A Study on Surface Wash and Runoff Using Open System ... PAPERS/JTAS Vol. 23 (1) Apr. 2000/06... · This study on surface wash and runoff using open system erosion plots was carried

Pertanika J. Trop. Agric. Sci. 23(1): 43 - 53 (2000) ISSN: 1511-3701© Universiti Putra Malaysia Press

A Study on Surface Wash and Runoff Using Open System Erosion Plots

SHARIFAH MASTURA S. A., and SABRY AL-TOUM1

Department of GeographyUniversiti Kebangsaan Malaysia

43600 Bangi, Selangordepartment of GeographyIslamic University of Gaza

Palestine.

Keywords: Surface wash, runoff, open system erosion plots

ABSTRAK

Kajian tentang hakisan dan aliran permukaan ini yang menggunakan plot hakisan sistem terbuka dijalankandi hutan simpanan Tekala, Hulu Langat, Selangor. Tujuannya ialah untuk mengukur satu proses geomorfologiyang penting tentang hakisan dan aliran permukaan. Kadar variasi bagi kedua-dua hakisan dan aliranpermukaan dianalisis dengan merujuk kepada sifat tanih kawasan kajian dan 14 parameter hujan. Hasilkajian menunjukkan bahawa kadar hakisan berada di antara 49.5 g m~2 yr1 ke 137 g mr2 yr~} dengan puratasebanyak 85.04 g mr2 yr1. Sedimen ampaian merupakan lebih 80% daripada jumlah sedimen yang diangkutoleh hakisan permukaan. Kadar purata tentang jumlah aliran permukaan ialah 133.8 lm~2 yt1. Hasil kajianjuga menunjukkan bahawa korelasi antara hakisan permukaan, aliran permukaan dan sifat tanih adalahberbeza-beza. Bagaimanapun, korelasi antara aliran permukaan dengan hakisan permukaan didapati sangatsignifikan dengan kebanyakan parameter hujan. fumlah keseluruhan hujan adalah parameter hujan yangterbaik untuk meramal kedua-dua hakisan dan aliran permukaan.

ABSTRACT

This study on surface wash and runoff using open system erosion plots was carried out in Tekala forest reserve,Hulu Langat, Selangor. Variations in the rates of surface wash and runoff were analysed with reference to soilcharacteristics of the site and 14 rainfall parameters. The results showed that the rate of surface wash ranged from49.5 g mr2 yr1 to 137 g mr2 yr1 with an average of 85.04 g m2 yr~l. Suspended sediment constitutedapproximately 80% of the total sediment transported by surface wash. The average rate of total surface runoffwas 133.8 Imr2 yt1. The results also showed that correlations between surface wash, surface runoff and soilcharacteristics varied. However, the correlation between surface runoff and surface wash was found to be highlysignificant with most rainfall parameters. The total amount of rainfall was most suitable rainfall parameter topredict both surface wash and runoff.

INTRODUCTION

Soil erosion resulting from deforestation andagricultural practice is a serious problem inMalaysia and the tropics in general. A recentstudy in the Sungai Tekala catchment, east ofSelangor, indicated that soil loss under forestson moderate to steep hills can range from 0.5 tha"1 yr"1 to 1.38 t ha'1 yr"1 (Sabry 1997) whileHatch (1981), in the Semangkok Forest Reserve,reported soil loss of 0.15 t ha"1 yr"1. The erosionis caused primarily by the process of rainsplasherosion and surface wash. Under tropical cli-

mate, the processes of erosion are accelerateddue to the fast pace of land use change such asthe replacement of natural forest by agriculturalland or from agricultural land to urban develop-ment. Soil loss estimates during land use conver-sion from forest to oil palm range from 220-250t krrr2 yr1 while that in construction sites wasreported to be phenomenal, from 40,000 - 50,000t ha"1 yr-1 (JAS 1996).

The low rate of soil loss in forest area isattributed to the function of the forest itself.Forest protects the underlying soil from direct

SHARIFAH MASTURA S. A., and SABRY AL-TOUM]

effect of rainfall. The rain forest vegetation in-tercepts between 21.8 % and 26.0 % of annualrainfall (Nik Mohamad et.al 1979). Runoff istherefore greatly reduced. Tree roots bind soil,and under undisturbed conditions, provide rela-tively high infiltrability rate. The litter layer inthe forest also protects the ground fromrainsplash erosion. This paper reports on a soilerosion study conducted in the Sg.Tekala forestcatchment using an open system erosion plot.

