INVESTIGATION OF TOPSOIL HORIZONS OF MOUNTAIN SLOPES

TitleINVESTIGATION OF TOPSOIL HORIZONS OFMOUNTAIN SLOPES AS A BASIS OF EXPERIMENTALGEOMORPHOLOGY

Author(s) OKUNISHI, Kazuo; IIDA, Tomoyuki

Citation Bulletin of the Disaster Prevention Research Institute (1981),31(3): 131-150

Issue Date 1981-09

URL http://hdl.handle.net/2433/124903

Right

Type Departmental Bulletin Paper

Textversion publisher

Kyoto University

Bull. Disas. Prey. Res. Inst., Kyoto Univ., Vol. 31, Part 3, No. 281, 1981 131

INVESTIGATION OF TOPSOIL HORIZONS OF

MOUNTAIN SLOPES AS A BASIS OF

EXPERIMENTAL GEOMORPHOLOGY

By Kazuo OKUNISHI and Tomoyuki IIDA

(Manuscript received June 18, 1981)

Abstract

The results of simplified penetration test, in situ permeability test for its vertical distribu-

tion and other supplementary survey which were carried out in six test fields demonstrate that

the microstructure of the soil horizons and their physical properties directly reflect the current

geomorphological processes active on the slopes in question. Investigation of the structure of topsoil horizons in and around the landslide scars reveals the critical conditions for their

occurrence. In situ measurement of the physical properties of finely stratified soil gives

important information about the dynamic balance between weathering and denudation. The

methods of investigation as described here will be useful as a preliminary procedure for more

detailed geornorphologicaI experiments on mountain slopes.

1. Introduction

In the normal cycle of the geomorphological evolution, topographic changes take place mainly as removal, deposition and deformation of loose material (topsoil) beneath the ground surface. Investigation of the occurrence, properties and behavior of topsoil horizons is, therefore, essential for the study of the geomorpho-logical processes. This paper describes a method of in situ measurement of the vertical variation of the mechanical and hydraulic properties in topsoil horizons and in the substratum as the preliminaries for the observation of individual process. Results from some test fields are presented to demonstrate their geomorphological significance.

2. Methods

Vertical variation of the mechanical strength of soil can be measured in situ by cone-penetrometers. A variety of cone-penetrometers have been proposed and used so far, among which a portable one is suitable for the investigation of topsoil horizons on mountain slopes. Doken-type (also referred to as PWRI-type) simplified pene-trometer was used by the authors. It has a cone resistor of an apex angle of 60° and of a diameter of 30 mm, which is driven into the ground by the impact of an

132 K. OKUNISHI and T. IIDA

Table 1 Physical properties of different horizons in Aotani Experimental Basin as compiled by Okimura1)

Soil horizons I II III IV V VI

Pedological identificationAB B—CC D

N10 by Doken-type 0-3 3-..7 4-- 7-25 25-•50 >50 penetrometer

N10 by Kobe Univ. 2--5 5-.40 > 10 type penetrometer

Permeability index 0. 33 0.011 0.0004 P (cm-•min-1) P-wave velocity i 350

Vp (m.sec-1) Electrical resistivity p 102 46 20

(Q.m) Dry bulk density ya 0.93 1.08 1. 19 >1. 30 (g.cm-3) Porosity n 0. 53 0.48 0.46 < 0. 41

Angie of internal 15-..23 26--29 >30 friction 4 (deg.)

Cohesion C (kg•cm-2) 0. 13-4.14 0.09-.-0.12 <0.05 Thickness 10 60 25

iron weight of 5 kg falling down from the height of 50 cm. Another type of portable cone-penetrometer (Kobe Univ. type) designed by Okimural) is based on the same specification except that the diameter of the cone resistor is 23 mm. Though the Swedish sounding penetrometer is often used in testing the mechanical strength of the ground, it is not very suitable for use on mountain slopes, because heavy weights are needed.

Cone-penetration resistance is expressed in this paper by N,0 which denotes the number of impacts needed for each depth increment of 10 cm. Okimuran has found that each soil horizon defined in pedological terms has a particular value of

penetration resistance (Table 1), and that the latter is well corelated to the dry bulk density Id of the soil (Fig. 1). Since the dry bulk density decreases as the weather-ing proceeds, the cone-penetration resistance is also a good measure of the degree of weathering.

Comparing the vertical profiles of N,0 in landslide scars with those of the adja-cent part in the Rokko mountains, Okimural) has revealed that N10 obtained by the Kobe Univ. type apparatus commonly shows the values of about 10 at the depth of the slip surface. Thus, the mechanical stability of a slope can be examined by measuring the depth of the topsoil horizons at which Nim attains the critical value. Hatano2) has found that the penetration limit of a boring stick coincides with this depth. Boring sticks are useful for rapid sounding on mountain slopes.

The mechanical strength of the soil is correctly determined by the triaxial test. However, it needs undisturbed samples of a certain size, which make the continuous

Investigation of Topsoil Horizons of Mountain Slopes as a Basis of Experimental Geomorphology 133

2.0

t.9 - -

la - 0 0

1.7 - 0 0

1.60 00

0 0 1.5 0 6?8) 0 o

1.4•60 0 0 0

000 1.3•

0 00g,0a0 0 12• 00- !

