In situ detern血蜘n of・field−sa加rated h雌a誼。伽面etiv．ity ofsubsurfaee soil of vadose zone by teehnique ofPressure lnfiltrometer

(1)

Journal of the Facuity of Environmental Science and Technology， Okayama University VoL4， No．1，pp81L92， February 1999

In situ detern血蜘n of・field−sa加rated h雌a誼。伽面etiv．ity of

subsurfaee soil of vadose zone by teehnique ofPressure lnfiltrometer

Ichiro Kohno C Yuj i Takeshita ＆ Ali Mohaminad Jafar

（Received November 20 ， 1998）

Abstract：For reducing or bridging gap between small scalC laboratory血vestigations and large scale

・field inveStigations， a simple and ／portable model which is ・based on real situation and has a

．compatibility to laboratory core sampling， the．pressure． i皿filtrometer technique．．is．． introdu6ed fbr determination of field−saturated hydraulic conductivity． The field 一saturated hydraulic conductivity，

KFs， is obtained from measurements of the steady flow rates applying constant head as well as falling head p血ciple． The device is also used fbr field core samphng on which laboratory constant head as．

well as falling head tests is carried out． The field device is validated through comparison to laboratory core sample experiment and other existing methods． This paper describes first time a versatile field device representing good performa皿ce f（）r in situ detem血ation of hydraulic parameters in a short廿me．

Key Words： ln−situ test， Field−saturated Hydraulic conductivity， Matric flux potential， Wetting front， 3 D−flow，

1． lntroduction

In situ determination of variably saturated sub−soil hydraulic parameter in real situation is till． now a challenge for geo−technical engineers． Avoiding of laboratory sampling disturbance and small scale laboratory specimens， field experimental procedure for finding hydraulic parameters in the soil of vadose zone is one of the prereqpisites for controlling and monitoring of applications like design of irrigation water for a particular area， pumping wells in water plants， drainage systems in building pits etc， Also the gradual increasing global demands to protect vadose zone water resource free from present environmental problems due to pesticides ＆ waste disposal， the most important hydraulic p arameter of soil water movement is field−saturated hydraulic co．nductivity （KFs）． Any logical basis fbr characterizing血the vadose zone Inust be based upon且1ndamental concepts of field hydraulic conductivity．

Downward variability saturated flow through the porous mediuin dnven primarily by gravity， hydrostatic head and capillary fbrGe， Such flow so血etimes diverted in vadose zone by baniers causing lateral transport， or aOcentuated by preferred pathways promoting rapid downward transport． Accurate predicting the attenuation and eventual location of solutes or constituents in the vadose zone are directly related to wetting front movement of unsaturated・ flow of water． For actual measurement of water flow behavior in unsaturated soil， the measurement of hydraulic conductivity in the field is an essential task．

Three of the ．most important factors governing liquid transmission in unsaturated soils are field−saturated hydraulic conductivity， KFs， matric flux potential，￠．， and Sorptivity， S． KFs or field−saturated hydraulic conductivity refers to the saturated hydraulic conductivity of soil containing entrapped air． KFs・ is more appropriate than the truly saturated hydraulic conductivity for yadose zone investigations． The K （w） relationship， known mathematically as the Kirchoff transfbrm， has been shown Gardner田（1958）and others to be particula■ly usefUI fbr describing soii water flow，

Matric flux potential．￠，． is a measure of the relative importanbe of gravity and capillary for soil−water movement in the particular soil． Fine−textured soils， where capillary tend s to predominate，． have small Alpha values； and coarse−

textured soils， where gravity effects manifest themselves most readily， have large Alpha values （Fig． 1）．

Theoretically Alpha （ct＊） parameters can be explained by Gardner ［1］（1958） exponential function，

K （yJ）＝KFsexp（ocyJ）， wnh O〈oc〈oo and yJ sg O ^（1）

Where ct （m i） iS an unsaturated soil parameter simply called the Alpha parameter． The flux potential ￠（m2．s−i） is defined as：

D6partment of Environmental and Civil Engineering， Okayama U血versity， Japan．

＊coctora！ Student， The Graduate School of Natural Science and Teclmology． Okayama University． Japan

81

(2)

82

J． Fac． Environ． Sci， and Tech．， Okayama Univ． 4 （1） 1999

φ（の≡∫κ（卿，Whe・eΨ・≦Ψ

ヴj

Integrating between vi and w ＝O giVes：

（2）

｛zs．，＝ Ef／｝1一s ［eaur］v，o

a

（3）

ct＝＝

W．s Ki．z−i

｛ZSnt c

（4）

X， is the macroscopic capillaiy lengtli （Philip ［2］， 1985）．．Here Ki is very small in comparison with KFs， neglecting Ki for air dry soil in field situation we get the flowing relation．．．