METHODOLOGYStudy Area

The 9.8 km2 Sungai Tekala catchment is locatedin the Hulu Langat, Selangor, at 3° 31 12" and 3°51 34" N and 101° 50' 18" and 101° 52' 32" W(Fig. 1). Situated about 40 km east of KualaLumpur, Sg. Tekala is a tributary of Sg. Langat,which drains the western flank of the MainRange.

ration of collected water. A divisor was fixed atthe highest position so as to channel the over-flow discharge to another lower collection bin asshown in Figure 2.

Fig. L Location of study area

Wash Trap Design

The wash trap used in this study was similar tothat used by Gerlach (1967) but with somemodification. The traps, made from sheets ofzinc tin has a collection tank which measures100 cm x 40 cm x 50 cm. The trap lip measured100 cm x 25 cm and the cover was 100 cm x 60cm (Fig.2). A cover was used to prevent directrainfall entering the trap and subsequent evapo-

Trap lip

~40 an !

Fig. 2. Wash trap design

Location of Trap Sites

Four wash traps were installed on each of thethree slope profiles, A, B, and C selected forstudy. Four wash traps were located on eachprofile. At profile A, wash traps were labeled asAl, A2, A3, and A4 and were installed on fourslope segments of 8°, 14.4°, 23.4° and 40.9° re-spectively (Fig.3). At profile B, four wash trapsnamely Bl, B2, B3 and B4 were also installed onslope segments of 70°, 16°, 24° and 39.7° respec-tively. Profile C had four wash traps namely Cl,C2, C3 and C4, located at 6.5°, 17.4°, 20° and37.2° respectively. The wash traps were locatedat convex points of the slope except for washtrap C3, which was located at concave points,while two other plots (Al and A3) were locatedon an exposed ground area without any vegeta-tive cover except for litter.

Sampling Procedure

Surface wash samples were collected from thetraps after individual rainfall events over a yearfrom 7th August 1994 to 17th August 1995. Ninety-nine samples were collected during the studyperiod. Prior to sample collection, the water andsediment in the collecting tank were mixed so

44 PERTANIKAJ. TROP. AGRIC. SCI. VOL. 23 NO. 1, 2000

A STUDY ON SURFACE WASH AND RUNOFF USING OPEN SYSTEM EROSION PLOTS

£ - ^Profik 13

S4'' , , 0 22* - -

2 4 " - -

a> -

< - « Pr.fi.eC

xari

Fig. 3. Location of the open system plotsat Tekala Catchment

that the samples, then collected using a one litrepolyethylene bottle, represented the surface washin the tank. The samples were then taken to the

laboratory for analysis of suspended and dis-solved sediment and concentration.

Rainfall Data Collection and IndicesTwo rainfall stations were established in thecatchment. A 16 cm rim diameter automaticreading Wilh. Lambrecht gauge was placed inthe middle of profile A. A storage gauge with a12.7 cm diameter was located nearby. The sec-ond station was sited in the Forest Departmentat Sg. Tekala catchment using one automaticrain gauge recorder. Both the stations werechecked daily after each event during the firstyear. The first station recorded rainfall datafrom August 17, 1994 to August 17, 1995 whilethe second station recorded rainfall data fromSeptember to November 1996. Fourteen rain-fall and erosivity indices used were based on therainfall data collected at 15 minutes interval(Table 1).

TABLE 1Rainfall and erosivity indices

Symbol Description

Rainfall Indices

AMMI

TKE

30

Erosivity indices

The amount of rainfall for each event in mmThe mean intensity of each event. AM/duration (mm h1)The kinetic energy (joules m'2 mm). Calculation on 15 min interval from KE = 29.8 -127.5/1; I is rainfall intensity.The total kinetic energy for each storm which was used to determine rainfall erosivity forall events together (J nr2)Rainfall intensity index for 15 minutes.Rainfall intensity index for 30 minutes.Rainfall intensity index for 45 minutes.Rainfall intensity index for 60 minutes.

I | 5 TKE the m a x i m u m sustained intensity for 15 minutes(Jeje 8c Agu 1990)

IM TKE the m a x i m u m sustained intensity for 30 minutes(Wischmeier & Smith 1958)

I45 TKE the maximum sustained intensity for 45 minuteslw TKE the maximum sustained intensity for 60 minutesEvd Daily erosivity - 16.64 Rd-173.82 where Rd is the daily rainfall

(Morgan 1974)API Antecedent precipitation index. API = pt. 1/t or pt.kt

Where pt is precipitation for a given day; t is time (number of days hours) since lastrainfall; k is recession factor that is less than one but ranges from 0.85 to 0.98 (Gregoryand Walling 1973)

PERTANIKA J. TROP. AGRIC SCI. VOL. 23 NO. 1, 2000 45


TABLE 2Physical and chemical soil properties of study plots

Characteristic ofthe Plot Site

Bulk densityPore space %Clay %Silt %Fine SandCoarse sand %Slope DegreeOrganic matter