. 0 0

1.1 - ooo0 OF82_ 10 8

0I os - •A.oo 9 0

0.9' • - MEAN 0

0.7 • 0 0

.6 • -

0.5( 3 5 7 10 30 50 70 100

N10

Fig. 1 Correlation between the dry bulk density yd and the cone-penetration resistance N10 measured by Kobe Univ. type apparatus. Closed circles

represent the average values from each horizons. after Okimurau

measurement of the vertical variation difficult. Moreover, a large effort needed for the laboratory procedure of the triaxial test make a dense measurement (in the vertical and horizontal direction) difficult so that one can obtain only a rough trend of the spatial variation of the mechanical strength of the soil through the triaxial test. It would be, therefore, practical to make a dense sounding by a portable cone-pene-trometer and to estimate, thereby, the three dimensional distribution of the mecha-nical strength. The triaxial test will serve as a standard apparatus there (see Table 1). Permeability represents a characteristic hydraulic property of the soil. Extreme variability of permeability of topsoil in the vertical direction, however, makes its laboratory determination difficult just for the same reason as with the case of mechanical streugth. Existing methods of determining the vertical distribution of the permeability in situ3) are not applicable to topsoil horizons, because they are oriented to the tests in great depths.

Okunishi4) has proposed a simple method of estimating the vertical distribution of the permeability by pouring water into an auger hole and by measuring the drawdown rate of the water level due to radial seepage. Accordingly, the value of the hydraulic conductivity K(z) at any depth z is given by

K(z) = r°2P ln (re) ( 1) 2ro

under ideal conditions, where ro is the radius of the auger hole, re the radius of in-

fluence in analogy to the pumping test for aquifer evaluation, and P the permeability

index defined as

134 K. OKUNISHI and T. HDA

P— d2v(z) (2) dz2

where v(z) is the time rate of the drawdown of the water level at the depth of z. Okunishi4) has shown that Eq. (1) is valid if the soil is loamy, but that if the soil is sandy, numerical treatment using a modified equation is needed because r, varies with time. In the latter case, P as calculated from Eq. (2) is no longer constant, even if K is constant. The values of P in different soil horizons are compared with those of K, Id and N10 in Fig. 2. The value of P could not be evaluated for the depth of 0-10 cm because the drawdown of the water level was too fast there. It is shown that the variability of P is much greater than that of K, which indicates that P is not necessarily proportional to K. It is, however, recognized that any disconti-nuous change in K is accompanied by a marked discontinuity of P. Therefore, this method of measuring P for K is effective for the identification of different soil horizons from the hydraulic point of view.

Some other measurements would be needed to complete the above procedure. Seismic prospecting has been modified, for example, by Nakagawa5) to examine a fine structure of smaller depths. Electrical prospecting by a portable apparatus is often convenient for investigating the subsurface distribution of moisture6).

N10 20 40 K (cm/s) Yd(g/cm3)

*--3.3x10-21.39

_Eno4-8 .3x10_31.36 .0040 ::1712.337E111:01-333111.434318 0.5:11." 5 .00024 <---2.7x10 3 1.47

0 -

---

a,

1 . 0 [.[ NISHIMIKAWA- I

L-10.7

1.5 - m

Fig. 2 Comparison of the values of N19 (by Doken-type penetrometer), permeability index P and the hydraulic conductivity K (by a laboratory apparatus)

for different horizons in Taruki Experimental Basin (cf. Sec. 3.4).


3. Results from some test fields

The methods described in the previous chapter were devised primarily for use in the granitic area. Granitic rocks often yield thick sandy regolith, called Masa-

soil in Japan, which makes the cone-penetration test and the auger boring easy for its homogeneity. It is noted that the disasters due to a dense occurrence of topsoil

slumps inducing debris avalanches and debris flows mostly take place in the region of weathered granite. It suggests that the topsoil is geomorphologically much more active in the granitic areas than in the other areas. Thus, most of this chapter deals with the result of the field survey in the granitic areas. The location of the test

fields is shown in Fig. 3.

3. 1 Aotani Experimental Basin in the Rokko mountains

Aotani Experimental Basin was established in 1973 at the southern foot of the Rokko Mountains for the research of landslides?). Geology and geomorphology

of the Rokko mountains have been summarized by Huzita et al.81 The regolith from granodiolite in and around this basin is rich in clay minerals, which provides

good conditions for vegetative growth. An intensive field survey of the structure of the topsoil horizons was carried out

TEST FIELDS I ROKKO MTS.

2 FUYUNOGAWA EXP. BASIN

3 TANAKAMI MTS. 4 N1SHIMIKAWA MTS.

5 MT. YAKE °57 6 R. TAKATOKI

6 \\,

41/ C. 1* 03

01 t

02 0

100 km

Fig. 3 A map for the location of the test fields.


0 10 20 30 40 50 60...,...,70.,..:Ow,.,....90 100m =NW11111 111911111),"'4'44.417.1,!beilrimiT--i

.-Imm,,...,.,tho..,,...0 -80m Id!IF.""',4s111II.,..!.F''''''''''',IrP'."%kirif4,,,.:.'-'qiii15tT.

.qui-....4.....4,..,. 20 LiJ

SOIL HORIZON IDENTIFICATION BY DOKEN-TYPE PENETROMETER 25°

-_

, -70 -'- 70 so 90 100

11 HORIZON_ 45 24013.2,,-1---1".,...via

LEGEND0.5_ _„ ,, .. ,. 74,;.0 20 tram,.1iiitr.7410-E- CT] I HORIZON ,,-°'--_,____,_4.11111.-.Mr,:x

-60 MIL/---.:.'41111 51- OMI,,360 146 104'111tie11:2 .02j- 6a)III HORIZON ..., , -.- MI III-IV HORIZON 70 SOIL HORIZON 25°

Egg IV HORIZON -` SPECIFIC RESISTIVITY IDENTIFICATION 1 .. EIM V HORIZON ,, (1). m I BY KOBE UNIV. TYPE — PRESENT GROUD SURFACE PENETROMETER