κ禽・

a x 一一＝一＝一 Ea ＝

ip nl

yf

¹ ^（5）

wf is the effective wetting front potential for Green and Ampt ［31 （191 1） in｛riltration． ct ， ct， X，， and wf a11 of which are equivalent and represent single parameter estimates of the unsaturated hydraulic conductivity，

The values of Alpha depend on soil type and flow behavior is shown by an ideal

line as in fig． 1．

There are a iot of methods for

deterntination of hydraulic conductivity in

laboratory and in the field．

Nonhomogeneity and anisotropy of sojls，

fissUres， tension．cracks， and root holes

commonly encQuntered・ in unsaturated

soils can not be represented in small scale

laboratory specimens． ln reality，

laboratory results of hydraulic

conductivity are not represefltative data of actual field situation． lt is very difficult to simulate natural field conditions in the laboratory， Considering tlxis point， field test results are more reliable for analysis of flow in vadose zone． Among the direct

and indirect field−methods for deterntination unsaturated hydraulic conductivity， ASTM園standard guide D

5126−90 reviews alteniative field methods

ct

Log of（K）

＊

@ α Sand ，

z，Clay

α説 fravity elow on

α＝0

Suction head， w

o

Fig． 1 An ideal relation ofsuction head 一 ln K，

values of Alpha

Capillary Flow only

Slope inclicates

for available teclmiques for measuring K in the yadose zone， ln these techniques tiiere are a lot of limitations ＆ assumptions for measuring hydraulic and transport properties at the unsaturated zone． Soil is assuined uniform，

homogeneous and non−sweliing in the most of experiments； but macropores， gradient in water content， soil bulk density， soil layering and changes in soil texture all occtir near the soil surface， wltich can result ．in negative calculations of hydraulic conductivity． To minimize these limitations at some extent， Pressure lnf7iltrometer niay be the most appropriate teclmique in actual field situation for determination of hydraulic conductivity． ln the field and laboratory experiments， scale．effect is an old problelll fbr all researcllers． As Pressure Ir血ltrometer is a simple apparatus， we can fix it s main ring dimension according to our needs． Consequently scale effects of in−situ experiments can be reduced to some extend of our goal． @When anY ．infriltra／ tion is occurred within a ring， there is ap．

edge effect on actual flow behavior tlirough any porous media of unsaturated soil． This edge effect can also be reduced by increasing the dimension of main ring of Pressure lnfi｝trometer while in−situ experiments are performed on any soil surface at field level

(3)

1， KOHNO et al．／ ln situ determinatton offield−sat rated hydraulde conductivity

83

ノノ 1 ／／ 1

ノ l t i t l l I t 1 1−I X

N NI N N N N N N X x x N N x X

TTN．

Head

x Axisymmetry and ／

＼＼＼

̲2Dn・wp・tt・mノ／

／／／／／1

！／ノ

X s．s

x ^{一一一．}

xx

コヘへ

しニこへも

Fi・ld−sa…a・・d・・n・一

黶E，d。。n，）

111−

w．

Fig． 2 Field−saturated and unsaturated fiow behaviour of Pressure lnfiltrometer

Water且ow into unsaturated soil from Pressure lnfiltrometer starts with−an mitial transient phase−and血en gradually approaches steady state， Durhlg the transipnt phase， both the field−saturqted zonβa耳d un呂atU■ated zone i皿crease in

．size by migraimg downward and outward from the infiltration surface． After steady fiow is attained the field−

sat肛ated zone（bulb）remains essentially constant in sir：e and shape， wh韮e the unsaturated zone con血ues to㎞crease in size by outward movement of wghing front． This outward movement is symmenical about the venical axis of Pressure Illfiltrometer ri皿g． The size of field−satulated bulb depends on the dtmensions of i面ltration surface， soil texture and structure， applied pOsitive head gp soil surface，．and mitial water potential （suction head of unsaturated