Sites on

Al

L.1257.4230.2

4.4

11.254.2

82.71

A2

0.7870.3346.0

5.0

10.039.014.47.06

profile

A3

0.9663.7343.0

6.09.0

42.023.4

5.45

A

A4

1.0859.2244.9

3.38.8

43.540.9

2.45

Sites on

Bl

1.0261.3333.6

4.4

1L051.0

7.04.09

B2

1.1556.631.5

4.1

12.252.116.0

1.76

profile

B3

0.9663.6736.7

5.611.546.224.0

3.36

B

B4

0.8866.7142.2

5.4

10.941.539.7

6.44

Sites on

Cl

0.7870.2137.5

5.310.746.5

6.5

4.97

C2

0.8267.4527.7

5.213.653.217.42.11

profile

C3

1.2054.5630.5

4.213.551.820

2.8

C

C4

1.1755.6331.4

4.511.552.637.2

3.19

RESULTS

Soil Characteristics and their Relationship withthe Rate of Surface Wash

Bulk Density

The average bulk density in the study area rangedfrom 0.79 to 1.20 (g cm3) with an average of0.99 g cmH (Table 2). Compared to other stud-ies in Malaysia, the values were comparable atbetween 0.94 g cm;) and 0.93 g cm3 for BukitLagong and Bukit Mersawa respectively (Peh1978). About 0.97 g cm3 was reported for theundisturbed soils at Tekam Forest Reserve(Kamaruzaman 1987) while 1.13 g crrr* wasmeasured at the undisturbed forests of the JengkaExperimental Basin in Pahang (Baharuddin etal. 1995).

High bulk density indicates a more mineral-ised, compact and coarse-textured soil and there-fore led to lower infiltration properties. Insig-

TABLE 3Correlation coefficient between sitescharacteristics with both surface wash

and surface runoff

Characteristics ofthe Plot Site

Bulk densityPore space %Clay %Silt %Coarse Sand %Fine Sand %Slope DegreeOrganic matter

Rate ofSurface Wash

0.31-0.290.370.14

-0.32-0.360.92**0.04

Rate ofSurface Runoff

0.20-0.200.100.27-0.170.100.97**0.04

Significant at the 0.001 levelSignificant at the 0.05 level

nificant correlation was found between bulkdensity and both the rate of surface wash andthe value of runoff at the study area (Table 3).

Total Porosity

The average porosity in the study area was 62.2per cent with a range of 54.5 % to 70.3 % at C3and A2 respectively (Table 2). High values ofpore space leads to higher infiltration rates andreduced runoff. This value was found to beclose to the value reported at Bukit Lagong andBukit Mersawa which were 61.6 % and 61.1%respectively (Peh 1978). Porosity in this studywere much higher than that found at JengkaExperimental Basin which was only 29%(Baharuddin et al. 1995). Correlation analysisshowed that the total porosity was significandyand positively correlated with the amount oforganic matter but insignificant and negativelycorrelated when calculated with the rate of sur-face wash and volume of surface runoff (Table3). At the study area, it was expected that therate of infiltration would be very high and henceled to very low surface runoff and may thereforebe potentially less erodible.

Soil Organic Matter

The average organic matter was low (3.8 %)when compared to Bukit Lagong (9.8 %) andBukit Mersawa (10.6 %) (Peh 1978). However,organic content was comparable with soils at theJengka Experimental Basins which was 2.8 %(Baharuddin 1995).

Correlation coefficient showed that a sig-nificant relationship was found between soil or-ganic matter and both bulk density (0.65) andpore space (0.69), but insignificantly correlatedwith both the rate of surface wash and the

46 PERTANIKAJ. TROP. AGRIC. SCI. VOL. 23 NO. 1, 2000


volume of surface runoff. In Alberta, Canada,soil organic matter and soil loss were found tobe negatively correlated with each other (Bryan1974).

This reflects, in general, the importance ofsoil organic matter in developing more poreshence leading to more infiltration and reducingthe volume of surface runoff which could beconsidered as the main geomorphic agent thathas led to surface erosion.

Soil Texture

Soils at the study area were generally classifiedas sandy clay texture. The distribution was bimo-dal with coarse sand and clay being the twopredominant textural classes (Table 4). Thiswas probably due to the removal of fine silt andfine sand by surface wash. Two textural classes,sandy clay and sandy clay loam were identifiedin the study area. The former was more preva-lent at profiles A and B while the latter was moreprominent at profile C.

