Water Quality Profile of Sg. Langatpsasir.upm.edu.my/...Quality_Profile_of_Sg._Langat.pdf · Langat...
Transcript of Water Quality Profile of Sg. Langatpsasir.upm.edu.my/...Quality_Profile_of_Sg._Langat.pdf · Langat...
Pertanika 11(2), 273-281 (1988)
Water Quality Profile of Sg. Langat
ANHAR SUKI, MORD KAMIL YUSOFF and MOK TREAN POEDepartment ofEnvironmental Sciences
Faculty ofScience and Environmental StudiesUniversiti Pertanian Malaysia
43400 , Serdang, Seiangor, Malaysia.
Key words: Water quality; Sg. Langat
ABSTRAK
Kajian ini telah dijalankan pada bulan Ogos hingga Disember 1985 yang bertujuan untuk menilaikesan-kesan pencemaran yang berpunca daripada corak penggunaan tanah yang berlainan atas kualiti airsungai. Sebanyak sembilan lokasi penyempelan telah dipilih sepanjang 60 km bermula dari hulu sungai.Beberapa parameter fizikal dan kimia telah diukur. Ujian statistik dan Indeks Kualiti Air diguna untukmenentukan perubahan kualiti air. Keputusan kajian menunjukkan nilai-nilai BOD, pepejal terampai,pengkonduksian elektrik dan kekeruhan meningkat selepas bandar Hulu Langat. Ujian statistik menunjukkan perubahan kualiti air sangat bererti (p < 0.05). Berdasarkan Indeks Kualiti Air, kualiti air sungai yangberhampiran bandar Kajang menurun dengan ketara jika dibandingkan dengan lokasi yang lain.
ABSTRACT
This study was conducted between August and December 1985 in order to assess polluting effectsfrom different landuse patterns. Nine sampling sites were chosen over a distance of 60 km. A numberof physical and chemical parameters were analysed. Statistical tests and Water Quality Index were used toevaluate the water quality variation. The result shows that relatively high BOD, suspended solids, electricalconductivity and turbidity occurred after Hulu Langat town. The statistical test indicated that the waterquality variation is significantly different (p < 0.05); the Water Quality Index shows that the waterquality near Kajang town has deteriorated compared to other stations.
INTRODUCTIONWater serves a multitude of uses. As the varioususes have grown in diversity, constrained by themore or less fIXed quantity available, significantinteraction of uses and reuses of water has alsodeveloped. Water quality is a very broad termwhen used by some people and quite specificfor others and is the complementary aspect ofwater quantity (Biggar, 1979).
Chemically pure water rarely occurs innature. Its content varies largely from region toregion, and is a reflection of the local geographyand climate (Hynes, 1970). According to Tanji
Key to authors' name: A. Suki; M. Kemil; T.P. Mok
(l979), water is often referred to as a universalsolvent and because of this solvent power, it tendsto pick up impurities as it comes into contactwith gases, liquids and solids. Water in the hydrologic cycle is highly mobile and acts as a conveyorwhich then transports the pollutant downstream(Tanji, 1979). The solvent power and the mobilityof water are the basic reasons why we have waterpollution problems in rivers or any water bodies.
In a developing country like Malaysia, rapidchanges are continually taking place. These includepopulation growth, urbanisation, agricultural,mining and logging activities, and industrialisation.
ANHAR SUKI, MOHO KAMIL YUSOFF AND MOK THEAN POE
Within the study area, there are six townslocated along the river; Kg Kuala Panson, DusunTua, Hulu Langat, Cheras, Kajang and Dengkil(Figure 2). The largest town is Kajang, located' inthe middle section of the catchment area.
The landuse pattern is depicted in Figure 1.The basin comprises urban and associated areas,cultivation of rubber, oil palm, mixed horticulture and coconut, tin mining and forest reserve.Percentage wise, agricultural land and forestoccupy 52 and 40 percent of the total acreagerespectively (Mok, 1986).
Most of the industrial areas are located inKajang and comprise various type of factories;the major factories being rubber, textile andpaper mills (Figure 2).
