ANASSESSMENTOFSTRENGHTPROPERTIESOF

VETIVERGRASSROOTSINRELATIONTOSLOPESTABILIZATION

Dit i HengchaovanichandNimalS.Nilaweera1

Introduction

Vetivergrass(Vetiveriazizanioides)hasbeenutilizedtoreducesoilerosioninmanycountriesthroughouttheworldforalongtime.Itiswellunderstoodthattherootpropertiesofvetivergrass can help reduce soil erosion and strengthen slopestability when planted properly. Vetiver hedgerows cultivatedacross slope soil can block the passage of soil particles anddevelopterracesbetweenthehedgesenhancingstabilityoftheslope. Some previous studies on vetiver have elucidated themorphological properties of the root and their qualitativesignificance for erosion control and slope stabilization(Grimshaw1994;Yoon1994).Theyemphasizetheearlydevelopingdeeplypenetrating(sometimesupto3.5m)fibrousroot system of vetiver and its capability of anchoringthemselvesfirmlyintoslopesoilprofiles.However,thestrengthpropertiesof vetiver root,whichalsoplayan important role interms of erosion control and slope stabilization by means oftheir influencesontheshearstrengthofslopesoilhasnotyetbeen adequately understood. When a plant root penetratesacrossapotentialshearsurface inasoilprofile, thedistortionoftheshearzonedevelopstensionintheroot;thecomponentof this tension tangential to the shear zone directly resistsshear, while the normal component increases the confiningpressure on the shear plane. Therefore it is essential todetermine tensile root strength properties in the process of

evaluating a plant species as a component in slopestabilization. Recently, in Malaysia the vetiver hedgerow techniquestarts to gain popularity in erosion control and slopestabilization. It has been andwill be used to stabilize severalroadembankmentsoftheEast-WestHighwayandsomeotherroad projects. Simultaneously, attempts have been made toanalyze the effects of vetiver root on slope stabilization anderosioncontrol.Thispaperdiscusses the tensile rootstrengthof vetiver and its contribution to soil strength throughexperimentsontensilerootdeterminationsandroot-permeatedsoil shearing,whichareapartofongoing researchespeciallydesigned to access both root strength properties andmorphological parametersof roots in relation to slopestabilityanderosioncontrol

1EroconSdn.Bhd.,KualLumpur,MalaysiaTensileRootStrengthorVit iverGrass Forthedeterminationoftensilerootstrength,maturerootspecimens were sampled from two-year-old vetiver plantsgrownonanembankmentslope.Thespecimensweretestedinfreshconditionlimitingthetimeelapsedbetweenthesamplingand the testing to two hoursmaximum. The unbranched andstraight root samples, about 15-20 cm long, were verticallyconnected to hanging spring balance via a wooden clamp atone end while the other end was fixed to a holder that waspulled down manually until the root failed. At failure, themaximum load was monitored. Subsequently, the mode offailurewasexamined foreachsampleand the resultsofend-sheared samples and those with unusually altered rupturepoints were discarded. To calculate the tensile root strength,the root diameterwithoutbarkwasusedsince thebark failed

before the root due to its weaker strength properties, andeventually the total tensile stress transferred to the root core.About 80 vetiver root specimens of different diameter classesvarying from0.2 to 2.2mmwere tested and the resultswereinterpretedas theultimate tensile forceand tensilestrength inrelationtorootdiameterwithoutbark. The ultimate root tensile force against the root diameterplot for the vetiver roots is presented in Fig. 1. The powerregression analysis of the relationship between the ultimateroottensileforceandtherootdiameterprovidesthebestfitwiththefollowingequation:

F1=46.93d1.4217 (1) whereisF1-ultimatetensilerootforce,and d-therootdiameter Thispowerregressionfunctioncanbeusedtopredicttheultimate tensile forceofa vetiver rootwith knowndiameter.Acomparisonof tensile resistanceof vetiver rootswith thoseofsomehardwoodspeciesismadeinFig.2.Theultimatetensileroot forces versus root diameter relationships for Japaneseceder, Dipterocarpus alatus, and Rocky Moutain Douglas firwereobtainedfromearlyworksofAbeandIwamoto(1986),Nilaweera(1994),andBurroughsandThomas(1977)respectively. The comparison clearly indicates that the tensileresistance of vetiver roots is as high as that of the hardwoodvegetation,sometimesevenhigher,contrarytothefactthatitisagrassspecies. The tensilestrengthof the root isdefinedas theultimatetensile root force divided by the cross-sectional area of theunstressed root. If the tensile root strength is constant forvetiver roots, the ultimate tensile force, F1, should beproportional tod2.AccordingtothefunctiongivenbyEquation(1),F1isproportionaltod1.4217implyingthatthetensileroot