Correlation analysis on soil texture showedthat clay and silt fractions did not have signifi-cant relationship with the rate of surface washand the amount of surface runoff (Table 3), afinding similarly reported elsewhere (e.g. Bryan1974 and Mutter and Burnham 1990). In con-trast however, Epstein and Grant (1967) re-ported that soil loss increased in soils with highclay content. But in another study, it has beenfound that soil loss increased with an increase inthe clay fraction up to its maximum at 19 %.Insignificant and negative correlations werefound between coarse sand fraction with both

the rate of surface wash and the volume ofsurface runoff (Table 3). In Africa, it was re-ported that there was a negative correlationbetween sandy soil and surface runoff and soilloss (Obi et al. 1989).

Particles Size Distribution of Suspended Sediment

The eroded soil contained a higher proportionof silt and fine sand compared to the insitu soil.Presumably the fine fraction (fine sand and silt)was more susceptible to erosion and transporta-tion by surface wash than coarse sand or clay(Table 4). The findings were similar to thatfound in Hulu Langat, Malaysia (SharifahMastura, 1988) and Nigeria (Lai, 1988). It mayalso be attributable to the clay fraction whichwas found to be less erodable than sand if onlythe factor of rainfall was taken into considera-tion (Le Roux and Roos 1988).

Surface Wash and Influence of Rainfall

The Rate of Surface Wash

Surface wash ranged from 20.3 g m2 yr1 for plotCl to 137.7 g m2 yr1 for plot A 4 (Table 5). Theaverage rate was 85.04 g nrr2 yr1 (0.85t ha-1 yr-1).There was considerable variations in the extentof surface wash due largely to differences inslope angles and soil characteristics.

In all plots, surface wash increased directlywith slope angles. The plots A 4, B 4 and C 4with slope angles of 37° - 40° showed rate ofsurface wash between 130-137 gm2 yr2. Surfacewash was highest in March 1995 and secondhighest in October 1994. Lower rates occurred

TABLE 4Suspended sediment texture of plot sites at Sg. Tekala (OSP)

Plot SiteTexture(%)

Clay Silt Fine Sand Coarse Sand

AlA2A3A4BlB2B3B4ClC2C3C4

222929251717162117171818

30.2464344.933.631.536.742.237.527.730.531.4

1111139111112121212910

4.45.06.03.34.44.15.66.45.35.24.24.5

232723303029343332273330

11.2109.08.81112.211.510.910.713.613.511.5

443335364243383439444042

54.239.042.043.551.052.146.241.546.553.251.852.6

Data in italic indicate before erosion

PERTANIKA J. TROP. AGRIC. SCI. VOL. 23 NO. 1, 2000 47


TABLE 5Rate of surface wash and surface runoff at plot sites

SiteSlope

(degree)Rate of Surfacewash (g m2 yr1)

Rate of SurfaceRunoff ( Im2 yr-1)

Rate ofRunoff/Rainfall

(Q/P)

AlA2A3A4BlB2B3B4ClC2C3C4

8°14.4°23.4°40.9°7°

16°24°39.7°

6.5°17.4°

20°37.2°

60.9871.3196.032

137.7456.4677.66

113.31135.67

20.3249.5470.96

130.51

80.17143.97123.7137.585.5

160.4

145.45189.43

69.6130.8145.6187.71

3.366.025.185.763.786.716.097.922.91

85.486.097.86

in February and July 1995. The monthly surfacewash distribution suggest bimodal with maxi-mum peaks occurring in March and Octoberand two minima occurring in February and July.This was similar to the monthly rainfall distribu-tion recorded at the study area.

In Malaysia, similar studies under forest coverland use have been carried out which usederosion plot techniques. Peh (1978) studied atBukit Lagong and Bukit Mersawa and foundthat the rates of sediment loss were 0.30 and0.31 cm3 cm1 yr * respectively. In his study at thePasoh Forest Reserve, Negeri Sembilan, Leigh

(1982) reported that the average suspended sedi-ment transport was 0.29 cm3 cm1 yr1. Otherstudies on the rate of soil loss in forest area isgiven in Table 6.

Correlation between Surface Wash and Rainfall

parameters.

