ARTICLE IN PRESS

Water Research 38 (2004) 2449–2459

Distribution of linear alkylbenzenes (LABs) in riverine and coastal environments in South and Southeast Asia Kei O. Isobea, Mohamad P. Zakariab, Nguyen H. Chiemc, Le Y. Minhc, Maricar Prudented, Ruchaya Boonyatumanonde, Mahua Sahaf, Santosh Sarkarf, Hideshige Takadaa,* a

Faculty of Agriculture, Tokyo University of Agriculture and Technology, Fuchu, Tokyo 183-8509, Japan Faculty of Science and Environmental Studies, Universiti Putra Malaysia, Serdang 43400, Selangor, Malaysia c College of Agriculture, Can Tho University, 3/2 Street, Can Tho City, Can Tho Province, Viet Nam d Science Education Department, De La Salle University, 2401 Taft Avenue, Manila 1004, Philippines e Environmental Research and Training Center, Technopolis, Amphoe Klong Luang, Pathumthani 12120, Thailand f Department of Marine Science, University of Calcutta, 35 Ballygunge Circular Road, Kolkata 700 019, India b

Received 20 October 2003; received in revised form 2 February 2004; accepted 2 February 2004

Abstract This paper reports the result of sewage pollution monitoring conducted in South and Southeast Asia during 1998– 2003 using linear alkylbenzenes (LABs) as molecular tracers of sewage contamination. Eighty-nine water samples collected from Malaysia, Vietnam, and Japan (Tokyo), and 161 surface sediment samples collected from Tokyo, Thailand, Malaysia, Philippines, Vietnam, Cambodia, Indonesia, and India were analyzed for alkylbenzenes. The P concentration range of LABs in river water particles in Southeast Asia (o0.005–0.913 mg/L) was comparable to or higher than those found in Tokyo (o0.005–0.638 mg/L). I/E ratios (a ratio of internal to external isomers of LABs) in tropical Asian waters were close to the value of LABs in raw sewage (B1) and much lower than those in secondary P effluents (3–5). This suggests that untreated or inadequately treated sewage is discharged into the water. LABs concentrations in sediments from South and Southeast Asia ranged from o0.002–42.6 mg/g-dry with the highest P concentration occurring at several populous cities. Low I/E ratios of the sediments with high LABs concentrations suggest a heavy load of untreated sewage. Clearly in view of the current data and evidence of the implications of sewage pollution, this paper highlights the necessity of the continuation of water treatment system improvement in tropical Asia. r 2004 Elsevier Ltd. All rights reserved. Keywords: Linear alkylbenzenes; Molecular marker; I/E ratio; Southeast Asia; South Asia; Sewage pollution monitoring

1. Introduction Rapid population growth and urbanization in most developing countries in South and Southeast Asia since the 1990s have been accompanied by high health risks due to increasing inputs of untreated domestic waste*Corresponding author. Tel.: +81-42-367-5825; fax: 81-42360-8264. E-mail address: [email protected] (H. Takada).

water to rivers and coastal environments. Establishment of municipal sewage treatment service cuts the load of sewage discharged into the aquatic environment, and it is one of the effective countermeasures to reduce the risk. However, the coverage of municipal sewage treatment is estimated to be low in tropical Asian countries (e.g., 21.5% in Bangkok, Thailand, 11.0% in Manila, Philippines, and 0.4% in Jakarta, Indonesia; [1]). In many areas, especially in the rural areas, such statistical information on the coverage of sewage