50 -- ORIGINAL GROUND SURFACE 1.- % --- PENETRATION LIMIT OF BORING STICK 1-.'' LANDSLIDE BOUNDARY IN V

, I m/sec )I,•..,IN 1967 BOUNDARY IN Vp 1 m / se c 1•

-40-.-BOUNDARY IN p Isl•rn) 4o0 ,....e.-'„, - i, '1'00 • -..t,... ''‘'00

0,..u1.- ...03. a.,0 '11106 ',3.,,,00 -30

1:70c, 1, 1,020 30 40 50 60 70 so so00 ioo

Fig. 4 Identification of soil horizons in a longitudinal cross section of a hollow named R-1

according to different methods of survey after Nakanishi et al.9)

m M 1 HORI ZON WE II H:P.ZON

411111:e11111(23 111NJHORIZON :::: v HORIZON

3-IsIlIIIA'. ee..110 2-•.III ..._ Io-Ar

mif'..r

le Air .„;h.;111111111111111111 ,16'

" -- - o -

O ; 1.0 :5 20m

Fig. 5 Identification of soil horizons in a transverse cross section of R-1 according to Nu.

by several researchers9,10) using their original techniques at a hollow named R-1

(in this paper concave-contoured unit slopes are referred to as hollows). The result is illustrated in longitudinal and transverse cross sections as Fig. 4 and 5, where the identification of tospoil horizons is made according to Table 1.

A landslide scar at a horizontal distance of 70 m in Fig. 4 is due to a landslide in 1967. It is, however, noted that the distribution of N10 and the elastic wave velocity shows more pronounced concavity in the underground structure, which is due to a fault crossing R-1 at this points). Historical descriptions on the other hand,

give an evidence of repeated occurrence of landslides. It is estimated that once a landslide takes place on a slope, the landslide scar will begin to be gradually buried by the loose material due to its production by weathering and its migration from the


N10 ,SOIL COLOR 0 20 40 60 (-0•-cm)(cm.Imin) 0

GREYISH BROWN SURFACE SOIL 5 0.2500 ---- LIGHT BROWN

.046 SAND

LIGHT BROWN 0.4 036 100LOAMY SAND

E ( REDDISH ) 0.6 .006 ..:• LIGHT BROWN

' SAND (HARD) a_

L.L1 0.8o 0o

0 BROWN SAND (DRY a HARD)

1.0 AOTANI. R - 6

Fig. 6 Comparison of the vertical profile of the topsoil horizons according to differ-

ent methods at a measure point at the center of the hollow R-6.

CASING PIPE GROUND

A SURFACE

MOR TAR A A A A HORIZON -I

A

A INSULATING A A

BRIM 4 A A

SAND 40:

HORIZON -II

• •

, • '. :• G.W.L.

.111111r/ A A A A HORIZON -Ill A

MORTAR 4 A &AA A A

A ^ti A /

Fig. 7 Structure of the observation well for the water level of throughflow (not to scale). The water in the well is connected solely with that in horizon-II.

upper part of the slope, preparing the next landslide. This interpretation is support-

ed by Fig. 5 which illustrates that the horizons of loose material is thickest in the

bottom of the hollow at which the transverse profile is most concave. More detailed

discussion about the mechanism of the development of hollows due to repeated occur-

rence of landslides has been given by Iida and Okunishin).

Similar investigation was carried out in another hollow named R-6, the result

of which is shown in Fig. 6. The auger hole for the in situ permeability test and

the sampling for soil color was drilled at the point where the cone-penetration test


8 - Aotani, 1976

6- 1 HOURLY RAINFALL (min)

4 - - -

2 -

1 0—1-IT-rl—r—r—I--1----if 1. 8- • •

6- •

4- . •

WATER LEVEL FROM THE BOTTOM ( cm) 2 - •

0, j 1

.

,

0 6 12 18 0 6 12 18 0 6 12 18 0

AUGUST 3 AUGUST 4 AUGUST 5 Fig. 8 A stage hydrograph of throughflow in an observation well at the same location with Fig. 6.

had been carried out. It has again shown that the soil horizons as identified by different physical properties essentially coincide with each other. Fig. 6 was used by Fukuol2) who observed the displacement, deformation and internal stress of the soil horizons due to rainfall at the same place.

Fig. 6 suggests that subsurface storm flow (or throughflow), which may cause surface landslides, will take place at a heavy rainfall at the depth of 65 cm, because the hydraulic conductivity K evaluated from Eq. (1) is only about 3 mmill below this depth, while that of the overlying horizons is estimated to be more than 40 mmih, much larger than the intensity of usual rainstorms. Thus, the observation of the water level of the throughflow was carried out in the auger hole described above. The structure of the observation well illustrated in Fig. 7 improves the response to the occurrence of throughflow by preventing the occurrence of dead water in the well. It also prevents shallow water to enter into the well through the interstices around the casing pipe. A multi-electrode water level detector designed by Okuda et a1.13) was set in the well for automatic recording. Only one occurrence of throughflow

(Fig. 8) was recorded during the observation period of August to October in 1976. The maximum water level took place about 20 hours after the peak of rainfall.

3. 2. Fuyunogawa Experimental Basin in the Ryumon mountains

Fuyunogawa Experimental Basin was established in 1975 for hydrogeomorpho-logical study of mountain slopes. Geomorphological processes in this basin have


0 10 20 30 40 50 60 70 80m 50m .i. N., •to Ala

0 0 10 20 20 48 58 60 70 80 90 0 10 20 70 40 0 10 20 30 40 0 10 20 30 40 , ........ y. 0 P • 0 — o 02 0

0 0 i 50m LO,:',,, .0;o.s —1* 4 0 410 ''' : :s..,...200 00 9 6 D 70 '.. ..,. ac,, -;- .,.., 0

r—J ' -..,.. cl- skt

,-. < 0 ••.`,....ci.• 40.a,,,, 030 050 30 ••,,

,..cz0-",::,,,... '4,,.. pl '....".... ir)

`;.., o

;, -- -,-,

.to 30.,,z., © 20 .:, ''- - g

----.,,..z.. -,....5.,. 0 ,...<10 '.......,

20. to 2'136 90 1? 20"1 *10 4ol02;"3, to 10 20' 4 011° 20.."'''',...„..<,,,,,,,,,

