table 3 Selected parameters of the water retentivity curve down to -100 kPa for a range of selected forestry soils at a range of bulk densities

Soil toml

Soil texture

Site No.

No

Sandy clay loam

·(6A)

la

Silty clay

(5A)

Kp

Clay

(4B)

Kp

Clay

(4B2)

Cf

Sandy loam

(I E)

Cv

Sandy clay loam

(15A)

la

Sandy clay loam

(12A)

Hu

Sandy clay loam

(IIA)

II'~, at - fo kPa

Bulk densi~y Porosity e, H",

(Mg m·J ) (m J m·3) (m J ni·3) (kg kg· l)

1.215 0.542 0.275 0.226

1.395

1.566

1.096

1.242

1.389

0.961

1.095

1.228

1.075

1.219

1.362

1.442

1.642

1.834

1.366

1.549

1.736

1.202

1.371

1.535

1.345

1.488

1.634

0,473 0.312 0.224

0,409 0.347 0.222

0.586 0.349 0.319

0.5310.402 0.324

0,476 0.465 0.334

0.637 0.365 0.380

0.587 0,421 0.385

0.537 0.476 0.388

0.594 0.371 0.345

0.540 0.426 0.350

0,486 0.491 0.361·

0,4560.1610.112

0.380 0.167 0.102

0.308 0.180 0.098

0,484 0.234 0.171

0,415 0.278 0.180

0.345 0.298 0.171

0.546 0.281 0.234

0,483 0.385 0.281

0,421 0.431 0.281

0,493 0.189 0.141

0,439 0.215 0.145

0.383 0.224 0.137

II'", at -I 00 kPa Crossover

e,. e", RA W" WRlh II',. (ml"",c)

(in3 m'3) (kg kg'l) (mmm-I) (mm m'l) (-kPa)

0.196 0.161

0.240 o.ln

0.295 0.188

0.278 0.254

0.345 0.278

0.406 0.292

.0.305

0.345

0.401

0.317

0.315

0.327

0.282 0.262

0.350 0.287

0.440 0.323

0.115 0.080

0.139 0.085

0.145 0.079

0.175 0.128

0.208 0.135

0.231 0.133

0.180 0.150

0.247 0.180

0.315 0.206

0.129 0.096

0.159 0.107

0.180 0.110

79

n 52

71

57

59

60

76

75

89

76

51

46

28

35

59

70

67

101

138

116

60

56

44

0,495

0.175

0.065

0.395

0.360

0.100

0.496

0.230

0.120

0.341

0.272

0.045

0.545

0.370

0.280

0.520

0.285

0.120

0.295

·0.300

0.155

0,420

0.375

0.155

3-5

2-4

2-4

3-5

4-5

3-10

1-5

1-5

\~m (mass)

(-kPa)

7-12

4-6

6-10

4-10

4-7

4-50

4-15

2-30

" Readily available water (-10 to -100 kPa); h Available water capacity (-10 to -1500 kPa); c Water release index (see text);

e" Volumetric water:content (m3 n1'3); e", Mass water content (kg kg'I); II'", Matric potential (kPa) 1 j I

different types of information related to pore geometry and I

available water capacity. ~or example, the flattening of the WRC during compaction i~ the result of a loss in total poro­sity and a modification in ppre size. The actual change in pore geometry is better reflecte:d by the expression of the water content on a mass basis ~ithout consideration of volume effects. From a practical and modelling point of view, changes in available water must be expressed on a volumetric basis.

Effect of compaction on field capacity ana wilting point

The data presented in Figure I (a - d) and in Tables 2 and 3 indicate that compaction may decrease or increase the FC when the water content is expressed on' a mass basis and when FC is assumed to correspond to a matric potential of -10 kPa. This is because the 'crossover' of WRCs for different levels of compaction occurred at a wide range of matric potentials, depending on soil type. This range of water con­tents corresponded to matric potentials of between -2.0 kPa and -1000 kPa. However,. no clear pattern of the effect of compaction on water content at -10 kPa with soil type emerged (Tables 2 and 3).

With the exception of three soils (two loams, 7 A and 2A, and one sandy loam 9A), the etfect of compaction generally was to increase the water content at FC on a volumetric basis (Tables 2 and 3; see also Figure I a, c and d). The reason for this is that increasing compaction results in the 'crossover' of the WRCs at various compaction levels within the range of water contents corresponding to -2.0 to -11.0 kPa. In all cases volumetric water content increased with increasing compac­tion at matric potentials lower than -I 1.0 kPa.