MATERIALS AND METHODS
9.9
31.0
31.0
18.0
17.5
16.7
17.3
14.7
11.9
River width(m)
On-site surveys were carried out determine thepotential sampling sites. The siting of the samplingsite was based on accesibility and homogeneity inorder to obtain a representative sample. Based onthe above criteria, nine sampling sites were chosen(Figure 2). Table 1 shows the distance of thesampling sites from the estuary (at the rivermouth).
Sampling of water from each site was carriedout at a frequency of once a week from Augustto December 1985 for a duration of 12 weeks.For the purpose of this study, grab samples werecollected for in-situ and laboratory analysis. Inorder to get a representative sample, the watersample was taken in the middle section of theriver at 0.6 depth from the water surface (i.e60% of the total depth).
The selected parameters fqr measurementwere pH, temperature, dissolved oxygen (DO),electrical conductivity (EC), turbidity, suspendedsolids and biochemical oxygen demand (BOD).The first three parameters were determined onsite and the rest of the analysis were performedin the laboratory within 24 hours. Nutrientmeasurements were not carried out. Table 2 showsthe analytical methods adopted throughout thestudy.
The analysis of variance (one way ANOYA)and Duncan Multiple Range tests were used todetect the significant difference of means betweensampling sites (Steel and Torrie, 1960). The interpretation of variation in water quality was evaluated with the Water Quality Index (Norhayati,1981).
PERTANlKA VOL. 11 NO.2, 1988
TABLE 1A description of the sampling sites or stations.
Station Distance fromEstuary (km)
Sl 118
S2 109
53 105
54 95
55 92
56 87
57 81
58 74
59 60
274
Basin CharacteristicsThe study area was confmed to a catchment
area of aproximately 700 sq. km within the Sg.Langat basin (Figure 1). The area is drained bySg. Langat with its headwaters rising in the hills inthe north and flowing generally southward towards the plains before turning westward towardsthe sea, covering a distance of approximately 120km. The headwaters of the Sg. Langat is made upof forested foothills and rugged mountain ranges.
These changes bring about complex environmentalproblems and the most important natural resource that is affected is water (Abu Bakar, 1985).The development of land and natural resources,and the discharge of waste products into the waterbody are the main water pollution sources inMalaysia.
In the early days the water quality of Sg.Langat was of little concern as demand for waterwas relatively low. Now, with increased demandfor water by the growing population, industry andagricultural sectors, there is an increased pollutionload in the river, and this affects its uses (Mok,1986). This study was carried out along Sg.Langat, to check on the variation in water qualitywith respect to landuse. Similar studies have alsobeen carried out on other river systems. Law(1980), concentrated more on faecal coliformcounts in Kelang River and Tan and Ng (1980),studied the effect of rubber effluent and domesticwaste discharges into Gombak River and a tributary.
N
WATER QUALITY PROFILE OF SG. LANGAT
101 047 '
KEY
Forest
Rubber
Scrub forest
Oil palmUrban andass areasHorticulture
[ill] Tin minesnewly cleared
0 Lands
till Swamps. -
~ Pasture
• Town
..".c.. Rivers
Scale I em : 2.5 km
Fig 1. LDnduse pattern in the study area.
PERTANIKA VOL. II NO.2, 1988 275
ANHAR SUKI, MOHO KAMIL YUSOFF AND MOK THEAN POE
101 0 47 .
Scale: I em = 2.5 km
N
(g)
•
KEY
Industrial Area. - "-Rubber Factory
Sampling station
·Road
276
Fig 2. Location of sampling sites.
PERTANIKA VOL. II NO.2, 1988
WATER QUALITY PROFILE OF SG. LANGAT
RESULTS AND DISCUSSION
5'
• -DO• - 100,
78 85 92 99 106 113 120Distance from Estuary (km)
Fig 3. BOD and DO values for the various sampling sites.
also contribute to the increase in pollution load(Table 3). Beside these, there are also a numberof other small industries and several saw millsalong the river.