, . -----0-,..03

1Hipri--,,,..„...'....0 .,.`,.

a A100 ,6

01.0 A20 A50 Ae°:-.,,OD < .:.•-, 10 A 5 -,, 0 ‘6•6

‘‘‘

kS 5)) 4s,,,.,1 ,, 0° *rte`̀ ..*..ao 0',... 10 20 30 40 50 60 70 BO.,.... • 0

Fig. 9 Longitudinal cross sections of Fuyunogawa Experimental Basin along the ridge (line-D) and along the bottom of a first order valley (line-A). The isopleth lines of N10-10 and

Nio = 50 are shown by broken lines and chained lines, respectively.

been discussed by Uemura14) on the basis of topographic analysis and the obser-vation of water and sediment discharge. The bedrock in this area consists of Ryoke

granite15). The topsoil horizons are rich in clay minerals and organic matter. It is estimated that an active tectonic movement related to the Median Tectonic Line crushed the bedrock and thereby stimulated the chemical weathering from which resulted clay minerals15). The topsoil horizons are rich in organic matter and suggest a high rate of plant growth.

The authors made an investigation of topsoil horizons along some measure lines within this basin. Fig. 9 shows the result in the longitudinal cross sections along the talweg of a first order valley (line-A) and along the ridge (line-D) border-ing the latter from another valley. Remarkable concavity of the ground surface is found at the horizontal distance of about 16 m on the measure line A, whereas the isopleth lines of N10 are fairly straight. Since surface slumps take place predomi-nantly in hollows and at valley heads, such concave feature are estimated to be landslide scars of the past. Convex profile found at the part of the horizontal dis-tance between 50 m and 80 m is attributable to local rock control or to an effect of minor faults. The pattern of the vertical change in N10 varies site to site suggesting a heterogeneity in the weathering zone.


3. 3 Experimental slope in the Tanakami mountains

Field experiments on the topsoil horizons were carried out in 1970 on a low ridge

(Fig. 10) in the Tanakami mountains. Though the result has been partly reported by Nakagawa5'6), Okunishi16) and Esumi and Okunishil7), an examination of the data from the geomorphological point of view has not yet been made.

The bedrock (coarse grained biotite granite named Tanakami granite) is covered by its weathering product (dry Masa-soil) which contains only a small

quantity of clay minerals. Fig. 11 shows the cross section of the weathering zones of this ridge according to the seismic prospecting by Nakagawa5) and to the ground survey by Matsuda and Okunishi18). The top zone (Zone-I) consists of Masa-soil, the P-wave velocity being about 400 m/sec. Zone-II consists of a mixture of Masa-soil and rock pieces, the mixing ratio varying with depth. At the bottom of this zone, the rock becomes the bedrock which contains Masa-soil in its dense fissures. Zone-III is the bedrock with the P-wave velocity of 4.0 km/sec. Though a ground observation gives such impression of deep weathering as revealed in the Rokko

mountains7'19), the cross sections in Fig. 11 present the structure characteristic of surface weathering. Because hydrological analysis of the land waters in this area

-5i-/ I / / . . \\'NN"------7\.97).).: /,, ,'-'--

-Pr' ,,

- .--i .

i 8 S.,

1‘. 300 J:'f, ' '

d/,......../„.57...:i,,,,,,,,,,, if&a, i

,,

I ,

i 1

• ;I : I ' I / HIC100"Crtl 5 '''' ^ 9 :149\ 41.111111°1 #***1.1 kJ, ,. :I: F 1 \/ e ...

,Ca ,"5 ) .'

li ,',' .'f:'11.. /..,-----'° 'C'''' -

1`.. 1 Ie'.. - \..../ : Vis ' ' .....-0,-__.-__-- -----------,_:,--- \ 50 10 0 m O

t e

Fig. 10 A topographic map of the experimental slope in the Tanakami mountains (contour interval: 4 m). Dotted line and chained lines on the slope show the boundary

between entirely bare part and sparsely vegetated part and the break in slope angle, respectively.


by Okunishi and Natsukawa20) suggests inactive

chemical weathering, the predominant agents of 3o1 ./

\

weathering in this area is estimated to be physical-`,, iiS 5 ones such as unloading, temperature change and 29o]

moisture change.------„_ III

The slopes shown in Fig. 10 present a typical1 300-—1--- landscape of waste bare-land in this region's). Trees .,

II are sparsely growing outside the dotted line stret- 290 S 4 ching their roots into the fissures of the bedrock. Al

There is a forest land to the north-east of the ridge 300

in Fig. 10, where topsoil horizons are obviously11 290 diff

erent from those of the bare-lands. Fig. 12----------S3 ill

shows that theparticlesize distribution varies with 300

,--1-`, depth from the ground surface to the depth of 80 cm in the forest land, whereas essentially no variation 290h] S 2 i

s found in the bare-land. Because the bare-lands in ̀.

in this area were caused by overlogging and accele-300--I

-

rated erosion21), such difference is attributable toa 290-

_---,.------ the removal of topsoil in bare-lands. Accumulation,' "' S 1 of coarse grains at the ground surface of the bare- 280-

land in Fig. 12 is caused by the selective removal of

- finer sand and silt by running water's). Vertical 270- -- 20 4bm'

profile of N,0 in the bare-land (Fig. 13) also shows-40m20 0 0 the lack of A and B horizons. A smallvalue of N10Fig. 11 Vertical cross section of the experimental slope along at the depth 0-10 cm is due to the loosening of different lines shown in

particles by freeze and thaw cycles's). Though Fig. 10.

FOREST AREA BARE SOIL AREA

00 0 50 100%0 50 100%

. ... — . o

E 0

0

.