The correlation coefficients were generally sig-nificant and positive for all the rainfall param-eters except for mean intensity (MI), where thecorrelates were low but significant (Table 7).The total amount of rainfall (AM) total kinetic

TABLE 6The rate of soil loss in forest areas

Study Area

Mesilau, KinabaluPark Sabah

Mendolong, Sabah

Semongkok F.R

Semongkok F.R(Primary Forest)

Semongkok F.R(Secondary Forest)

SouthwesternNigeria(Secondary Forest)

Usambara MtsTanzania

Central HimalayaIndia

Period ofStudy

4 months

10 months

5 months

One year

One year

2 years

2 years

8 months

Slope Angle

36-38%

19.6-42%

25°-30'

25°

25°

10%

10VL5*20°-25°

5°-25°

Plot Size

25m x 6m

Open Plot

10m x 4m

10m x 4m

10m x 4m

25m x 4m

12m x 2m12m x 2m

20m x 20m

No. ofPlots

1

7

3

3

3

1

1

6

Soil Loss

0.408 t ha-1 yr1

38 kg ha'1 yr1

0.3553 t ha-1 yr1

0.1491 t ha 1 yr1

0.0573 t h a ' yr1

78.9 kg ha 1 yr1

14.2 g m 2 yr-1

10.1 g rrv2 yr1

15.3-57.2 gha-1 yr1

Author

George (1987)

Malmer (1996)

Hatch (1978)

Hatch (1981)

Hatch (1981)

Jeje (1987)

Lundgren (1980)

Pathaket al (1984)

48 PERTANIKAJ. TROP. AGRIC. SCI. VOL. 23 NO. I, 2000

TABLE 7

2

I5O

P.

ip

IOL

.

NS0>-*4O

. 1

RainfallParameter

AM

MI

TKE

^ 5

^ 0

™so

ho

hs

*m

EVD

API

Al

**0.83

**0.39

**0.74

**0.81

**0.73

**0.72

**0.73

**0.74

**0.68

**0.73

**0.77

**0.78

**0.79

**0.80

Sites on

A2

**0.87

**0.49

**0.74

**0.86

**0.73

**0.74

**0.74

**0.75

**0.70

**0.80

**0.84

**0.86

**0.84

**0.81

Profile A

A3

**0.82

**0.38

**0.78

**0.82

**0.79

**0.78

**0.79

**0,80

**0.64

**0.72

**0.78

**0.79

**0.80

**0.78

Correlation

A4

**0.86

**0.46

**0.85

**0.86

**0.84

**0.85

**0.86

**0.86

**0.69

0.77

**0.84

**0.85

**0.86

**0.83

coefficients between surface

Bl

**0.75

**0.37

**0.65

**0.73

**0.64

**0.63

**0.60

**0.60

**0.66

**0.69

**0.68

**0.68

**0.69

**0.69

Sites on

B2

**0.80

**0.41

**0.68

**0.78

**0.67

**0.67

**0.63

**0.64

**0.64

••O72

••074

**0.74

**0.76

**0.65

Profile B

B3

**0.82

**0.40

**0.76

**0.80

**0.75

**0.74

**0.75

**0.76

**0.72

**0.74

**0.78

**0.78

**0.78

**0.80

wash and

B4

**0.79

**0.35

**0.74

**0.77

**0.72

**0.72

**0.72

**0.72

**0.68

**0.72

**0.75

**0.74

**0.76

**0.80

rainfall parameters

Cl

**0.88

**0.51

**0.77

**0.85

**0.76

**0.75

**0.74

**0.74

**0.73

**0.81

**0.82

**0.81

**0.82

**0.69

Sites on

C2

**0.86

**0.43

**0.75

**0.83

**0.74

**0.73

**0.73

**o.74

**0.70

**0.76

**0.79

**0.80

**0.82

**0.82

Profile C

C3

**0.85

**0.47

**0.66

**0.83

**0.65

**0.64

**0.62

**0.62

**0.72

**0.78

**0.78

**0.79

**0.81

**0.77

C4

**0.73

**0.50

**0.79

**0.73

**0.77

**0.73

**0.73

**0.73

**0.76

**0.76

**0.73

**0.72

**0.66

**0.72

Average

**0.82

**0.43**074

**0.80

**0.73

**0.72

**0.72

**0.73

**0.69

**0.75

••0.77

**0.78

**0.78

**0.76

n

77

77

77

77

77

56

45

37

77

56

45

37

44

75

>

""V.

C

I<

JOG§

1o01i

** Significant at 0.001 level* Significant at 0.05 level


energy (TKE), storm erosivity (EVd) and maxi-mum intensity for 60 minutes (If>0) were thehighest correlated rainfall parameters of surfacewash. The EI^ index was therefore highly corre-lated with the rate of surface wash as comparedto EI]V EI^ and EI45. The correlation coeffi-cients between the rate of surface wash withmaximum rainfall intensity for 15, 30, 45 and 60minutes were higher than the rainfall parameterthat contained the interaction of energy andintensity (EIjr., EI.W, EI45 and EI^). Other studiesreported positive correlation between one ormore rainfall parameters and the rate of surfacewash for different land use such as forest andagriculture (Lai 1976, Sharifah Mastura 1988,and Armstrong 1990.