0043-1354/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2004.02.009

ARTICLE IN PRESS 2450

K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

treatment service is not available. Therefore, it is important to measure the types and magnitude of wastewaters discharged into the aquatic environment. Sewage pollution monitoring is an urgent necessity for water quality control and future plans for sewage improvements in these regions. Despite the increasing application of molecular markers to trace domestic waste inputs in several developed countries [2], the use of molecular markers is yet to be implemented in tropical Asian countries. Previously, we reported the fecal pollution monitoring using coprostanol in the Mekong Delta, Vietnam, and western Malaysia, as part of a monitoring project covering a large part of South and Southeast Asia [3,4]. In this article, we will present the result of the project for sewage pollution monitoring using linear alkylbenzenes (LABs) in water and sediments. LABs with C10–C14 normal alkyl chains are industrially sulfonated to produce linear alkylbenzene sulfonates (LAS), one of the most widely used surfactants in the manufacture of detergents since the early 1960s [5]. Approximately 1–3% of LABs escape sulfonation during the synthesis process of LAS, and they are impurities in commercial products [6,7]. Thus, domestic wastewater contains LABs which will be subsequently discharged into the aquatic environment [8]. Because LABs are more resistant to microbial degradation than LAS [7], they have been widely utilized for monitoring sewage inputs [7–10]. LABs are hydrophobic (Log Kow of C10 homologs: 6.90–7.06; C14 homologs: 9.16–9.29) [11] and associated with particulate matter in sewage and water column which are readily incorporated into bottom sediments. Detections of LABs have been reported in water column particles [12], riverine and marine sediments [13–19] and marine organisms [20,21]. Recently, it has been proposed that alkylbenzenes in mussels from South and Southeast Asian coasts could be used as molecular tools to assess sewage impact in tropical regions [22]. LABs homologs consist of isomers with different phenyl-substitution positions on the alkyl chains. External isomers (i.e., isomers whose phenyl substitutional positions are near the terminal end of the alkyl chain) are more susceptible to aerobic microbial degradation than internal isomers (i.e., isomers whose substitutional positions are near the center of the alkyl chain). Therefore, the isomeric distribution of LABs provides information related to the degree of biodegradation of LABs [23]. Furthermore, the isomer distribution of LABs can be applied to obtain information on types of sewage discharged into the aquatic environment (e.g., raw sewage vs. secondary effluents) [22]. Understanding the types of sewage input is essential for the establishment of sewage treatment systems. This is especially important in tropical Asia where sewage treatment plants are still under construction.

This paper is the first report on sewage pollution monitoring using LABs as molecular markers in water and sediments in tropical Asia. Use of hydrophobic molecular marker such as LABs is advantageous in extensive monitoring studies which require transport of samples to distant laboratories. Moreover, estimation of the quality or type of sewage treatment based on the information obtained from the isomer distribution of LABs is important for assessment of water quality control policy in the future.

2. Materials and methods Water samples were collected from 15 sites in western Malaysia (September, 1999), 44 sites from the Mekong Delta, Vietnam (October, 2000), and 30 sites in the Tokyo metropolitan area (July 2001). The samples collected in the Mekong Delta included nine groundwater sites, and other samples were mostly freshwater samples from rivers and canals. The detailed map of the sampling sites shall be referred to elsewhere [3,4]. Additionally, for the purpose of examining the concentrations, isomeric composition, partitioning between particulate and dissolved phase, and removal efficiencies of LABs, grab samples of raw sewage (influents) and secondary effluents were taken from five sewage treatment plants (STPs) located in the Tokyo metropolitan area (Fig. 1). Surface sediment samples were collected from 19 sites from Tokyo (Japan), 36 sites from Thailand, 13 sites from Malaysia, 15 sites from the Philippines, 35 sites from Vietnam, 11 sites from Cambodia, 20 sites from Indonesia, and 12 sites from India (Fig. 1). The location of each site was initially chosen to represent all the major regions of the countries including both urban and rural areas. In tropical Asian countries, most raw sewage is directly discharged into rivers and coastal zones, especially in rural areas due to lack of sewage treatment systems. On the other hand, modern sewage treatment systems have been partly installed in some urbanized areas such as Kuala Lumpur (Malaysia), Bangkok (Thailand), and Manila (Philippines). However, insufficient coverage of the system, poor management, and overflow of sewage caused by frequent heavy rain introduce large amounts of sewage to rivers and coastal waters. All water samples were collected using a stainless steel bucket and stored in solvent-rinsed 3 L amber glass bottles. The samples (0.2–2 L) were filtered with prebaked 0.7 mm pore size glass fiber filters (GF/F, Whatman, UK) within 6 h after collection. The filters (with trapped particulate matter) were then stored at –30
C until analysis. The dry weight of particles on the filter was measured for all samples. The surface sediments (top B5 cm layer) were collected using an Ekman dredge

ARTICLE IN PRESS K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

2451

Fig. 1. Maps showing the location of sampling sites of (a) a part of India, Indonesia, Thailand, and Philippines, (b) Cambodia, (c) Tokyo, Japan, (d) Manila, Philippines, (e) Jakarta., Indonesia, (f) Calcutta, India, and (g) Thailand. The detailed location of sampling sites of Malaysia and Vietnam is shown elsewhere [2]. Open circles on (c) indicate locations of water sample collection whereas solid circles indicate locations of sediment sample collection.