0 0

?0 .,, (2 (7 i-I0 a _ 0 i5C",_',)o .0o '8

gq.1 . . - i . 0

a_T1,.. ino o0co a.o acr, O..2a,0 w_.co. . 00

o-/ ,I .O0000 ...1- _ vcoN.41-

/ 0

N

100-1Oa, 11050.°a )a 'C 1 ;7'''1:1'7c''

!o

,00 I i. ...

Fig. 12 Vertical variation of the particle size composition (in weight) of the soil in a bare-land and in a forest land.


not very obvious, a jump of N10 is found at the N10 0 25 50 depth of about 80 cm, which is thought to corre-

0 1 spored to an discontinuity in electrical resistivity

at the depth of 90-100cm reported by Nakagawa6). Behavior of soil moisture within the topsoil

horizons in this area is discussed by Esumi and Okunishi17) and Okunishi22'23). Accordingly, the retention capacity of the topsoil in the bare-lands

E isso small that the suction of soil moisture easily 50 -

exceeds the permanent wilting point when there are several successive fine days. In the forest lands,

on the other hand, the topsoil is always kept moist

f2i because A and B horizons have a large retention capacity.

100 - 3. 4 Experimental basins in the Nishimika-

Fig. 13 Typical N10 profile on the wa mountains ridge in Fig. 10. A

swarm of landslides took place in the Nishi- mikawa mountains in 1972. According to Okuda

et a125), the density of land-slides attained as much as 200 km-2. The bedrock of this area is coarse-grained granitic rock referred to as hornblende bearing biotite adamelite24). Because the weathering product (Masa-soil) contains only a little

amount of clay minerals, most of the hillslopes are usually dry, so that the vegetative cover is poor, consisting mainly of red pine and shrubs. It is estimated that waste bare-lands were once predominant in this area21). Some ridges have a very sparse vegetative cover even today.

In the research project on landslides organized by Tanaka7), three test fields were established for the investigation of landslides, among which the field experiments on topsoil horizons were mainly carried out in an experimental basin called Take-daira-NW Basin25'26). Fig. 14 shows the result of the in situ permeability test carried out there. It is shown that the soil color is consistent with P-value and that the topsoil is thinner in the hollow than on the ridge reflecting the erosional processes acting in the hollow (note that the longitudinal profile of this hollow is convex at this point).

Another experimental basin (Taruki Experimental Basin) was established by the authors and field experiments on the topsoil have been carried out there since 1977. Structure of tospoil horizons in the representative cross sections was examined using Doken-type cone-penetrometer, and on this basis, the occurrence of landslides and resultant topographic evolution were analyzed by Iida and Okunishill). Comparison of the results of the in situ permeability test and the laboratory determi-nation of hydraulic conductivity of topsoil horizons in this basin is shown in Fig. 2.

The structure of the topsoil horizons shown in Fig. 2 is different from that


P (cm-I—mirk-1) BLACKISH 0.0 DARK BROWN

SURFACE SOIL SURFACE SOIL DARK BROWN BROWN

SURFACE SOIL ---- SURFACE SOIL BROWN 06 (3

SURFACE SOILLIGHT BROWN LIGHT BROWN .068•SANDY LOAM SANDY LOAM - 0.5 -

LIGHT BROWN LIGHT BROWN LOAMY SAND LOAMY SAND .0014

LIGHT BROWN .0009 SANDY LOAM LIGHT BROWN -LIGHT BROWN I 0

LOAMY SANDSAND .0004

LIGHT BROWN ].-• • • SAND LIGHT BROWN

SAND (HARD)

TAKEDA1RA — NW I 5 m- ..• .0004

Fig. 14 Vertical profile of the permeability index P and the soil color on a ridge (right) and in a hollow (left) on the same contour line crossing the side slope of a first order valley in Takedaira-NW Basin.

observed on most slopes in Taruki Experimental Basin in that the horizons where N10 value is about 10 are fairly thick. Because a landslide removed the soft layer

(the horizons where N10 is less than 10) of a thickness of about 1 m at this site in 197227), the original topsoil prior to the 1972 landslide had been thicker than that shown in Fig. 2. Though some possible causes could be estimated for such a struc-ture, no conclusive evidence has been obtained yet. The surface 20 cm of the toposil in Fig. 2 in which N10 is less than 10 and K is around 10-2 cm/sec is estimated to have been formed in the period of five years after the landslide of 1972.

Observation of soil moisture and throughflow using tensiometers and obser-vation wells has been carried out by the authors28) in this experimental basin since 1979. In order to investigate how the topography and the structure of the topsoil affect the movement of soil moisture, tensiometers were set at several depths of the top and the bottom part of a ridge and a hollow. Accordingly, soil moisture is highly variable and its flux is upwards except during and immediately after the rainfall on the ridge, whereas in the hollow, soil moisture is constantly high and its flux is always downwards feeding the baseflow. Structure of the observation wells for the throughflow was the same as shown in Fig. 7, but the bottom was set at the boundary between the regolith (10< Ni0�50) and the bedrock (N10>50). Though the permeability of bedrock is not correctly known, a high contrast in N10 between the regolith and the bedrock suggests a high contrast in permeability. Occurrence of throughflow was detected in this well even at the time of moderate rainfall (Fig. 15). Since the throughflow makes a preferable conditions for chemical


ur.1, Exp. Sas 0 II ' fl"

5

HOURLY RAINFALL (mm)

10-

1 5

20- '111

1 WATER LEVEL FROM

10- THE BOTTOM (cm) -

6 12 18 0 6 12 18 0 6 12 18 0 October 18 October 19 October 20

Fig. 15 An example of the stage hydrograph of throughflow in an observation well in the Taruki Experimental Basin.

weathering28), it is expected that more detailed investigation into the hydrologic

properties of topsoil horizons will offer valuable information about weathering condi-tions.