Wischmeier and Smith (1958) showed thatthe product of the kinetic energy of the stormand the maximum EI^ was most significantlycorrelated with soil loss. The study in Rhodesiareported that in a bare area within agriculturaluse, the use of EI^ was a better predicator forsoil loss than EI i r In Malaysia, Peh (1978) foundthat EI^ had the highest correlation with surfacewash at Bukit Lagong. Hudson (1981) however,found that total kinetic energy of rainfallintensities which was greater than 25 mm hr1 tobe a more significant predictor of soil loss thanthe EITO index. Mutter and Burnhan (1990) re-ported that kinetic energy was the most success-ful rainfall parameter related to soil erosion. AtPasoh, Malaysia, Peh (1978) found that totalkinetic energy, EI1V EIM EI4r. and EI^ were sig-nificantly correlated with sediment transport butthe correlation coefficients were lower than thatbetween the amount of rainfall and sedimenttransport. At Bukit Mersawa, the same authorfound that the total kinetic energy index andthe maximum intensity for EI r had the highestcorrelation with sediment transport.

Pinczes (1982) and Ruangpainit (1985)found positive correlation between soil loss andboth amount of rainfall and intensity. In SouthAfrica, Garland (1987) found that the best rela-tionship was between sediment transport and 1^.But in America, Wischmeier and Smith (1958)found that the correlation between soil loss andthe amount of rainfall as well as the maximumintensity for 5, 15, or 30 minutes intervals waspoor. This result was supported by Armstrong(1990) who found that the correlation betweenvarious erosivity indices and soil loss was poor.

Highly significant and positive correlationwas found between surface wash and antecedentprecipitation index. Chang and Ting (1986)also reported a significant positive correlationbetween sediment transport and the antecedentprecipitation index while Jeje (1987) found nosignificant correlation between antecedent pre-cipitation index and soil loss in Nigeria.

Overall other results did not show consist-ency in identifying the best rainfall and erosivityindices. This is probably because the experi-ments have been carried out in different cli-matic regions with different techniques appliedto it. This is also an indication of the complexi-ties involved in erosion studies and more re-search is required in the future.

Relationship between Surface Runoff and SurfaceWash

Rate of Surface Runoff

The rate of annual surface runoff for all erosionplots at the study area is shown in Table 5. Theaverage rate of total surface runoff at the Sg.Telaka was 133.8 lm2, with a range from 69.6 atplot CI to 189.4 lm* at plot B4. The low surfacerunoff at the study area was probably due to thevegetative cover that intercepted a certainamount of total rainfall that subsequently evapo-rated to the atmosphere. It was found that inPeninsular Malaysia the amount of rainfall inter-cepted ranged from 20% to 25% (Ffolliott 1990)and 27% according to Nik Muhammad andShaharuddin (1979). The presence of canopyprovide the soil with more time to receive therainfall and thus allowing time for more infiltra-tion before surface runoff starts. Soil character-istics, such as low bulk density, high porosity andhigh percentage of sand, also led to a higherrate of infiltration, hence reducing the amountof surface runoff.

In order to compare with other studies, therate of surface runoff was converted to rainfall(P)/ runoff (Q) coefficient or Q/P. The averagepercentage of Q/P for all plots at Sg. Tekala was5.6 %, with a range of 2.91 to 7.92 %. In otherareas, the Q/P coefficient is given in Table 8.The results varied from 2.3% at Jengka Pahangto as high as 6.8 % in Thailand.

As discussed above different percentages ofQ/P have been reported for different rain forestareas. This showed that the tropical forest wasfar from being uniform, as there existed differ-

50 PERTANIKAJ. TROP. AGRIC. SCL VOL. 23 NO. 1, 2000


The value

Study Area

Jengka,Pahang

Sabah

Sabah

Nigeria

Thailand

TABLE 8of rainfall/Runoff coefficients

P / QCoefficients

2.3

2.9

4

3.2

6.8

Author

Baharuddinet al (1995)

Malmer (1996)

George (1987)

Jeje and Agu(1990)

Thongmee andVannapraset (1990)

ences in canopy structure, in the composition ofthe ground layer and in the rate of litter fall.Differences in the experimental sites and differ-ences in plot sizes used have also caused differ-ent results. In addition, other climatic factorsalso contributed to the differences in the Q/Pcoefficient, such as the total amount of rainfalland its annual pattern of distribution.Analysis of the monthly surface runoff was bimo-dal with two maximum results registered inMarch and October. The highest maximum valueoccurred in March. The two minima occurredin February and July. Overall the rate of surfacerunoff distribution was highly linked to the rain-fall distribution of the study area.