ARTICLE IN PRESS 2452

K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

and stored in solvent-rinsed stainless steel containers with Teflons liners at –30
C until further analysis. A precisely weighed sediment sample (1–5 g-dry) or a filter sample containing suspended particles was freeze dried, spiked with alkylbenzene recovery surrogate and consecutively ultrasonically extracted with three solvents, methanol, dichloromethane/methanol (1:1, v/v), and dichloromethane, as described elsewhere [3]. A part of the samples were extracted using Soxhlet apparatus with dichloromethane/methanol (2:1, v/v) for 12 h at cycling rate of 10–15 min/cycle instead of ultrasonic extraction. The difference in the analytical results was compared between ultrasonic extraction and Soxhlet extraction using a standard reference material, SRM 1941a (National Institute of Standards and Technology, Gaithersburg, MD). The difference in extraction efficiency between the two methods was tested by four replicate analyses to confirm that it was negligible (relative percent difference=3.2%). The extracts were then subjected to purification and instrumental analysis as described elsewhere [24]. Briefly, the extracts were purified and fractionated using two-step silica gel column chromatography, and the alkylbenzene fraction was determined by gas chromatography-mass spectrometry (GC–MS) in selected ion monitoring mode at m=z ¼ 91; 92, and 105. Twenty-six congeners of LABs were quantified by comparing the integrated peak area of the summed quantification ion (m=z ¼ 91+92+105) with that of the injection internal standard (biphenyl-d10, m=z ¼ 164). Structures of linear alkylbenzenes are expressed as ‘‘m-CnLAB’’ where ‘‘m’’ refers to the phenyl substitution position on the alkyl chain and ‘‘n’’ refers to the number of alkyl carbons. The P sum of all the 26 congeners is expressed as LABs. A typical chromatogram of LABs in a sediment sample is shown in Fig. 2a. In some sediment samples, measurements of LABs were interfered by tetrapropylene-based alkylbenzenes (TABs) since they overlap C10 and C11 congeners under the present chromatographic condition as shown in Fig. 2b. Fig. 2c shows a chromatogram of TABs standard. Concentrations of TABs are expressed as the sum of the peaks designated by letters of L, M, P, Q, R, S, T, U, V, Y, and Z (Fig. 2c). Because O, S’, and X coeluted with 6-, 3-, and 2-C11LAB, respectively, and peak N was often difficult to identify, they were excluded from quantification [25]. All samples were spiked with 1-CnLAB mixture (n=8–14) in isooctane (100 mL of 5 ng/mL) as an alkylbenzene surrogate internal standard prior to extraction and the concentrations were corrected based on surrogate recovery. For quality assurance, the recovery and reproducibility were tested by four replicate analyses of SRM 1941a and satisfactory results (recovery=83–109%, coefficient of variation=0.3–10.2) were obtained. Procedural blanks were prepared along with every batch of sample analysis, and usually B1 ng

Fig. 2. Gas chromatogram of alkylbenzenes in sediments collected from the Philippines, PR-2 (a), Indonesia, JK-9 (b), and TABs standard (c). IIS (internal injection standard, biphenyl-d10) was monitored at m=z ¼ 164: Surrogates (1-Cm, m:8-14; from left to right) indicated by asterisks are partially erased during figure processing. Subscripts indicate the alkyl chain length. Numbers on the peaks indicate the phenyl substituted position on the alkyl chain. Letters (LBZ) indicate peaks of TABs.

P of LABs per sample corresponding to B0.5 ng/L or B0.2 ng/g-dry sediments were detected. Quantification limits were set at 10 times the procedural blank values.

ARTICLE IN PRESS K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

3. Results and discussion 3.1. LABs in sewage influent and effluent To investigate the removal efficiencies of LABs in STPs, influent and effluent samples taken from five STPs located in the Tokyo metropolitan area were analyzed. All STPs employed activated sludge treatment followed by physical treatment (settlement) in their treatment P process. Concentrations of LABs in the particulate phase of influent and effluent samples ranged from 5.70 to 12.5 mg/L and from 0.092 to 0.180 mg/L, respectively (Fig. 3), and the calculated removal rate was 97.0– 99.2%. This was consistent with the previous reports [7,26]. The present study also examined the LABs partitioning in water by analysis of both dissolved and particulate phase of the STP influents and effluents, and confirmed that greater than 88.6–97.3% were present in the particulate phase for both influent and effluent samples. This was consistent with physico-chemical nature of LABs, i.e. Log Kow ranging from B7 for C10 homologs to B9 for C14 homologs. On the basis of incubation experiments and field observations, it has been suggested that the isomeric composition of LABs provides information regarding the degree of biodegradation [6,23,25]. In detergents and untreated sewage, the abundance of each isomer with a given alkyl chain length is almost evenly distributed,