A problem with interpreting the effects of compaction on FC is that the WRCs for various bulk densities of the same soil 'crossed' between matric potentials of -1.0 and -11.0 kPa which is within the range commonly designated for FC (Smith & Thomasson, 1974; Cassel & Nielson, 1986). This 'crossover' po'mt is of importance particularly if FCis taken at a higher matric potential than -10 kPa which would be the case for soils where an impeding layer is present. For exam­ple in Figure Id the FC for the Cartref soil (IA) would decrease rather than increase (see Table 2) if an FC of -5kPa were used to reflect more poorly drained conditions.

The creation of meso- and micro-pores at the expense of macropores accompanies reduction in porosity during the

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

96

compaction process (Hill & Sumner, 1967). However, the absolute change in pore size distribution and its relevance to the resultant WRC are complicated and reflect, complex changes in pore geometry following soil compaction.

With increasing compaction only a marginal rise in wilting point (WP) is noted when expressed on a mass basis but a large rise occurs when expressed on a volumetric basis. The contention of Reeve et al. (1973) that A WC on a volumetric or depth basis essentially increases with increasing compac­tion because of the associated rise in FC while WP remains fairly constant, is curious. The data presented here show that although FC increases for most soils with increasing compac­tion, WP also rises markedly with increasing compaction for each soil in the study when expressed on a volumetric basis. Thus increases in A WC (mm m· l ) caused by increases in FC were offset by increases in wilting point.

Water release index

Table 2 shows that the water release index (WRI) decreased substantially with increasing compaction. In other wordstfli~,. slope of the central portion of the WRC decreased with increasing bulk density (Figure I, a - d). This decrease in WRI corresponds to a decrease in large pore sizes which may have the effect of substantially lowering infiltration rates (Akram & Kemper, 1979). O'Sullivan and Ball (1993) sug­gested that WRI was a sensitive indicator of changes in soil structure and reported a noticeable decrease in WRI for a lim­ited number of topsoils from field experiments in Scotland. It is not clear what practical value the water release index has, if any, for plant growth since it consistently decreases with increasing compaction whereas A WC increases or decreases with increasing compaction level. Although large macro­pores are lost during compaction many are converted into mesopores in the available range. This is reflected by the 'shoulder' of the S-curve being moved to the right with increasing compaction (e.g. Figure I, a - d).

Available water capacity

Owing to the small range of bulk densities covered per soil,a clear effect of compaction on A WC,expressed in mm m· l .

was difficult to ascertain precisely but it is believed that a very useful insight was obtained. Table 2 shows that increas­ing compaction lowered A WC for most soi Is. Of the eleven soils studied three showed a continuing increase in A WC fol­lowing increa,sed compaction. These were a, Cartref sandy loam (I A), Nomanci sandy clay loam (lOA), and a Fernwood loamy sand (8A)(Table 2). The WRC of all of these soils crossed over with increasing compaction at matdc potentials greater than -10 kPa. Therefore, increases in A WC were due to an increase in Fe. Two soils, a Kranskop clay (13A) and a Cartref loam (7 A) first showed an increase in A WC before decreasing with increasing compaction (Table 2). This is sim­ilar to the behaviour repOlted by Archer and Smith (197'2) and Reeve et at. (1973). Tables 2 and 3 show that if all the soi Is are considered, compaction has a similar effect on readily available water (RA W) for only 4 out of 18 of the soils in this study, i.e. an increase with increasing compaction and then a decline with further compaction. Clearly, for these soils, moderate compaction converts many of the larger pores in the unavailable range (> -'10 kPa) into smaller pores in the availa-

S. Afr. Tydskr. Plant Grond 200 I, 18(3)

ble range. When expressed on a volumetric basis, A WC generally

decreased following soil compaction except for sandy soils and clayey soils. Increases in FC following compaction did not necessarily result in increases in A WC as suggested by Reeve et al. (1973) as wilting point also increased with com­paction. The effect of compaction on A WC expressed on a mass basis was more complicated since the effect of compac­tion on FC was more pronounced than on witling point. For a range of bulk densities, WRCs 'crossed' when' expressed on mass basis at matric potentials of between -2.0 kPa and -1500 kPa depending on soil type.

In the case of the Fernwood loamy sand (8A), increasing compaction resulted in increasing A WC (Table 2) presuma­bly owing to the macropores being increasingly modified into mesopores. Interestingly, on the site where this sample was taken no significant decreases in tree growth (Eucalyptus grandis x camaldulensis) following compaction occurred. Increases in A WC with increasing compaction were accom­panied by significant increases in root length density and sfight increases in tree growtIi (Smith, Du Toit & Sibisi, 1999). These results are attributed to less water and nutrients draining out of the rooting zone and being available for tree growth (see for example Agrawal et aI., 1987 and Agrawal, 1991).