Correlation between Surface Runoff andSurface Wash

A significant and positive correlation was foundbetween the rate of surface wash (WS) andsurface runoff (VW) at all the 12 erosion plotsin the study area as shown in Table 9. Anotherstudy showed similar results (Mutter andBurnham, 1990). The strongest correlation wasregistered at plots C 3 anf C 4 with r = 0.96 andhighly significant at 0.001 level. A lower corre-lation was registered at plot A 3 with r = 0.68 but

the correlation was also highly significant at0.001 level.

CONCLUSIONSoil loss from the Sg. Tekala catchment is low.Changes to its forest cover would result in con-siderable increase in erosion rates. The relation-ships between both surface runoff and surfacewash with soil characteristics and 14 selectedrainfall parameters were analysed. Out of thefourteen used rainfall amount, total kinetic en-ergy, h}jstorm erosivity and maximum intensity60 minutes indices were found to be highlycorrelated with the rate of surface wash. Theseindices, therefore, can be recommended as pre-dictor indices for erosion studies in tropicalregion. Comparisons of results made with previ-ous studies carried out in Malaysia and else-where showed that the results differed in somebecause of differences in land cover, plot sizeand methods adopted.

The relationship between surface wash andrunoff was significantly correlated, hence theclose link between the rate of surface runoff andrainfall distribution in the study area. Althoughthe relationships between surface wash, surfacerunoff with soil characteristics were not clearexcept for slope angle, further research in thisarea is suggested to establish a better under-standing between these parameters.

REFERENCES

ARMSTRONG, J. L. 1990. Runoff and soil loss frombare fallow plots at Inverell, New SouthWales. Aust. J. Soil Res. 28 : 659-675.

BAHARUDDIN, K. MOKHTARUDDIN, A.M. and M. NIK

MUHAMAD, 1995. Surface runoff and soil lossfrom a skid and a logging road in a tropicalforest. / Trop. For. Sri. 7(4) : 558-569.

BRYAN, R.B. 1974. Water erosion by splash andwash and the erodibility of Albertan soils.Geogr. Ann. 56A (3A) : 339-346.

TABLE 9Correlation coefficient between rate of surface wash and surface runoff.

Study Plot

Correlationcoefficient

Al

**0.92

A2

**0.93

A3

**0.68

A4

**0.79

Bl

**0.89

B2

**0.83

B3

**0.76

B4

**0.81

Cl

**0.95

C2

**0.92

C3

**0.96

C4

**0.78

Al!

**0.75

**Significant at 0.001 level



CHANG, M. and J.C TING, 1986. Applications ofthe Universal Soil Loss Equation to variousforest condition. IAHS Publ. No. 157 : 165-174.

EPSTEIN, E. and J. GRANT. 1967. Soil losses andcrust formation as relayed to some soil physi-cal properties. Soil Sci. Soc. Am. Proc.61 : 547-30

FFOLLIOTT, P.F. 1990. Manual on Watershed Instru-mentation and Measurement. Asean-US Water-shed Project. College Langua Philippines.

GARLAND, G.G. 1987. Rates of soil loss from moun-tain footpaths : An experimental study inthe Drakensburg mountain, South Africa.Applied Geography. 7 : 41-54.

GEORGE D.A. 1987. Measurement of rainfall andsoil erosion from agricultural and primaryjungle plots in Sabah, Malaysia. In SteeplandAgriculture in the Humid Tropics, p. 309-331.eds. T.H.Tay, Mokhtaruddin, AM. andZahari, AB. Kuala Lumpur.

GERLACH, T. 1967. Hillslope troughs for measur-ing sediment movement. Rev. Geomorph. Dyn.Xvii (4): 173-174.

HATCH, T. 1978. Preliminary results from soilerosion and conservation trials at SemangkokA. R. C. Tech. Paper No. 6, Dept. Of Agricul-ture Sarawak.

HATCH, T. 1981. Soil erosion and shifting cultiva-tion in Sarawak. In Workshop on HydrologicalImpact on Forestry Practices and Re-afforestationUPM. No. 2-6, 1981. UPM, Malaysia.

HUDSON, N. W. 198L Instrumentation for studiesof erosive power of rainfall. IAHS Publ. No.133 : 383-390.

Jabatan Alam Sekitar, Malaysia. 1996. Guidelinesfor Prevention and Control of Soil Erosion andSiltation in Malaysia. JAS Kuala Lumpur.

JEJE, L.K. 1987. Runoff and soil loss from erosionplots in Ife area of South Western Nigeria.In International Geomorphology 1986. Part II.ed. V. Gardiner. Chichhester : John Wiley 8cSons Ltd : 459-472.

JEJE, L.K and A.N. AGU. 1990. Runoff frombounded plots in Alakowe in SouthwesternNigeria. Applied Geography 10 : 63-74.