2453

whereas in treated sewage effluent, internal isomers dominate over external isomers [7]. This is due to the selective biodegradation of the external isomers relative to the internal isomers [23,27]. Takada and Ishiwatari [23] proposed the internal/external ratio for the C12 homologs (I/E ratio; a ratio of sum of 6- and 5-C12LAB relative to sum of 4-, 3-, and 2-C12LAB) to quantitatively evaluate the degree of LAB degradation [23]. A high I/E ratio indicates a high degree of LAB degradation. Consistent with previous reports [2], I/E ratios of influent samples ranged from 0.9 to 1.1, whereas those of effluent samples increased to the range of 3.0–5.6 (Fig. 3).

3.2. LABs in water P Concentrations of LABs in water samples collected in Tokyo, Malaysia, and the Mekong Delta ranged from o0.005–0.638 mg/L, from 0.007 to 0.913 mg/L, and from o0.005–0.867 mg/L, P respectively (Fig. 4). Significant concentrations of LABs were detected at all stations except for four out of nine groundwater samples in the Mekong Delta and six headwater samples in the Minamiasakawa River and Yajigawa River in Tokyo. Higher concentrations of LABs were more often found in Southeast Asian waters than in the urban rivers of Tokyo. The highest concentrations in Southeast Asia were observed at urban sites such as Kuala Lumpur, Malacca, and George Town in Malaysia, and Can Tho

ΣLABs concn. ΣLABs concn. Influent Japan Effluent

Malaysia 0.01

0.1

1

(a)

10

100

Vietnam

(µg/L) 0.0

I/E ratio

0.2

0.4

(a) Influent

0.6 (µg/L)

0.8

1.0

I/E ratio Japan

Effluent 0

1

2

3

4

5

6

Malaysia

(b)

(I/E ratio) P Fig. 3. LABs concentration (a) and I/E ratio (b) in influent and effluent samples from STPs in Tokyo metropolitan area. Solid circles indicate influent samples and open circles indicate P effluent samples. [Note: LABs: sum of 26 LAB congeners; I/E ratio: a ratio of (6-C12LAB+5-C12LAB) relative to (4-C12LAB+3-C12LAB+2-C12LAB).]

Vietnam 0 (b)

1

2

3 (I/E ratio)

4

5

6

P Fig. 4. LABs concentration (a) and I/E ratio (b) in water samples collected from Japan, Malaysia, and Vietnam.

ARTICLE IN PRESS 2454

K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

City in the Mekong Delta. On the other hand P at rural sites in Malaysia and in the Mekong Delta, LABs concentrations were generally lower than 0.1 mg/L which suggest less input of sewage and/or dilution by large mass of river water. I/E ratios in Tokyo, Malaysia, and the Mekong Delta ranged from 0.7 to 5.5, from 0.6 to 1.6, and from 0.7 to 1.3, respectively (Fig. 4). The samples with I/E ratios lower than 2.0 in Tokyo were all collected in the Minamiasakawa River and the Yajigawa River which receive less efficiently treated sewage P in its lower stream (M-5 to M-11, Y-2 to Y-5). The LABs concentrations in these sites ranged from 0.113 to 0.638 mg/L, and I/E ratios ranged from 0.7 to 1.2 which clearly indicate inputs of untreated domestic waste. Other surveyed rivers in Tokyo such as the Sumidagawa River and the Tamagawa River had much higher I/E ratios (2.3–5.5) probably due to almost 100% coverage of sewerage service. Interestingly, lower I/E ratios (1.1–2.7) compared to the present study have been reported in 1980s in the same rivers [7], indicating the consequence of the upgraded sewage coverage from B70% in 1980s to B100% in the present. On the other hand, in Malaysia and the Mekong Delta, the I/E ratios were all close to the values of poorlyPdegraded LABs including the sites with the highest LABs concentrations, indicating heavy loads of untreated domestic waste which definitely contain a variety of harmful substances such as pathogenic organisms, steroid hormones, pharmaceutiP cals, and personal care products. LABs concentrations tended to be higher in canals than in large rivers probably because of lower water flow and less dilution in canals. In addition, lower I/E ratios were more frequently observed in canals because urban canals often function as wastewater drainage, which means that they are close to the source of pollution. This suggests that not only major rivers but also local canals should be included when selecting water quality monitoring sites in order to grasp the maximum pollution level in the region. Moreover, the results emphasize the necessity of improved sewage collection system and pollution control in urban canals. 3.3. LABs in sediments P Concentrations of LABs in sediment samples collected in Tokyo, Thailand, Malaysia, Philippines, Vietnam, Cambodia, Indonesia, and India ranged from 0.003 to 5.86 mg/g-dry, from 0.003 to 14.1 mg/g-dry, from 0.004 to 8.59 mg/g-dry, from 0.056 to 13.0 mg/g-dry, from 0.003 to 8.65 mg/g-dry, from o0.003 to 4.20 mg/g-dry, from o0.003 to 42.6 mg/g-dry, and from 0.002 to 4.45 mg/g-dry, respectively (Table 1, Fig. 5). Significant concentrations of LABs were detected in all samples except for two stations each in Cambodia, Indonesia, and India. LABs chromatograms were interfered by