Conclusions

The data presented here forms only a small part of the type of information that is required when evaluating the effect of tim­ber harvesting operations on soil and long-term site produc­tivity but nevertheless provides an insight into the likely changes in soil physical properties following intensive site operations. When considering tree or plant growth, changes in soil water retention properties during the compaction proc­ess, as reported here, must not be seen in ,isolation. Other soil physical properties affecting tree root devel9pment which need to be considered during the compaction process are soil aeration, soil strength and hydraulic conductivity.

, Although clear relationships between soil type and the effect of compaction on A WC were not forthcoming, several response patterns were elucidated. While for many soils increases in compaction result in an immediate decline in A WC, this is not true in all cases. Since indicators of long-term forest sustainability are increasingly sought (Franc, Laroussinie & van Bueren, 2000) the information presented he're demonstrates the necessity of understanding the complex inter-relationships between site impacts and soil physical quality. Many of the relationships presented here, coupled with field measurements, have been utilised in explaining the relationship between tree growth and harvesting impacts on a site specific basis and providing essential guidelines for the management of harvesting operations in commercial planta­tionsin South Africa (e.g. Smith, 1999).

Acknowledgements

The authors wish to express their appreciation to Mrs Mary Galbraith and Mr mnoceru Mchunu for their considerable assistance with field and laboratory work. The authors grate­fully acknowledge the assistance of Mr Ben du Toit in pro­viding a translation of the Abstract into Afrikaans and Mr

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14

S. Afr. J. Plant Soil 2001', 18(3)

Steven Dovey for help in the preparation of the graphical material. The comments of an anonymous referee also added value to the manuscript.

References AGRA WAL, R.P., 1991. Water and nutrient management in sandy

soils by compaction. Soil Till. Res. 19, 121-130. AGRAWAL, R.P., lHORAR,B.S., DHANKAR, l.S. & RAJ, M.,

1987. Compaction of sandy soils for irrigation management. Irrig. Sci. 8,227-232.

AKRAM, M. & KEMPER. W.D, 1979. Intiltration of soils as atl'ected by pressure and water content at the time of compaction. Soil Sci. Soc. Amer. 1. 43. 1080-1086.

ARCHER . .I.R. & SMITH, P.D .. 1972. The relation between bulk density, available water capacity, and air capacity of soils . .J. Soil Sci. 23.475-480.

ARYA, L.M. & PARIS . .I.F .. 1981. A physioempirical model to pre­dict the soil moisture characteristic from particle-size distribution and bulk density data. Soil Sci. Soc. Amer. 1. 45. 1023-1030.

BENNIE, A.T.P. & BURGER, R du T., 1979. Grondverdigting onder besproeiing op die Vaalhartsbesproeiingskema. Volume II: Die invloed van grondverdigting op die grond - plant sisteem. Dept. of Soil Science Report No 79/2. University of the Orange Free State. Bloemfontein. .

CASSEL, D.K. & NIELSON, D.R., 1986. Field capacity and availa­ble water capacity. In A. Klute (ed), p90 I - 926. Methods C?/,Soil ' Analysis. Part I. 2nd Edition, Agronomy Monograph 9. ASA and. SSSA, Madison, Wisconsin.

CRONEY, D. & COLEMAN, J.D. 1954. Soil structure in relation to soil suction (pF). 1. Soil Sci. 5, 75- 84.

DE .lONG, R., CAMPBELL, C. A. & NICHOLAICHUK, W., 1983. Water retention equations and their relationship to soil organic matter and particle size distribution for disturbed samples. Can. 1. Soil Sci. 63,291-302. .

FRANC, A., LAROUSSINIE, O. & VAN BUEREN, E.L.. 2000. About criteria and indicators for sustainable forest management. In Krishnipillay, B. et al.. XXI IUFRO World Congress. 7 - 12 August, 2000, Kuala Lumpur. Malaysia. V.I Sub-plenary ses­sions: Forests and Society:: the role of research: 288 - 294. Kuala Lumpur,IUFRO.

GREY, D.C. & .IACOBS, E.p., .1987. The impact of harvesting on forest site quality. S. Afr. nor. 1. 140,60-66.

GREY, D.C., SCHONAU, Alp.G., & HERBERT, M.A .. 1987. For­est soils and their implica~ions for management. Forestry Hand­book, South African Foresfry Research Institute, p.70 - 81.