KAMARUZAMAN, J. 1987. Impact of ground-basedlogging machine on soil physical propertiesand tree growth. Unpublished M.S. ThesisUPM Serdang.

LAL, R. 1976. Soil erosion on Alfisols in WesternNigeria, I. Effects of slope, crop rotationand residue management. Geoderma 16 : 363-375.

LAL, R. 1988. Erodibility and erosivity. In SoilErosion Research Methods, p. 141-160, ed. Lal,R. Ankeny. USA. Soil & Water Cons. Soc.

LEIGH, C.H. 1982. Sediment transport by surfacewash and throughflow at the Pasoh ForestReserve, Negeri Sembilan, Malaysia. Geogr.Ann. 64A (3-4) : 171-180.

LE ROUX, J.S. and Z.N Roos. 1986. The relation-ship between the size of particles in surfacewash sediment and rainfall characteristicson a low angle slope in a semi-arid climateZ. Geomorph. N. F. 30(3) : 357-362.

LEVY, G.J., LEVIN, J. and I SHAINBERG. 1994. Sealformation and interrill soil erosion. Soil Sci.Soc. Am. J. 58 : 203-209.

LUNDGREN, L. 1980. Comparison of surface run-off and soil loss from runoff plots in forestand smallscale agriculture in the UsambaraMts., Tanzania. Geogr. Ann. 62A (3^) ; 113-148.

MALMER, A. 1996. Hydrological effects and nutri-ent losses of forest plantation establishmenton tropical rainforest land in Sabah, Malay-sia. The Malaysian Forester 42(3) : 174-201.

MUTTER, G.M. and C.P Burnham. 1990. Plot-studies comparing water erosion on chalkyand non-calcareous soils. In Soil erosion onagricultural land, p. 15-24. ed. J. Boardman,IDL. Foster, &JA Deary, Chichester : JohnWiley 8c Sons Ltd.

NIK MUHAMAD, M. and A. SHAHARUDIN. 1979. Rain-fall interception, throughfall and streamflowin a secondary forest. Pertanika. 2 (2) : 152-154.

OBI, M.E., SALAKO, F.K. and R. LAL. 1989. Rela-tive susceptibility of some Southeastern Ni-geria soils to erosion. Catena. 16 : 215-225.

PATHAK, P.C., PANDEY, A.N. and J.S SINGH. 1984.

52 PERTANIKA |. TROP. AGRIC. SCI. VOL. 23 NO. 1, 2000


Overland flow, sediment output and nutri-ent loss from certain forested sites in theCentral Himalaya, India./. HydroL 71 : 239-251.

PEH, C.H. 1978. Rate of sediment transport bysurface wash in three forested area of Pe-ninsular Malaysia. Occasional Paper No. 3.Department of Geography. UM. KualaLumpur.

PINCZES, Z. 1982. Variation in runoff and erosionunder various methods of protection. IAHSPubl. No. 137 : 49-57.

RUANGPAINIT, N. 1985. Percent crown cover re-lated to water and soil losses in mountousforest in Thaoiland. In Soil Erosion and Con-servation, p. 462-471. ed. El-Swaify, SA.Moldenhauer, WC. And Lo, A. Ankeny.Lowa: SoilCon. Soc. Am.

SHARIFAH MASTURA, S.A. 1988. Soil loss from bareslope using simple erosion plots. In TropicalUrban Ecosystem Studies. Volume 5. Contempo-rary Urban Issues in Malaysia, p. 38-45. ed.Abdul Samad., Mazlan, O. and H. Khatijah.Working Group on Urban Ecosystem.Malaysian National MAB Committee.

SABRY AL TOUM. 1997. Surface erosion study inthe Granite Area of Hulu Langat, Selangor,Malaysia. Ph.D Thesis. Depart, of Geogra-phy. UKM. Unpublished.

SINGER, MJ and P.H WALKER, 1983. Rainfall-runoff in soil erosion with simulated rain-fall, overland flow and cover. Aust. J. SoilRes. 21 : 109-122.

THONGMEE, U and M. VANNAPRASERT. 1990. Soiland Water losses on plots with differentland use in the Phu Wiang Watershed. IAHS-AISH Publ No. 192 : 48-52.

WISCHMEIER, W.H., and D.D SMITH. 1958. Rainfallenergyand its relationship to soil loss. Trans.Am.Geophy.Union. 39 : 258-291.

Received: 13 July 2000Accepted: 18 October 2000


Documents

A Study on Surface Wash and Runoff Using Open System ... PAPERS/JTAS Vol. 23 (1) Apr. 2000/06... · This study on surface wash and runoff using open system erosion plots was carried