soil）．

2． Methodology 2．1 Experimental model

Our experimental model is a device designed to represent a simplified version of reality． This device can be used fdr the field experiment as well as in the laboratory experiment （see Fig．3 ＆ Fig． 4）． The previous device was in different foml． The Mariotte reservoir was fixed at the top of血g． In our device， Mariotte reservoir is separated from 血g，Consequentiy the device becomes more portable and Stable at血e㎞一sitU experiMent． There is a possibility of distUrbance of soil wh丑e血g insenion into the soil sulface is done；but separat血g the Mariotte reservior， degree of distutbance has been reduced at some extent． During field experiment soil is assumed incompressible by any force applying丘om the surface． The possibilities of compression of surface layer of soil have been reduced血our modified arrangement Due to compression of su血ce layer， soil hydraUlic propenies may be changed and it may be

difficult to get our desired result． The field model having血g dimension，95㎜血side伽．＆65mm dep血of

insenion， is attached with Matiotte reservoi．r which control the positive constant head in situ test as well laboratory core sample test． The laboratory model is the modi血ed assembling of field device． A bottom cap， a top cap and a spacer porous disk are assembled with the laboratory model． Here constant head as well as fallmg head pr血ciple is．to be applied． Sbil core sample having dimensioned 95mm by 65mm is be taken by inserting ring into the soil surface of vadose zone ＆ laboratoty test is to be canied out saturating soil sample and keeping same condition of the field．

The clear standpipe is very usefu1 for falling head recordmg of field and laboratory・data． in vadose zone research works the falling head technique in the field is not established and available． This proposed modification is an evolution of in−situ falling head technique of measurement of field−saturated hydraulic conductivity in vadoSe zone．

For relatively less permeable soil， cohstant head technique can not give accurate measurement of血丘1tration rate．

Conside血g血s point for in−si加test鉛r血e soil applying制血g head a clear stand pipe having 20mm diameter is incorporated with Pressure lnfTiltrometer ring．

(4)

84

J． Fac， Environ． Sci． and Tech．，．Okayama Univ． 4 （1） 1999

Sea1 20 cm

AirTube

Acrylic φ35mm

Clear Stand Pipe

?・mド＼

．・・一＝一・．．

bontrol．Valve

一 11

幕g行1・… 1圏

．イ

φ20mm

．150cn lariotte

qeser＞．6i

P ＝5 早v oo

@ B

@ 艮

Cork

Stand

・．「「，曾 ■ ■

D．．． п･，噸．．．．．．．．．．．

Fig． 3 Field Model of Modified Pressure lnfiltrometer， Constant ＆ Falling head．

Acrylic

Clear Stand Pipe

￠20mm

Conむ二〇l Va玉ve 20 cm

＝nt

邑霧￠20mm

田艮

1一．．．岡．．．．・且 l l

v睡 rピ

＼

幽

⁴

Il・

ll

150 cm Mariotte Reservoir

Stand

Fig． 4 Laboratory Model of Modified Pressure lnfiltrometer， Contant ＆ Falling head．

For both the cases， field and laboratory model， air tube is to be used to control constant head． The control valve is closed while falling head test in the field ＆ laboratory is done． As a result the modified device is capable to measure in−situ hydraulic conductivity of a wide range ・of soil． ln the falling head technique in the field， time requirements is aii important factor． lt is very usual that if we need long time for recording field data of fine Soil using constant head principle， we have to talce special precautions at site． But it is possible to record field data by this device using falling head technique aiso， Therefore we can propose to measure field−saturated hydraulic conductivity of different types of soil （ 10 2cm／s 2 KFs．2 10−6cmis ） using this simple device．

2． 2 Basic equation：

Considering a point source out side of a ring，， an analytical expression ［7］ for steady flOw out of ring into rigid，

homogeneous， isotropic， uniformly unsaturated soil is described as following：

Qs＝2za （KFsH＋￠．）＋xa2KFs ^（6）

The above equation was developedご。且sidering a point source in a se血一infinite flow dolllain；but this was an approximation． Now if flow is considered withip a ring， the flow geometry differs significantly from that of a point source． The steady flow equation within the a ring is written as following as

Qs＝ a／G （KFs H＋O．）＋za2KFg．（7）

An implicit requirement of above equation is that no surface ponding （flooding） occurs outside the ring． The presence of a wetting front on the surface is， however， permissible and expected．

Now field−saturated hydraulic conductivity （KFs） can be calculated by two approaches−such as inultiple head aiid sing！e head． For multiple・head approach， equation （7） can be written as

(5)

1， KOHNO et al．／！n situ determinatton of fteld−sat rated hydraultc cond ctivity

85

Q，i＝ a／G （KFs Hi＋￠．）一Prca2KFs （8）

Qs2＝ a／G （KFs H2＋O．）＋7ta2KF， ^（9）

For si皿gle approac瓦equation（7）can be w亘tten as equation（10）using an impoltant relationship o負msaturated soil water fiow as O．＝ KFs／oc （eq 5）