strengthdecreaseswiththeincreaseinrootdiameter.Figure3illustrates theactual relationshipbetween tensile root strengthandrootdiameter.Thetensilerootstrength,Ts,decreaseswiththeincreasingrootdiameter,d,followingthepowerregressionrelationship. Ts=59.80d-0.5785 (2) Similar relationships were reported from many previousworks on hardwood roots. This phenomenon implies thatstronger, finer roots provide higher resistance than larger-diameter roots with comparatively low tensile strength for agivenrootcross-sectionalareaofspecies.AccordingtoFig.3,thetensilestrengthofvetiverrootsvariesfrom180to40MPafortherangeofrootdiameterof0.2-2.2mm.Themeantensilestrengthisabout75MPaat0.7-0.8mmrootdiameterwhichisthemostcommondiameterclassforvetiverroots.Comparedtomany hardwood roots, the average tensile strength of vetivergrass isveryhigh.Eventhoughsomehardwoodrootsprovidehighertensilestrengthvaluesthantheaveragetensilestrengthofvetiver roots in the rootdiameterclassof0.7-0.8mm, theiraverage tensile strength values are lower since the averagerootdiameterismuchhigherthanthatofvetiverroots.

DirectShearTestsonVetiverRoot-PermeatedSoil

Rootsof treesandother vegetationprovidea reinforcingeffect to soil through tensile resistance and frictional oradhesionalproperties.Thereinforcingeffectortheincreaseofshear strength in soil due to roots can be quantified byconduction in-situ direct shear tests on root –permeated androot-free soils at the same location. The difference betweenshearstrengthvaluesof root-permeatedsoiland root-freesoilsheared under the same conditions gives the shear strengthincrease due to the roots. In order to determine the root-

reinforcement effect of vetiver grass, large-scale direct sheartestswereperformedonaslopedsoilprofileofanembankmentvegetatedwith vetiver. The test apparatus comprised a shearbox,ahydraulicjackingsystem,aprovingringanddialgages.Theshearboxwasmadeof8mmthicksteelplatescapableofholding firmly a soil block of 50 cm x 50 cm x 50 cm indimensions.Ahydraulicjackingsystemwithcapacityof10tonsproduced theshear load through theproving ringof3 tonsofmeasuringcapacitywhichcontrolledtheshearforcewhilefourdialgagesmeasuringthesheardisplacement(Fig.4).

Thetestplantswereselectedfroma50-cm-longvetiverhedgerowthatusuallyincludes3plantsplantedataspacingof15cm.Thesoilsurroundingtheplantswasremovedleavinga50cmx50cmx25cmroot-permeatedsoilblockcenteringthehedgerow.Subsequently,theshearboxwassetsoastocoverthe soil block and the loading and displacement measuringsystemswereassembled.Thesoilblockwith25cmheightthenshearedhorizontally towards the slopedirection under stress-controlled-condition.After shearing, theshearsurfaceand theorientation of failed rootswere examined carefully in order toestimatethesheardistortionduringfailure.Itwasobservedthattheaveragesheardistortionduringfailurewasabout30º.Thetotalcross-sectionalrootareaontheshearplaneandthebulkweight of roots in the sheared soil block were measured inorder to determine the root area ratio and the biomass,respectively. This procedure was followed for each 25 cm ofdepthundertheentirevetiverhedgerowlengthof50cmupto1.5mdepth. For each depth level of shearing, a root-free soilprofile adjacent to the root-permeated soil profile was alsosheared under the same shearing conditions. Each soil blockwasshearedunderitsself-weightasthenormalload.Thebulkdensity of test soil was determined before each test for a

comparisonofthenormalloadontheroot-permeatedsoilblockwith that on the counterpart root-free soil block. Each pair oftestswasmadeunderequal,normalstressconditions.

Table 1. Shear strength increase in soil profile due to rootpenetrationoftwo-year-oldvetiverplantswithspacing15cminahedgerowof50cmlengthDepth(m)

DR(kg/m3)

AR(mm2)

Ar/AX10-4

∆S(kN/m2)

%∆S ∆S(kN/m2)

0.250.500.751.001.251.50

1.5220.7010.5210.3780.1810.135

331.0176.2137.8106.871.251.6

3.311.761.381.070.710.52

8.924.173.462.611.941.28

90.239.334.626.319.012.5

5.394.735.024.895.454.96

DR–bulkweightof root inunitsoilvolume,AR– rootareaonshearsurface,A–areaof theshearsurface,∆S–shearstrength increase insoilduetoroots,∆S-shearstrengthincreaseinsoildueto1cm2rootarea.