The BOD values were also obtained alongthe river. The statistical analysis shows that themean values between each sampling station werefound to be very significantly different (p <0.01).This is also true for the other parameters discussedbelow. There is a distinct increase in the BODvalue after sampling station S3. (Figure 3) At
15.0
13.5
12.0
10.5
~ 9.0
S8 7.5
~ 6.0Cl
~ 4.5
3.0
1.5
050 57 64 71
BOD and DOThe average DO profJJ.e for the period of study isgiven in Figure 3. As shown, the dissolved oxygenvalue varies from 8.1 to 6.1 mg/I. Statisticalanalysis shows that the mean DO at each samplingstation is significantly different (p < 0.05). Atthe upstream regions (Sampling station Sl-S3),the value is above 7.8 which represents greaterthan 95% saturation. This is expected as the riverin the upper region is mainly shallow and turbulent allowing ample reaeration. There is alsolimited organic pollution load in that section ofthe river as the area is mainly covered by rubberplantation and forest. The BOD value is on averageless than 1.0 mg/I. This compares well with dataobtained by the Department of Environment(l980-l984).
Further downstream the DO value reducesto 6.9 mg/l near Cheras area and to 6.2 mg/l atKajang Town. The latter represents a reductionof the dissolved oxygen content by as much as23% as compared to the upstream region (Sl).The built-up areas from Cheras to Kajang towncontribute to the increase in organic load particularly via the open drains. There is also a rubbermill downstream of Hulu Langat town and a paperand a textile mill at Kajang town. These industries
TABLE 2Analytical methods adopted
Parameter Method of analysis
pH
TemperatureDissolved oxygenTurbidity
Electrical conductivitySuspended solids·BOD·
Orion digital pH meter model 201 with glass electrode (± 0.1pH unit)Thermistor probe (YSI 58) ( ±O.))Dissolved oxygen meter model YSI 58 ( ±O.I mg/!)Turbidimeter (HACH Mode12100A)(±2% of the full scale:-Ranges:- 0 0.1 NTU
o 1.0 NTUo 10 NTUo 100 NTUo 1000NTU)
Digital conductivity meter model PCM 3 ( ± 0.01 umhos/cm)Filtration with Whatman glass microfibre filtersAzide modification of the Winkler method and incubation timeof 5 days at 20 C.
• - - - APHA,1975
PERTANIKA VOL. II NO.2, 1988 277
ANHAR SUKI, MOHO KAMIL YUSOFF AND MOK THEAN POE
station- S4, the BOD had increased 10 fold compared to the value upstream. This is partially dueto developed areas after Hulu Langat town andthe rubber mill. The discharge from the mill andits BOD value is given in Table 3. The BOD valueactually peaks at Kajang town with an averagevalue of 9.0 mg/I. A BOD value of 9 mg/1 wouldbe considered a moderately polluted river according to a classification by Mok (1985). Here theopen drains from the town normally dischargeinto the river and there are also industries likethe paper and textile mills which discharges theirwastes. The 1985 data on these discharges are alsogiven in Table 3.
After Kajang town, the mean BOD valuesdecreases and at sampling station S9 the valuehas dropped to an average of 3.2 mg/I. This isdue to the reduction in urban activity downstream and the absence of other major pollutionsources. The DO as expected did not increaseimmediately and has a minimum value at stationS7. This is due to the time taken for biologicalactivity to reduce the oxygen content of thewater. The Streeter-Phelps equation which calculates oxygen deficit in rivers due to organicpollution also predicts a similar prof11e (Metcalfand Eddy Inc. (I 979)).
It is interesting to note that a BOD of 9.0mg/l at Kajang represents an organic mass transport of about 68 kg/h (assuming lowest streamflow for the period of study (Table 4) and theinput from the two major mills (paper and textile) at Kajang is about 4 kg/h (using data fromTable 3). This rough calculation shows that themajor organic input to Sg. Langat comes fromother non point sources. It is suspected that thedrain serving the developed areas are the maincontributers.
Turbidity and Suspended SolidsThe suspended solids and' turbidity results areshown in Figure 4. The trend obtained withrespect to distance is similar to that obtainedfor BOD and DO. Stations SI, S2 and S3 huelower suspended solids as the area is mainlycovered by forest. As such there is little erosionoccuring. This result is also reflected in the lowturbidity values. The suspended solids value ofless than 26 mg/1 is in the class of good qualitywater which- requires little or no treatment forpotable ust:.
The suspended solids and turbidity areexpected to rise as the river flows past developed and developing areas from Cheras andKajang town. The mean suspended solids valueat station S8 is a high 377 mg/I. This representsmore than a 10 fold increase compared to the
455 100
410 Turbidity 91
Suspended Solid360 82
320 73
~215s.