Q，＝［（aH／G ＋7ta2．＋ af （otG）］ K，，（Reynolds ＆ Elrick ［5］ 1990）， _（10）

where Q is the steady state flow rate， H is the constant heact a i s the ring radius， G is a shape factor and ct an unsaturated soil parameter represeming the capillaiy component of 3D−flow of vadose zone． rlhe equation （7） can be written as for estimatmg shape factor， G

G＝ a（ H KFs＋￠．）／（Q，一za2Kfs） Where 1｛｝iO ^（11）

The values of G were determined numerically substitut血g the proper values of KFs，φm， H， d， and Qs values from numerical simulation ［7］． lt waS found that G is independeni of H greater than O．05m．（Positive head on soil surface）．

WD． Reynolds and D．E． Elrick［7］ developed an empirical relationship among the shape factor， G，血g radius， a， and depth of ring血sertion， d，㎞t6 the soil surface as following．

G＝ O，316 d／a ＋ O．184

This relation is also used fQr our calculation of hydraulic conductivity in the field．

（12）

3． Description of site for in−situ test

The site was selected at Okayama University campus． A ho！e having 160cm diameter and 150cm depth was dug． By b血ging granite soil from other place， it was depos孟ted near the hole side and to remove a11 gravel and stones having diam greater than 2mni，total soil was sieved by 2mm sieve． By using面s脚te soi1血e dug was五11ed． D曲g filling of soil layer by layer， manual compaction was done． ln this way an artificial land having

ロ箋

鵡 100

90 80 70 60

50 7

40 30 20 10 0

0．Ol むロま Grahl size in．

高

10

Table 1 lndex propenies of site soil．

Description

Est㎞ated value Un届omity coef匝cient， Cu ^4．8

Coefficient of Gradatio几

b。

152

Bulk densitV（wet）ρb ^1．83

Drv densitv．γd ^1．66．

Degree of saturation， S，

84％

．Spec澁。 Gravitv， G、 2．67

Void Ratio， e 0，608

Porositv， n 0，375

Fig． 6 Grain size of Granite soil of in−situ experiment，

cylindrical shape． of．160cm diameter＆150cm depth and．free drainage bouロdary was made．皿le site fbr field eXperiment was kept open under normal weather conditions about six mon血sL By natural rain．fa11 and compaction this arti血cial la皿d become a stable situation fbr field experiment． Five stationS−A，：B， C， D and E， having distance 40cm frem each other were chosen for conducting in−situ experiments． Soil index propenies （i．e．， soil grain properties ＆ soil aggregate propenies） had been checked and found as in table−1． After checking soil grain size the modified Pressure h而ltrometer device was set at the soil surface a皿d field measurements of steadv flosl・rate and falling head were performed and temperature correction at field was done to adjust the field result．

(6)

86

J， Fa6． Environ． Sci． and Tech．， Okayama Univ， 4 （1） 1999

From index propenies of soil， it may be concluded that our experirnental site is medium loose sandy soil． Grain size analysis was done by taking air−dry sample from the eXPe血lental site． There are a lot o．f血e血ods for checkmg hydraulic conductivity倉om g面n size culve． The prediction of hydraUlic conductivity ffom ・gra血size curves have to．

face a lot of limitations； but to get preliminary idea this grain size curve was developed．

4． Calcuiation approach

Conside血9血e simultai｝eous equations（M皿廿ple Head），血e values of KFs and inatric flux potential，φ、， are calculated by solving equations （8）＆（9）． Multiple head technique is that independent measurements of hydraulic conductivity and fiux potential are obtained； but limitation due to sQil heterogeneity in the form of layering can give us unexpected values of field−saturated hydrauhc conductivity（KFs）。 But血ouτcase soil was unifonn and

homogeneous．

Considering si皿gle head apProach and apPlying one positive head and site es血nating Alpha parameter， fiel（箕一 saturated hydraulic conductivity is calculated by using equation （10）． ln single head teclmique， only one water potential need to be applied to thg infiltration surface of unsaturated soil apd tlris tactic avoids the occurrence of negative value of hydraulic ． conductivity apd flux potential． lt ． also requiTes the independent measurement or estimation of Alpha （ct ） parameters The previoUs study suggests that the single head method yield field−saturated hydraulic conductivity （KFs） which is usually accurate to within a factor of 2 when Alpha （or ） is site−estimated and selected from the categories in Table 315］（Elrick et． al．， 1989）． Our calculated values obtained by multiple head and