3. 5 Catchment of the Toge-zawa valley on the eastern slope of Mt. Yake

Field experiments on debris flows has been carried out in a gully called Toge-zawa (also referred to as Kamikamihorizaw a) on the eastern slope of Mt. Yake by Okuda and co-workers29). At the time of the vapor eruption in 1962, volcanic ash brought about a thick deposition at the head part of this gully. Thereafter, this part has remained waste because of heavy erosion.

A cross section of the geologic formation can be easily observed at the side walls of the gully since it has entrenched into the ground as deep as 10 m. Andesite basement is covered by a strata of andesite boulders (pyroclastic deposition) lami-nating nearly parallel to the ground surface. Above the reddish horizons of soil and boulders which overlie the pyroclastic deposits, there are found the roots of bamboo grass and trees which suggest the level of the past ground surface. The top hozizon is a whitish deposit of volcanic ash and volcanic sand of a thickness of 60-100 cm which occurred at the time of the 1962 eruption30).

Field survey of topsoil horizons was carried out by Okuda et al.30,31) as the

preliminaries for a hydrological investigation into the occurrence of debris flow. Distribution of NI0 and P in a vertical cross section along a longitudinal measure line is shown in Fig. 16. The N10 profile at each measure point is shown in Fig. 17. Though the topsoil is soft, the existence of stones in it caused larger values of N10 thus introducing a kind of noise in this profile. However, the trend of N10 varying


HORIZONTAL DISTANCE Y,4 KEDAKE

00 0 100/2 200 300m ^ ...____!„-------- 2jillIllIllMll 3 E05 _ —3 32 ir-.---.. 4/*--4

Lu 1—•.--...4. , o_,4

1------ ....,-\3\%4j-.,^./.\\ // CLASSIFICATION - ..

. n 4 'ss..'""\BY NVALUE -.../ 3- BOUNDARY IN pV10

I 5- I I I I I I I 1 I I I I I

IA 1B 2A 2B 3A 3B 4A 4B 5A 5B 6 7 7B 8 0,1 II I I I I 1 1 '

.003 .01-2,01......_——4'.17)31 — —----—_ m"-----62-1—— — E80'1 0027 .0059 °°30003°'.0003 I 0 5- *008^ .006 ."- ......._,„—-------__.000L_ 1— .013 .0004-.. ......—0018 •-•. --... — --

w ---------------- CLASSIFICATION ° I - BY P VALUE

Fig. 16 Identification of topsoil horizons by N10 and P in the vertical cross section along a

longitudinal measure line on the eastern slope of Mt. Yake.

PROFILE OF N10 ALONG LINE- I Y,4KEDAKE 0 5 10 20 50 Igo 3 5 10 20 50 fop 2 5 10 20 50 100 3 5 10 20

I-1 I 0.4• I-1

r-i1 1 0.8-1 1_1

7 L_

IA IB 1.2-1 2A 2B ----

E 0 2 5 lo 20 50 I 10 20 50 5 10 20 5.0 5 t0 20 50 ...-.. 1 `-l\li•;C) 1 _ ! I

0.4- I 1 j

1 LL _?I-1 08- 1 H • A 3A 3B

4B 4B 12- a .

Lu O 5 1020 5010 20 50 109 lo 20 50 Igo 5 to 20 so lo, •

I-I o.(____, 1-1_7_, 0.4----1 1

.___, i . 8-7

. 5 A . 5 A6 i

. Fig. 17 N10 profiles at each measure points shown in Fig. 16. Measurement was twice carried out at the point-7.

with depth can be roughly classified into several standard patterns which reflect

the conditions of deposition and the processes of weathering and migration of the

material. Since the observed values of N10 present a log-normal distribution,

topsoil horizons have been identified in Fig. 16 according to the criterion 2Th-1< N10

<P for the n-th horizon.

At the in situ permeability test, the side wall of the auger holes became irregular


30- RAINFALL INTENSITY 20- ,Thrim/— (mm/h) 10-

0 • 30-

RILL DISCHARGE

20' ( m3/h )

10-

0

6 - WATER LEVEL IN THE

4 /\WELL ( cm ) 2 •

0•• 610

SEPTEMBER 4, 19 7 8 Fig. 18 An example of the stage hydrograph of throughflow (bottom) in an observation

well installed near the measure point 2B in Fig. 16.

because of the stones in the soil. It may possibly have introduced a large error into the values of P. Classification of topsoil horizons into several subdivision was, however, possible because of great vertical variation (with the range of variation of

about 103). Four observation wells for throughflow as explained in Sec. 3. 1 were

prepared according to the vertical distribution of the P-value. An example of the observational results from a well installed near the measure point 2B (see Fig. 16) at a depth of 44 cm is shown in Fig. 18. The water level in the well, or the stage of

throughflow, attained its peak about one hour after the peak of the rainfall, when the discharge of the adjacent rill had already decreased to nearly zero. A large rate of the recession of throughflow shown in Fig. 18 is attributable to the perco-

lation of water into the underlying horizons.

3. 6 Catchment of the Takatoki river

The Takatoki river pours into Lake Biwa from the north. The bedrock of the drainage area is paleozoic rocks interbedded with chert, shale, sandstone and

schalstein. The weathering zone is highly heterogeneous because the dip of the strata is great and a dense fault-joint system occurs. Many lineaments which reflect

a particularity in vegetation along linear geologic structures can be easily traced in an aerophotographs. It is, therefore, suggested that the heterogeneity in the

bedrock has a significant effects on the topsoil. According to a hydrological study by Okunishi32,33), the coefficient of direct runoff of this river is somewhat larger than that of other rivers pouring into Lake Biwa, though the base flow is highly

• stable. The valley-side slopes of the Takatoki river is very steep except at its head

reaches. A longitudinal profile of one of such steep slopes along the southern divide


of Obanashi Experimental Basin established by Okunishi33) is shown in Fig. 19.