GUPTA, S.C & LARSON, W.E., 1979. Estimating soil water reten­tion characteristics from p~rticle size distribution, organic matter percent and bulk density. Water Res. 15, 1633 - 1635.

GUPTA, S.c. SHARMA, P.P. & DEFRANCHI, S.A., 1989. Com­paction effects on soil structure. Adv. Agron. 42, 311-338.

HILL, .I.N.S. & SUMNER, M.E., 1967. Effect of bulk density on moisture characteristics of soils. Soil Sci. 103, 234-238.

HUTSON, l.L., 1986. Water retentivity of some 'South African soils in relation to particle size criteria and bulk density. S. Afr. 1. Plant & Soil. 3, 150-155

97

KATOU, H., MIYAJI, K. & KUBOTA, T., 1987. Susceptibility of undisturbed soils to compression as evaluated from the changes in the soil water characteristic curves. Soil Sci. Plant Nutr. 33, 539- 554.

MIRREH, H.F. & KETCHESON, .I.W., 1972. Intluence of soil bulk density and matric potential on soil resistance to penetration. Can. 1. Soil Sci. 52; 477-483 ..

MOFFAT, A..l., 1991. Forestry and soil protection in the UK. Soil Use Manage. 7,145-151.

O'SULLIVAN. M.F. & l3ALL, B.C .. 1993. The shape of the water release characteristic as affected by tillage. compaction and soil lype. Soil Till. Res. 25. 339-349.

RA.lKAI, K., KABOS. S .. VAN GENUCHTEN, Th. M. & .IANS­SON, P., 1996. Estimation of water- retention characteristics from the bulk density and 'pmtic1e-size distribution of Swedish soils. Soil Sci. 161. 832-845.

RAWLS. W . .I.. GISH. T..l. &! l3RAKENSIEK. D.I., 1991. Estimat­ing soil water retention from soil physical properties and charac­teristics. Adv. Soil Sci .. 16. 213-234.

REEVE, M.J, SMITH, P.D .. & THOMASSON, A.J., 1973. The effect of density on water retention properties of tield soils. 1. Soil Sci .. 24,355-367. :

SCHULZE. R.E .. HUTSON . .I.L & CASSo A. 1985. Hydrological .. ,characteri,stics ~nd pr~perties~f soils in southern Africa 2: Soil

; t water retention models. Water SA. II, 129-136. SMITH, C.W .. JOHNSTON. M.A. & LORENTZ, S., 1997. Assess­

ing the compaction susceptibility of forestry soils. II. Soil proper­ties affecting compactibilily and compressibility. Soil Till. Res. 43.335- 354.

SMITH, C.W., DU TOIT, B. & SIBISL N.N .. 1999. The effect of harvesting operations on soil physical properties and growth of Eucalyptus spp. on tive ditl'erent.forestry sites in South Africa. In: Kellogg, L. (Ed); Timber harvesting and transportation tech­nologies for forestry in the new millennium p.140 - 151: Pieter­maritzburg. South Africa. .I line 10 - I I. 1999: Conference Proceedings [N.P.]. International Union of Forest Research Organisations (IUFRO).

SMITH, C.W .. 1999. Site damage and long-term site productivity of forest plantations in South Africa: Impacts of harvesting opera­tions and suggested management strategies. ICFR Handbook II 99. Institute for Commercial Forestry Research, Pietermaritz­burg.

SMITH. P. & THOMASSON. A..I.. 1974. Density and water-release characteristics. In B.W. Avery & c.L. Bascomb, (eds): Soil sur­vey laboratory methods. Tech. Monograph No.6, Rothamstead Experimental Station. Harpendon. Herts.

SOIL CLASSIFICATION WORKING GROUP. 1991. Soil classiti­cation: a taxonomic system for South Africa. Memoirs on the agricultural resources of SOllth ;!li'ica No 15. Department of Agricultural Development. Pretoria. 257pp.

SOIL SURVEY STAFF, 1990. Keys to Soil Taxonomy. 4th Edition. SMSS Technical Monograph .. Blacksburg, Virginia, 422pp.

VAN GENUCHTEN. M. Th .. 1980. A closed-form equation for pre­dicting the hydraulic conductivity of unsaturated soils. S9i1 Sci. Soc. Amer. J. 44: 892-898.

Dow

nloa

ded

by [

Mor

aine

Com

m C

olle

ge]

at 0

1:16

11

Nov

embe

r 20

14