．single head are compared graphically血figure l O． These values are obtained from in−situ test on granite soil．

5． Experimentai results and discussion

At fhlst ㎞ our 血一situ experiments

dimensions of the device were fixed on the basis of previous researchers ［5］ as radius of img， a＝ 4．75 cm， depth of ring insertion into ・ soil， d＝ 6．5 cm then the device was installed・

at the site and field data was recorded． A good relationship between steady flow rate

（Qs） and applied head CH） was found （fig．7）．

The consistency i皿． steady． floW rate to applied head results better performance of ． our in−situ deviCe・ than any． conventional field techniques． Wlule field experiment was conducting there was・no instability

problem due to high head as Mariotte

reservoir is separated from top portion of ring

which need care血1 ㎞sertion into the

unsa1瞳ated soil suriiage． In previou忘technique

斡奪 §

冨 α ぎ

￡あ還 2

の O．4 0．35 0．3 0．25 0．2 0．15 0．1 0．05 0

20

30 40 ・50

Applied Head， H （cm）

Fig．．7 A stable relationship between steady fiow ＆ aPplied head at in−situ experiment on Gtanite soil．

60

of Pressure lnfiltrometer th Mariotte reserVoir wa．s fixed on the．血g． As result it was very d亜cult to keep stable gondition in the field during・ the ring insenion into the unsaturated sdil surface of vadose zone rthrqe forces are act血g on unsat皿ated s6il surface． while experiment is nmn血g i皿．tlle field−such as gravity fbrce， hydraulic fbrce

．and capillary force． When higti ． head is applied the hydraulic pUsh has a great cOrttribution on hydraulic conductiVity． Constant head as well as fallng head pri皿ciples were applie．d血the field．soil at血e statiohs A， B， C，

D，and E． Core samples were taken from each site and constant head as well as falling head p血ciple were apphed at the laboratory model． lhe results are shown in table 2， Our results for G． ranite represenfs good performance of Pressure㎞filtrometer；but to reduce scale effect and edge effect for㎞一situ tesち the r血g dhnension should be aS large as．plactica1． From the present field observations du血g in−si加expe血nents oll Granite soil， tlle depth of r血g insenion （d） should be greater than 5 cm from soil surface to avoid surface ponding out side the ring． lt is ・also our observations that insenion of lnfTiltrometer ring may alifect the accurate results of field infTiltration， So during ilsertion of血g， soil surface shoUld be．leveled carefUlly a皿d．Pressure lnfiltrometer ring should be kept in venical．

positioh and gradu搬ly downward push is to be．applied． It is better to use a wooden smaU hammer to i皿sert the t㎞g into the soil up to the fU11． height of r血g which can avoid any water flooding out side tlle ring．

(7)

1． KOHNO et al．／ ln situ determinatton of field−saturated hydraMltc conductivity

87

Table 2 Field saturated （KFs） and Saturated （KLs） hydraulic conductivity of Grapite soil．

Site Field−satumted， KFs， cm／sec Lab saturated， KLs， cm／sec Constant head

Fal㎞head

Constant head

Fal㎞head．

A

^450x10 ⁴

^452xlO−4 ^{7．84xlO層4} ^{7．13xlO曹4}

B

^{4．．90xlO−4} ^{4．28xlO曹4} ^{6．64x10−4} ^{7．38x10曹4}

C

^{5．68xlO−4} ^{4．40x10層4} ^{6．48．x104} ^7．80x10¶4

D ^5．58x10 ⁴ ^{4．60x10−4} ^{7．25x10−4} ^{7．50x10−4}

E

^{6．12x10−4} ^3．98x1σ4 ^{6．32x10−4} ^{8．63xlO冒4}

Average

^5．36x10¶4 4．36x104 ^{．6．71x10−4} 7。70x10−4

Mea皿

^{4．90xlO雪4} ^{7．30x10−4}

SDF α0007

0．0001

To estimat，e the magnitude of variation of field and laboratory results at dii］ierent stations， mean value of KFs ＆ SDF

（Staiidard Deviation Factor） was calculated． For field experiment， mean of KFs＝＝4．90xlO cm／s， SDF＝O．OOO7， For Laboratoly experimenちmean of KLs＝7．30xlO−4，SDF＝0．0001． Laboratoly results are 1．5 times greater出皿field results；because du血g field experiment entrapped ah Ihake disturba且ce to the flow of water within pore space． The staiidard deviation is negligible； bechuse our experimental site Was ari anificial land of same tyPe of Granite soil． The sta皿dard deviat童on calcu重ation of KFs is the representative of variability of the site；but we need sufficient field data of KFs to check the variability of any site．