The slope segments bordered by some breaks become gentler as they become higher,

which suggests that this area has been upheaved by geodynamic stress34).

The result of a preliminary survey of topsoil horizons with a boring stick is also

shown in Fig. 19. At the bottom part of the slope, the survey was carried out at an

interval of several meters. Because a high content of angular pebbles in the topsoil

horizons caused a large random variation in the penetration limit of the boring stick,

600

LLI

500 I— I= --1

SLOPE PROFILE 400

300 • • 80• THICKNESS OF TOP-SOIL

• HORIZONS (cm) ---- 40 #‘41^("\--r--

0 0 100 200 300 400

HORIZONTAL DISTANCE (m)

Fig. 19 Thickness of topsoil horizons as measured by a boring stick along the southern divide of the Obanashi Experimental Basin.

N10 N10 0 20 40 0 20 40 60 80 0

0.5

LL

I

T

SITE-2 15-

SITE - rl

2•

Fig 20 N10 profiles at two measure points on the northern divide of the Takatoki

river basin.


the trend of the variation along the measure line is not obvious. Several measure-ments were averaged for the measure points at the horizontal distance of 320 m and 460 m, respectively. Because these measure points belong to older geomorphic cycle than the bottom part, the topsoil horizons at these points are thicker than at the latter parti8).

The result of the sounding with the Doken-type cone-penetrometer carried out on another ridge of an altitude of about 750 m which is regarded as a relic of the upheaved peneplain34) is shown in Fig. 20, where site-1 represents a flat but slightly inclined strip on the ridge and site-2 the bottom of a small rill entrenching the strip to a depth of about l m. Because the topsoil contains plenty of pebbles, N10 at site-1 is subject to a random variation. However, the horizons can be roughly subdivided into three horizons at depths of 0-30 cm, 30-160 cm and more than 160 cm, accord-ing to the values of Nlo. The difference in N10 among these horizons is attributable to the difference in the content of pebbles in the soil. It is also suggested that a

periodic vertical change in N10 reflects the geologic structure of the mother rock. The horizon where N10* 10 at site-2 is estimated to have been formed by the deposi-tion in the ri1132).

Though the investigation of topsoil horizons in this area was not complete as a basis for geomorphological experiments, it has been shown here that the outline of the structure of the topsoil horizons can be surveyed with such simple instruments notwithstanding the significant heterogeneity in the physical properties of the materi-al. Further field experiments are needed to explain the difference in topsoil hori-zons between different geomorphic conditions.

4. Discussion and conclusions

There is some difference in the abundance of data of the field survey of topsoil horizons among the six field tests. It was, however, commonly suggested that the structure of the topsoil horizons of mountain slopes reflects the predominant geomor-

phic processes acting on the pertinent slope. In the granitic areas, the microstructure of the topsoil horizons was revealed by

the combination of cone-penetration test and in situ permeability test, and the result was interpreted in terms of geomorphic processes. In the paleozoic area and on the volcanic slope, fundamental structure of topsoil horizons was revealed by the same method notwithstanding the noise component in N10 due to the admixture of

pebbles in the topsoil. It was shown that the boring stick method is useful in the survey of topsoil horizons over a wide area, though the fine structure of topsoil hori-zons can not be obtained.

Detailed analysis of the structure of topsoil horizons makes it possible to test the hypothesis about the geomorphic evolution by examing whether the theoreti-cally predicted processes of weathering and mass transport can produce such topsoil horizons as actually observed. Iida and Okunishill) have successfully carried


out this procedure concerning the geomorphic processes which cause landslides. Measurement of the vertical distribution of the physical properties (e.g. mechanical strength and permeability) would lead to a quantitative discussion about the mass balance of topsoil due to the production by weathering and the removal by various transporting agents.

Appropriate instrumentation of the field experiments on mountain slopes would be possible on the basis of a preliminary survey of the topsoil. In this paper, it was exemplified by the instrumentation of the observation well for throughflow. It is expected that the field experiments on weathering, surface erosion and soil creep will be effectively designed on the basis of such preliminary survey as described here.

References

1) Okimura, T.: Structure of the soil horizons and landslides. in Tanaka7), 1977, pp. 27-32 (in Japanese).

2) Hatano, S.: Slope morphology as a factor of landslides. in Tanaka?), 1977, pp. 23-27 (in Japanese).

3) Milligan, V.: Field measurement of permeability in soil and rock. In Situ Measurement of Soil Properties, Vol. H, an ASCE Publication, 1976. pp. 3-54.

4) Okunishi, K.: In situ measurement of permeability of different horizons on the mountain slopes. Correspondence at the First Symposium of the IGU Commission on Field Experiments

in Geomorphology, unpublished 1978. 5) Nakagawa, A.: A study on the ground of mountainous region (I). Ann. Disast. Prey. Res.

Inst., Kyoto Univ., No. 14A, 1971, pp. 589-597 (in Japanese with English abstract). 6) Nakagawa, A.: A study on the ground of mountainous region (II). Ann. Disast. Prey. Res.

Inst., Kyoto Univ., No. 15B, 1972, pp. 183-196 (in Japanese with English abstract). 7) Tanaka, S. (ed.): Study on the Relationship between Landslides and the Geological and

Geomorphological Structures. Report of the Special Research of Natural Disaster Sciences, A-51-4, 1977, 135p. (in Japanese).

8) Huzita, K., T. Kasama, M. Hirano. T. Shinoda and M. Tanaka-Yamashita: Geology and geomorphology of the Rokko area, Kinki District, Japan. Jour. Geosci., Osaka City Univ.,

Vol. 14, Art. 4, 1971, pp. 71-124. 9) Nakanishi, S., Y. Fukuo, S. Tanaka, K. Okunishi, A. Nakagawa, T. Okimura and S. Hatano:

Investigation of landslides in Aotani Experimental Basin. in Tanaka7), 1977, pp. 53-79 (in Japanese).