The test results were processed in order to obtain therelationshipbetween theshearstressandsheardisplacementforeachtest.Figure5presentstheshear-stressversusshear-displacement curves, each plot representing the relationshipsforroot-permeatedandroot-freesoilsforeach0.25mofdepthup to 1.5 m of root penetration. The difference between themaximumshearstressofroot-permeatedsoilandthatofroot-freesoil ataparticulardepth isdefinedas theshearstrengthincreaseinsoilduetothepresenceofvetiverroots(∆S).Accordingtothetestresults,itisobviousthatthepenetrationofvetiverrootsinasoilprofileincreasestheshearstrengthofsoilsignificantly.Foreach testdepth, theshearstrength increase,

thecorresponding rootcross-sectionalarea,and thebulk rootweightperunitvolumeofsoilweredeterminedandtabulatedinTable1.The∆svaluedecreaseswithindepthfrom8.90kN/m2at0.25mdepth to1.82kN/m2at1.50mdepthdependingonthe number of roots penetrating through the shear surface. Acomparisonof thevariationof∆s and the root cross-sectionalarea on the shear surface in given in Fig. 6 for the depth ofroot-penetration. The vetiver root penetration of a 2-year-oldhedgerow with 15 cm plant spacing can increase the shearstrengthofsoilinanadjacent25-cm-widestripby90%at0.25m depth. At 0.5m the shear strength increase is about 39%and is thengradually reduced to12.5%at1.50mdepth.Theshearstrengthincreasedueto1cm2rootarea(∆s)iscalculatedforeachtestdepthandpresentedinTable1.The∆svaluevariesveryslightlywithanaverageof5.1kN/m2fortheanalyzedroot-penetrationdepth.

Discussion

Theoretically,theaveragetensilestrengthofrootscanbeused to compute the shear strength increase in soil due topenetration of roots across a shear plane. The computationadapts the simple model of root-reinforced soil subjected todirectshear(Wu,1976).Accordingtothismodel,thetensileforcethatdevelopsintherootswhenthesoilisshearedcanberesolved into a tangential component which directly resistsshearandanormal componentwhich increases theconfiningstressontheshearplane.Themodelsimplyassumesthattherootsarefullymobilizedduringshearing.Themobilizedtensileresistance in the roots translates into an increase in shearstrengthinthesoilasexpressedbythefollowingequation:

∆s=tR[cos𝜃tan𝜙+sin𝜃] (3)

Whereare: 𝜃 –angleoesheardistortion

𝜙–angleofinternalfriction

tR – average tensile strength of roots per areaunitofsoil

Theaveragetensilestrengthofrootsperareaunitofsoilcanbedeterminedbymultiplying theaverage tensilestrengthoftheroots(TR)bythefractionofthesoilcross-sectionoccupiedbyroots,ortherootarearatio(AR/A). UsingTR=75MPa,𝜙=30ºand𝜃=30º, the increase intheshearstrengthinthesoilduetovetiverrootwascalculatedforeachtestdepthandtabulatedinTable2withexperimentalresults.Table2.Theexperimentalandcomputed∆svaluesatvariabledepths.

Depth,m ∆s,kN/m2(experimental)

∆s,kN/m2(computed)

0.250.500.751.001.251.50

8.924.173.462.611.941.28

24.8313.2010.358.035.333.90

Thecomputed∆svaluesareabout3timesashighasthevalues obtained from the field experiments. The disparitybetween∆svaluesobtainedfromexperimentsandcomputationcan be attributed to assumptions made on the root-reinforcementmodelandthenatureofrootspecimensusedinthe tensile tests. During shearing of root-permeated soil, the

tensile strength of each and every root was not mobilizedcompletelyasassumed in themodel.Somerootswerepulledoutcompletelyorpartlybyaruptureatafinerpointbelowtheshear surface providing a lower resistance to shearing thanexpected. Even though the root penetration of vetiver isgenerally vertical, as assumed in the model, some rootorientationsobliquetotheshearsurfacecangiverisetolowershear-strength increase in soil. In actual conditions, the rootcrookedness, jointing and the presence of young roots yieldlower∆svaluesthanthoseexpectedfromstraight,unbranchedandmaturerootwhicharestrongerthantheformer. Thought the adaptation of the root reinforcement modeldoes not compute the shear strength increase directly, anestimationofshear-strengthincreasecanbemadebydividingthecomputedvaluesbyafactorof3forvetiver-root-permeatedsoil with the angle of internal friction of 30º. Furthermore, thecorrelation of∆swith the root area ratio and bulk rootweightperunitofsoilvolumeclearlyindicatelinearrelationshipswhichcanbeusedtopredicttheshearstrengthincreaseinsoilduetovetiverroots(Fig7).Thevalue∆sincreaseslinearlywiththerootarea ratio in theorderof 2.7 x104 for vetivergrass.Therelationship between the∆s and bulk root weight per volumeunit of soil indicates somepositive intercept of shear-strengthincreaseowingtothenatureofrootsandrootpenetrationwithdepth.Atshallowerdepths,thefractionsofrootweightgivenbyobliquelyorientedrootsandbytherootsterminatingbeforetheshearzonearehigher than thoseatdeeper levels.Therefore,atshallowdepthstherootweight isnotdirectlyproportionaltothe ∆s value. As a consequence, a positive intercept in therelationshipbetweenthe∆svalueandthebulkrootweightperarea unit of soil appears though the intercept should betheoreticallyzero.Foraknownroot-arearatiooraknownbulk