64:;
"0 ~'g 230 55 ~VI"0 >-
~ 185 • 46~<=It~140 37 ~
95 28
50 19
$.
50 57 64 71 78 85 92 99 120Distance from Estuary (km)
Fig 4. Suspended solids and turbidity values for thedifferent sampling sites.
TABLE 3Characteristics of industrial effluents in the study area
Type of Discharge Average Monthly Concentration
Factory (m3/day) Suspended solid BOD(mg/!) (mg/I)
Rubber 256 34 189
Textile 360 236 55
Papermill 246 20 265
Source: Department of Environment (1985)
278 PERTANlKA VOL. II NO.2, IS88
WATER QUALITY PROFILE OF SG. LANGAT
TABLE 4Weekly river discharge at Kajang (S6) and Dengkil (S9)
Month
August
September
October
November
Source: Mok (1986)
StationKajang (S6)
(cumecs)
3.02.13.73.07.72.72.63.47.54.43.54.3
18.821.4.17.315.8
Dengkil (S9)(cumecs)
10.17.45.54.95.75.76.0
15.536.518.534.235.463.371.863.457.8
value at Sl to S3. Between S6 and S8, developmental projects are underway. These includeland clearing and housing development. As suchthe suspended solids and turbidity continue torise.
At .station S9, the suspended solids decreasein concentration. This last station is about 14 kmfrom S8, and the area is mostly palm oil plantation, forest and swamps. Therefore there islittle additional input of solids and sedimentationwould reduce the settleable component. Two tributaries namely Sg. Chua and Sg. Semenyih jointhe Sg Langat 4 km and 9 km upstream of stationS9. These two factors partially explain the reduction in the suspended solids values.
Electrical Conductivity (EC)The results are given in Figure 5. The electricalconductivity range from 33 to 88 uS/em. This isconsidered low and does not affect the waterquality, as river water has an EC value of between50 to 150 uS/em. The EC trend downstream issimilar to the other parameters as given above.
pH and TemperatureThe mean pH for the various sampling stationsrange from 6.3 to 6.8 (Figure 6). This is a relatively narrow range and well within the pH of naturalwaters. Even though the range is small, the mean
at each station was found to be significantlydifferent (p < 0.01). The temperature of thesampling points are apprOXimately constant,ranging from 26 to 28oC. The variation is mainlydue to the time of sampling rather than any realdifference between the sampling stations. Thesampling was carried out from morning to evening.
Water Quality Index (WQI)The water quality index can be computed usingthe various mathematical equations. For example
100
91
~ 82
~ 73§;64:Eg 55"0
"846-;;
'S 37.l!UJ
28
9\
50 57 64 71 78 85 92 99 106 113 120Distance from Estuary (km)
Fig 5. Electrical conductivity values for the differentsampling rite,.
PERTANIKA VOL. 11 NO.2, 1988 279
ANHAR SUKI, MOHD KAMIL YUSOFF AND MOK THEAN POE
the water quality index can be based on a weighted or unweighted arithmetic mean, weighted orunweighted geometric mean of partially weightedarithmetic or geometric mean (Horton, 1965 and;Ball and Church, 1980). The easiest to use isprobably the arithmetic mean which can be givenas (Ott, 1978);
\00 r----------------.,
80
60
o~
40
WQI-= 1.n
n
~
i = 1qi (1)
20
where - qi is the quality of the i th parameter(subindex quality) and;
- n is the number of parameter to be inclu-
51 52 53 54 5S 56 57 58 59Sampling Station
Fig 7. WQI values for the various sampling sites.
ded.
Fig 6. pH and temperature values for the differentsampling sites.
CONCLUSIONThe results show that the changes in water qualityof Sg. Langat can be explained by the landusecharacteristics of the river basin. Similar trendswith respect to distance were obtained for thevarious water quality parameters. The waterquality at the upper catchment is very clean witha BOD less than 1.0 mg/l and DO at greater than95% saturation. This makes the water a goodsource of potable water. Unfortunately the uptakefor potable water supply is further downstream atCheras (between S3 and S4) and another afterDengkil. The most polluted section of the riveris between S4 and S8. This section includes theKajang town, the worst point in the river. Theorganic loading to the river is mainly from nonpoint sources rather than specific large industries.However, the discharges by the industries aresignificant.