Validity of field results was checked by laboratory core sample test； but the core sample test was done by a new laboratory model of Pressure．Iniiiltrometer technique． The mean yalue of field−saturated hydraulic conductivity （KFs）

of each statiOn are compared to mean value of core sample results

セ

の

。．

．台

・忌

暮

昭

8 ヨ

壽

も彫

芭塁

の

9．5E−04

8．5E−04

7．5E−04

6．5E−04

5．5EO4

4．5E−04

3．5E−04 3．5E−04

y＝0．8384

@ R3＝0．

＋0．0003

P94

Best hne

om test dat

A

▲

B

｣

▲

△

Ide lline

4．5E−04 5．5E−04 6．5E−04 7．5E−04 8．5E−04 9．5E−04

F重eld−satしじated「hydraulic c onductivity， cm／s

Fig． 8 Relationship between saturated and field−saturated hydraulic conductivity for Granite soil．

The regression 1ine B： is obtained from experimental data and it represents a good relationshiP between saturated and field−saturated hydlaulic cqnductivity；becaロse the line B is near about pala11el to the li皿e of agreement（i．e．，

ideal l血e A ）． The variation of mean value丘om tlle．estimated value of regression li le was compared to the variation of mean value from individual value at different points （R2＝SS！SST）．

(8)

88

J． Fac． Environ， Sci， and Tech，， Okayama Univ． 4 （1） 1999

The equation of best fitted lmq is used to predict field−saturated values using laboratory core sample experimental

「esults ofG「血te soiHtis obsewed血t e即e血e al results agree wi血．P edicted vai es・

A Experimental values o Predicted values．

8．E−04

7．E−04

6．E−04 養 LJ一 s．E−04 濱 v 9 4．E−04 罵！二霧 3．E−04 竺，9 tr−t 2．E−04

1．E−04

0．E＋OO

123 4 56789 10

10 cases at in−s itu test

Fig．9C。mp、，i、。n。fp，edi、t，d。、i。，S。fKFs一高一si加，。p，㎞，姻・e、ult、．

tMultiple Head 一Sipg16 Head

l．20E−03

1．OOE一一〇3 臣

ii・i．． s．ooE−04 薫

g 6．ooE−04

・［； 4．00E一一〇4 ，範

m 2．00E−04

O．OOE＋OO

O 123456

Five cas es dn Granite s o il

Fig． I O Comparison of single head and rriultiple head calculation approach

Tr

^nd lin ^．f士om ^redlct

dvalu

es

●

o

^●

● ▲

▲

盈 ^▲●

T

^end li _ef㌃om ．??垂? enta1 alues

(9)

1．KOHNO et al．／∫n sit determtnatf切n qプ万属4−sαtUTαted妙d「a銘lde cond％ctivity

89

It is observed from the resultg． of 5 cases of otir site that single head approach of estimation of hydrauiic conductivity is more consistence ＆ less sensitive to soil variability （Fig． 10）． For．Granite soil of our site was homogeneous． Therefore in both the cases results were acceptable； but single head results are more stable indicating good performance． ln single head approach it is・ possible to measur．e field−saturated hydraulic conductivity in the field applying only one positive

head on soil surface and considering the value of Table．3［5］（Elrick et．al．，1990）

Alpha pararr｝eter in site estimation suggested values （table−3） and this measurement is needed short time

Single head approach of calculation

conductivity needs in

from Elrick

approach of

of hydraulic dependent measurement of A】pha

parameter or si．te estimatlon from Elrick Table 3，0ur estimation of Alpha pararneter was done（Fig lO）by fitting field data of lnK and suction head（Ψ）of Instantaneous proflle lechnique at the field of Tottri dune sand． Estimated value of A且pha．parameter was found O．39 cm匿1． According to Elrick（1990）suggestion

（Table 3）， the site estimation of A｝pha．v alue was O．36

ミ

cm 聖D It iS fouhd Ihat the． field竜aturated hydr律ulic．

conductivity（K，，：）． calculated by bonsidering O36 cm−l from E且rick table and by estimating from field data is shown in the table 4。 The variation gf ｛he resuhs of KF、

is．12％， whicb represents acceptable values of Alpha in table 3． Therefore our estimation agrqes with．the suggested table 30f Elrick et． al．，（1990），

Soi．蓋texture！structure category ^{＊一1}

ｿ cm

Com acted structureless cla e soil _0．01 Flne texture cla e and unstructured _0．04