weightofrootspervolumeunitofsoil, theserelationshipscanbe used to predict the∆s value instead of rather difficult andexpensivedirectsheartests.Carefullyextrudedrootsystemsofvetiverplantsbywaterjettingcanbeusedtodeterminetherootareaandtherootweightatdifferentdepthsofrootpenetration. In the present study, the shear-strength increase in thesoil by the root penetration of a vetiver hedgerow at differentdepthsofupto1.5mwasdeterminedfora0.5mwidestripofsoil across the slope. In general, for a 1 m-wide hedgerowspacingthese∆svaluescanbeuseddirectlyatrelevantdepthintervals throughout theslope.However, forgreaterhedgerowspacings the ∆s  values should be corrected according to thepertaining areas of influence. It was unable to investigate theinfluenceofvetiverrootsontheshearstrengthofsoilbelow1.5depthduetothedifficultiesencounteredduringexcavationandthe setting-up of testing equipment. Field evidences indicatethatagradualandslowdecreaseinrootpenetrationwithdepthafter the upper most 0.5 m where a rapid decrease in rootpenetrationoccurs.According to the trendof the∆s decreasewithdepthitcanbepredictedthatashear-strengthincreaseofabout1kN/m2at2mdepthbelowthevetiverhedgerowtakesplace.

Conclusion

The tensile root strength properties of vetiver grass inassociationwith its inheritedmorphological rootcharacteristicsimprove the resistanceofsoil slopes toshallowmassstabilityandsurfaceerosion.Thetensilestrengthofvetiverroots isasstrongas,orevenstrongerthan,thatofmanyhardwoodrootswhichhavebeenprovenpositive for root reinforcement insoilslopes.The root tensilestrengthofvetiverdecreaseswith theincrease of root diameter as in the case of hardwood roots.

Compares to hardwood vegetation, the smeller average rootdiameterofvetiverfurnishesaveryhighmeantensilestrength(75MPa),indicatingthatvetivergrassismoreeffectualinrootreinforcementinsoilslopes.Thepenetrationoffineandstrongvetiverrootsinasoilprofilecanincreasetheshearstrengthofsoilsignificantlyatshallowdepths.Theshearstrengthincreasein soil due to the root penetration of a 2-year-old vetiverhedgerowwithplantspacingof15cmvariesfrom90%at0.25m depth to 1.25 % at 1.50 m depth. The shear-strengthincrease in soil due to vetiver roots can be approximated byusing the average tensile root strength and the existing rootareaoccupiedbyvetiverrootsonapotentialshearsurfaceatacertain depth or by using the relationships of shear-strengthincreaseinthesoilversustheroot-arearatioorthebulkweightofrootpervolumeunitofsoil.

ReferencesAbe,K;andM.Iwamoto.1986.Anevaluationoftreerooteffectonslopestabilitybytreeroot

strength.J.JapaneseForest.Soc.68:505-510.Burroughs., E.R. ; and Thomas, B.R.. 1977. Declining rootstrengthinDouglasfirafterfellingas

a factor in slope stability. Research Paper INT-190,Intermountain Forest andRangeExperimentStation,USForestService,Ogden,Utah,USA.

Grimhaw, R.G. 1994. Vetiver grass-Its use for slope andstructuralstabilizationundertropical

andsemi-tropicalconditions.Proc.IntConf.onVegetationandSlopes,Oxford,England

Nilaweera, N.S. 1994. Effects of tree roots on slope stability:ThecaseofKhaoLuangMountain

area,SouthernThailand.DoctoralDissertation,Diss.No.GT-93-2, Asian Institute of Technology, Bangkok,Thailand

Yoou,P.K.1994. Importantbiologicalconsiderations inuseofvetivergrasshedgerows(VGHR)

for slope protection sna stabilization. Proc. Int. Conf. onVegetationandSlopes,Oxford,England