10) Okunishi, K.: Field experiments in Aotani Experimental Basin. Excursion Guide-book for the Third Symposium of IGU Commission on Field Experiments in Geomorphology, 1980,

pp. 37-42. II) Iida, T. and K. Okunishi: On the slope development caused by the surface landslides. Geog-

raphical Review of Japan, Vol. 52, No. 8, 1979, pp. 426-438 (in Japanese with English Abstract). 12) Fukuo, Y.: Deformation of weathered soil as a factor of landslides. in Tanaka7), 1977, pp.

37-41 (in Japanese). 13) Okuda, S., H. Suwa, M. Nakano and K. Yokoyama: Synthetic observation on debris flow,

Part 2. Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 20B-1, 1976, pp. 237-263 (in Japanese with English abstract).

14) Uemura, M.: Water and sediment discharge from a small mountainous drainage basin. Proc. Ann. Meeting, Assoc. Jap. Geogr., No. 12, 1979, pp. 40-41 (in Japanese).

15) Huzita, K.: Tectonic development of Southwest Japan in the Quaternary Period. Jour. Geosci., Osaka City Univ., Vol. 12, Art. 5, 1969, pp. 53-70.

16) Okunishi, K.: Characteristic erosional processes in granitic drainage basins as found in


Tanakami mountain range, Shiga Prefecture, Japan. Bull. Disast. Prey. Res. Inst., Kyoto Univ., Vol. 24, Part 4, 1974, pp. 233-261.

17) Esumi, S. and K. Okunishi: Hydrological studies on small mountainous drainage basins (V). Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 18B, 1975, pp. 283-292 (in Japanese with English abstract).

18) Matsuda H. and K. Okunishi: Geomorphological characteristics of bare-lands in Tanakami mountain range. Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 13A, 1970, pp. 541-555

(in Japanese with English abstract). 19) Suzuki, T., M. Hirano, K. Takahashi and E. Yatsu: The interaction between the weathering

processes of granites and the evolution of landforms in Rokko Mountains, Japan, Part 1. Bull. Fac. Sci. and Eng., Chuo Univ., Vol. 20, 1977, pp. 343-389 (in Japanese with English abstract).

20) Okunishi, K. and T. Natsukawa: Hydrological studies on small mountainous drainage basins

(I). Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 11B, 1968, pp. 89-101 (in Japanese with English abstract).

21) Chiba, T.: Culture in Bare Mountain Areas. Gakuseisha, Publ. Co., 1973, 233p. (in Japanese). 22) Okunishi, K.: Hydrological studies on small mountainous drainage basins (II). Ann. Disast.

Prey. Res. Inst., Kyoto Univ., No. 13A, 1970, pp. 587-599 (in Japanese with English abstract). 23) Okunishi, K.: Hydrologic change due to afforestation in waste bare-land. Publication of

IAHS, No. 117, 1975, pp. 465-473. 24) Nakai, Y.: Compositional variations of the Inagawa granitic rocks in the Asuke area, Aichi

Prefecture, Central Japan. Jour. Jap. Assoc. Min. Petr. Econ. Geol., Vol. 69, No. 6, 1974,

pp. 215-224. 25) Okuda, S., S. Nakanishi, S. Tanaka, K. Okunishi, A. Nakagawa, S. Hatano, N. Oyagi and

K. Yokoyama: Investigation of landslides in Obara Village and its neighborhood, Aichi Prefecture. in Tanaka?), 1977, pp. 102-126 (in Japanese).

26) Tanioka, S., S. Hatano and S. Niwa : Topographic changes in landslide scars. Proc. Ann. Meeting, Assoc. Jap. Geogr., No. 16, 1979, pp. 76-77 (in Japanese).

27) Okunishi, K. and T. Iida: Study on the landslides around Obara Village, Aichi Prefecture (I). Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 218-1, 1978, pp. 297-311 (in Japanese with

English abstract). 28) lida, T., K. Okunishi and A. Nakagawa: Study on the landslides around Obara Village, Aichi

Prefecture (II). Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 23B-1, 1980, pp. 395-403 (in Japanese with English abstract).

29) Okuda, S., H. Suwa, K. Okunishi, K. Yokoyama and M. Nakano: Observation on the motion of a debris flow and its geomorphological effects. Zeitschrift fur Geomorphologic,

Suppl. Bd., No. 35, 1980, pp. 142-163. 30) Okuda, S., H. Suwa, K. Okunishi, K. Yokoyama, M. Nakano, K. Ogawa and S. Hamana:

Synthetic observation on debris flow, Part 4. Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 21B-1, 1978, pp. 277-296 (in Japanese with English abstract).

31) Okuda, S., H. Suwa, K. Okunishi, K. Yokoyama, K. Ogawa and S. Hamana: Synthetic observation on debris flow, Part 5. Ann. Disast. Prey. Res. Inst., Kyoto Univ., No. 228-1,

1979, pp. 157-204 (in Japanese with English abstract). 32) Okunishi, K.: Hydrology of the Takatoki river. Report of the Special Research of

Environmental Science, B57-R12-4, 1980, pp. 17-20 (in Japanese). 33) Okunishi, K.: Hydrology of the Takatoki river (continued). Report of the Special Research

of Environmental Science, B105-R12-12, 1981, pp. 74-81 (in Japanese). 34) Mizuyama, T.: Geomorphology and climatology of the upper reaches of the Takatoki river.

Report of Academic Investigation of the Upper Reaches of the Takatoki River, 1973, pp. 1-22 (in Japanese).

Documents

INVESTIGATION OF TOPSOIL HORIZONS OF MOUNTAIN SLOPES