Most． @structure soi置s行om clays

Dthrou h loam lncludi．n flne sand

0．12

Coarse and gravelly sands 036

Table． 4 Values of K，，，． on Alpha consideration

Alpha selection for

hn−situ test．

Calculated value of KFs

Elrick suggested

魔≠撃浮?Cα承＝0．36cm 1

2．4x10 cm／sec

Estimated value，

ｿ＊≠O．39cm冒1

2．7x1．0 crnlsec

SUCTION HEA D（m）

一〇．5 一〇．4 一〇．3 一〇2 ^一〇，1 o

r ^T

^T

y ＝ 39．19x 一 2．8272

t

！

（e

T

^o

一5

一10

一15

一20

一25 り

嗣．

D〈

a

o隅

oo

房E

ジ．

目z

瞠りD

A

z oり

の

Fig．11Estimation of Alpha（α）by using in−sltu experimental data．

Influence of Alpha parameter on K！：s was checked by using・field data of steady flow rate on Granite soil at our site of artificial land． Our field parameters were as following：

Steady flow rate （Q，）＝ O．22 cm3／sec， Radious ring ＝ 4．75 cm， Positive head on soil surface ＝ 50．6 cm， Shape factor （G）＝ O．631 and the range of values of Alpha were from O．1 ｛o 1 assuming rate of increment O．05． A relationship between field−saturated hydraulic conductivity and Alpha parameter is found （Fig． 12）． From this relationship it is observed ｛hat the valve of site estimation of Alpha parameter can be assumed within a range

(10)

90

J， Fac， Environ． Sci． and Teeh．， Okayama Univ． 4 （1） 1999

Fig．

6．00E−04

5．ooE−04

4．ooE−04 契 e

gf・ 3．ooE−04 臣

2．ooE−04

1．ooE−04

0．ooE−H）O

O O．1 O．2 O．3 O．4 O．5 ．0．6 O．7 O．8 O．9 1 A IPhal c m一

12 ln血uence of Alpha values ori field−saturated hydrauhc conductivity（KFs）when shape factor， G＝α631．

Sha efac

or，＝0．6 1

from O．25 to O．30 cm−i； because the calculated value of field−saturated hydraulic conductivity （KFs） becomes same beyond the value of O．3 of Npha．

1］㎡1uence of Alpha parameter was also checked us血g field data recorded during in一＄itu experi血ents． our site at dune sand has high hydraulic conductivity and rate of steady flow was also very high． Steady flow rate， Q，＝10．72 cm3／seg，

Radious of iing ＝ 4．75 cm， Positive head on soil surface ＝ 23．50 cm， Shape factor （G）：0．384 and the range of values of Alpha were from O．01 to 1 assuming rate of inerement O．01．

署g 2

3．50E−02

3．00E−02

2．50E−02

2．00E−02

1，50E−02

1．00E−02

s．ooE−03．

0．00E＋GO

hape

actor

G＝

^．384

e O．1 O．2 O．3 O．4 O．5 O．6 O．7 O18．．O．9

Alpha， cm i

1

Fig．13 h】血uence Alpha valu6s on field−saturated hydraulic conductivity（KFs）when shape fact6r， G＝0．385

It is evident from fig，（12） and fig．（13） that the Alpha parameter has a tittle influence of field−saturated hydraulic

盤臨撚膿鑑翻｛1 羅￡躍膿器諾：盤，島継課詳「avi噸e A so qpe

(11)

1． KOHNO・et al，／ Jn situ dete7minatlon offield−saturated hydrautic conductivity

91

6．Comparison of in−si加experimental results

Results of in−sim experiment ．with the results of other exisimg and conventional methods are shown below：

Table． 5 Results of different methods．

Site

Methods

Hdrauhc conductivi i

Totton dune sand In−situ test by．

oressure I皿丘1頓ometer

KFs＝2．80x10 2cm／s

To賃od d㎜e sand Ihstantaneous profile

狽?モ?獅遠

KFs＝2．00x10−2cm／s

Okaya卑a， Granite

唐盾奄P

In−situ test bV 7

oressure h面ltrometer

KFs＝4．90x10 4cm／s

Okavama， Gral並te 7

唐盾奄

LaboratorV p．I core− 7

唐≠?le

KLs；7，30x10−4cnl／s

Okavama，．

franite soil

Labofato琢

bonventional method

KLs＝3．60xlO弓cm／s

Laboratory core sainple result is 1．5・ times greater than field result as core sample was fully satuated．