BOD and suspended solids were used in the com·putation. The highest value of 87 is for stationS1 and station S6 at Kajang has the lowest valuei.e. 66. At the last station at Dengkil, the waterquality appears to have recovered slightly. Thewater quality index gives a relative indication ofthe quality of the water along the river. A riverwith a WQI value in the ranges 0-31, 32-57,58-78 and 79-100 would be considered grossly,moderately, slightly polluted and clean respectively. Using this classification, the river would onlybe termed slightly polluted at Kajang. However,these are only rough indications and the divisionsare rather subjective.
12.0
33.0
29.S
40.0
26.0_~
22.5i
"19.0~
5IS.Sf-
..1! S2 ---1'.....
• _ Temperature 36.5
• - pH
..
2.3
The subindex quality (qi) is obtained fromrating curves which range from 0 (worst waterquality condition)' to 100 (best water qualitycondition). The rating curves are determined bythe Delphi's method, which is based on a questionaire survey carried out on a panel of expertsin water quality (Ott, 1978). The WQI calculatedfrom the above equation ranges from 0 to 100.A higher value would indicate a better qualityof water. The qi value for Malaysian conditionscan be obtained from Norhayati (1981). Usingthese rating curves and equation (l) the WQ1obtained for various points in Sg. Langat is givenin Figure 7. Based on the available data, 5 parameters namely pH, temperature, dissolved oxygen,
8.5
1.0 l--~_~__---.._~_.-------.._........__~ S.OSO S7 64 71 78 8S 92 99 106 113 120
Distance from Estuary (km)
7.5
3.6
6.2
49
8.8
10.1
11.4
12.7·
14.0
280 PERTANlKA VOL. II NO.2, 1988
WATER QUALITY PROFILE OF SG. LANGAT
REFERENCES
ABU BAKAR JAAFAR. (1985): The state of the environment in Malaysia: Public sector perspective. Paperpresented at the Seminar on the State of Environment Quality in Malaysia. Kuala Lumpur. Ministryof Science, Technology and Environment. 8-11July 1985.
APHA. (1975): Standard Methods for the Examination ofWater and Wastewater. 14ed. Washington. American Public Health Association.
BALL, R.O and R.L. CHURCH (1980): Water qualityindexing and scoring. J. of the EnvironmentalEngineering Divi9ion. 105(4), 757-771.
BIGGAR, J.W. (1979): Soil and Water Salinity. Department of Land and Water Resources. University ofCalifornia, Davis.
DEPARTMENT OF ENVIRONMENT. (1980-1984):River water quality data. (unpublished)
DEPARTMENT OF ENVIRONMENT (1985): Factorydischarged data (unpublished).
HORTON, R.K. (1965): An index - Number system forrating water quality. J. Water Poll. Control. 37(3),300-306.
HYNES, H.B.N. (1970): The Ecology of Running Waters.Toronto. University of Toronto Press.
LAW, A.T. (1980): Sewage pollution in Kelang River anditsestuary.Pertanika. 30U, 13-19.
METCALF AND EDDY INC. (1979): WasterwaterEngineering: Treatment Disposal and Reuse.Second edition. New York. McGraw-Hill BookCo.
MOK, T.P. (1986): Water quality changes in Sg. Langat.B.Sc.(Env.). Project Report 1985/86. UnivenitiPertanian Malaysia. 108 p.
NORHAYATI MUSTAPHA. (1981): Indices for waterquality assessment in a river. Master's Thesis.Environmental Engineering Department. Bangkok.Asian Institute Technology.
OTT. W.R. (1978): Environmental Indices - Theoryand Practice, Michigan. Ann Arbor Science, AnnArbor.
STEEL, R.G.D. and J.H. Torrie (1960): Principles andProcedures of Statististics. New York. McGrawHill Book Company.
TAN, W.T. and B.S. NG, (1980): A chemical survey ofthe Batang Belungkung River and the GombakRiver. Pertanika. 3(1),40-46.
TANJI, K. (1979): Water quality and pollution. InIntroduction to Water Science by Silk, W. et a1. ,.Department of Land and Water Resources. University of California, Davis.
(Received 22 May, 1987)
PERTANIKA VOL. 11 NO.2, 1988 Z81