Laboratory conventional method gives us 10 times higher valtie than the field value as in conventiollal methoct soil was disturbed and conditions were difi7erent than core sample．

7．

@Adyantages

1）Falling head as well as constant Ilead technique can be applied at二一situ hlvestigation of hydra曲。 conductivity using this device very Smoothly． 2） By making the device compatible to laboratory core sample test， Pressure Iufiltrometer device can be used in the field as well as in the laboratory． 3） The method is handy， less expensive， easy to installation．4）By usi lg only one positive head（耳）， we can est㎞ate field−saturated hydraulic conductivityσくFs）

with血lhour．5）The device is less susceptible to soil h6terogeneity than other conventional and exising methods．

8． Limitations

1） Ilhis device can not be applied at gravelly and stony soil．

2） Additional water flow along the wall of血g inserted㎞to unsatUrated d理． soil may affect the accuracy of measurement of actua1 amount of infiltration du血g in−sitU investigation of hydraUlic parameters of vadose zone．

9． Conclusion

if we look for last two decades， it has become clear that the in situ measurement of soil hydraulic properties is essential for dealing 4th the complexities of water and solute movement in the field． lt is therefore eSsential that

accurate field methods be developed for meas曲g出ese prope血es． if we s曲zed，血e 1曲res皿ts of面s

research works are as following：

1） The results of field−saturated hydraulic conductivity estimated by considering site estimation of Alpha parameter agree with the results of field−saturated hydraulic conductivity estimated by indePendent measurement of Alpha pararneter．

2） The value of laboratory ＄aturated hydraulic conductivity （KLs） is found approximately 1．5 times greater than the value of field−saturated hydraulic conductivity （KFs） which was determined by our in−situ device compatible to laboratory core sample test method．

3）

4）

5）

Alpha pararneter has a little influence on field−saturated hydraulic co 獅р浮モ狽奄魔奄狽?（KFs） in case of Granite soil．

Single head approach of calculation is more stable and less sensitive to soil． heterogeneity than multiple heads approach of calculation．

Our device is more stable to applied positive head and less vulnerable to the assumption of incompressible of surface soil during in−situ experiments．

(12)

92

J． Fac． Environ， Sci． and Tech．， Okayama Univ． 4 （1） 1999

10． References

1． Gardner，1958． Some．steady−state．solution of the unsaturated moisture flow equation with．apphcation to evaporation from a water table． Soil Sci．85，228−232．

2．P櫨ip，1985． ApProximate analysis of the bore hole permeameter in pnsaturated soil． Water ResQ皿ces Research 21．1025−1033．

3． Green and Ampt，1911． Studies i皿soil physi．cs． I Th．e flow of air and water through soils． JQurnal of Agric， S．ci・4・

1−24．

4．AS IIM Standard guide D−5 126−90， Handbook of Vadose Zone characteriza廿on＆Monito血lg． P−180−18L．

5．D．E．Elrick and W．D． Reynolds． Infiltration from C6噸nt Head． Well．Permeameters alld infiltrometers．

Advances measurement qf．Soil Physical Proper丘es：B血ging Theoτy into Practice． Soil Science Society of America， Special PUblication Nubber−30，1992．

6．D．E E㎞ck， W．D Reynolds， G．W Parkm and D．J Fallow．1992， Ponded lnfiltration from．撃№?and Auger． holes：

an historical perspective， Switzerland．

7．WD．Reynolds and pE． Elrick 1990． Ponded infTiltration from a single ri皿g I． Analysis of steady．flo、v， Soil Sci．

Soc． Of American Joumal 541233−1241，

8．D．J，Fallow and Elrick l 996． Field measurements of pressure heads with Guelph．lnfiltromet6r． Published in so重1 sci， soc． Am J 60

ラ

9．W．D． Reynolds＆DE． Elrick， 1987， A laboratory and numerical assessment of血e Guelph permeameter me血od．

Soil Science、 October 1987．．．、

10．WD． Reynolds and D．E． Elrick．1991． Determination of HydraUlic．Conduc廿vity．浮唐奄獅?a tension infiltrometer．

Soil Sci． Soc．of American Joumal 55：633−639．

11． 1．White and K．M． Perroux． Measurement of surface−soil hydraulic properties： Disc permeameter， Tension Infiltrometer and other techniques．

12．IWhite and M．Sully l 987． A lapid in situ meth6d fbr measu血g soil hydraulic conduc廿vity an面eld test Qf the bore hole permeameter． Water resource research 23 ： 2207−2214．