several unknown peaks in four out of 11 sites in Cambodia (CASRKC, CAKKPK, CASVTR-1, CASVTR-2) and their concentrations were not determined. An unknown peak overlapped on 2-C12LAB was found in sediments with extremely low LABs concentrations, in which case 2-C12LAB was excluded from quantification. A number of stations in each country exceeded the maximum P concentration found in Tokyo. In Thailand, high LABs concentrations of 11.2, 11.9, and 14.1 mg/ g-dry were found P in several canals at St.I, St.C, and St.B, respectively. LABs concentrations increased from 0.003 mg/g-dry at CP-13 in the upper Chao Phraya River in Ayuthaya province to 0.747 mg/g-dry at CP-3 in the lower stream in Bangkok. This indicates that the main contribution of sewage inputs to the Chao Phraya River is ascribed to the lower stream in the vicinity of Bangkok. In Malaysia, high concentrations of 8.59, 3.00, and 1.08 mg/g-dry were detected in the Klang River (Kuala Lumpur), Pinang River, and Malacca River, respectively, while the rest were generally low, i.e., below 0.2 P mg/g-dry. This observation was consistent with LABs found in the water samples collected at the same locations. In the Philippines, the concentration increased from 0.089 mg/g-dry at Laguna de Bay (PR-13) to 13.0 mg/g-dry at the mouth of the Pasig River (PR-2). This suggests that the Pasig River carries a large amount of sewage to Manila Bay. P The Mekong Delta in Vietnam had generally low LABs concentration of below 0.1 mg-g/dry except for canals adjacent to human habitation. This observation was also consistent with P LABs in the water samples P collected at the same locations. The maximum LABs concentrations in sediments collected from Hai Phong Province in Vietnam was P0.892 mg/g-dry (HP-3). The mean concentrations of LABs in Cambodia and India (0.014 and 0.014 mg/g-dry, respectively) were one order of magnitude lower P than that in Japan (0.217 mg/g-dry), while the maximum LABs concentrations reached to the similar level in Phnom Penh City (CAPPTT: 4.20 g/g-dry) and in Kolkata (HG-5:P 4.45 mg/g-dry). Extremely high concentrations of LABs were observed in Jakarta, Indonesia ranging from 5.29 to 42.6 mg/g-dry (JK-1–JK-9). LABs concentrations in sediments and suspended particles collected at the same locations during the surveys in Malaysia and Vietnam were compared on a dry P weight basis (mg/g-dry). The respective ranges of LABs concentrations in the suspended particles from Malaysia and Vietnam were 0.007–16.9 mg/g-dry and 0.013–40.6 mg/g-dry, while those in the sediments were 0.004–8.59 g/g-dry and 0.003–8.65 mg/g-dry. The P LABs concentrations in the sediments were one order of magnitude lower than those in the suspended particles. This can be explained by a mechanism in which LABs in the rivers and coastal waters are

ARTICLE IN PRESS K.O. Isobe et al. / Water Research 38 (2004) 2449–2459 Table 1 Concentrations of alkylbenzenes in sediments P aP No. Site LABs bI/E ratio c TABs (mg/g-dry) (mg/g-dry) Japan 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

T-7 T-8 T-9 T-10 T-11 T-12 T-13 S-5 S-6 S-8 S-9 S-10 S-11 E-1 E-2 A-1 K-1 TS-1 TB-1

Thailand 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

CP-2 CP-3 CP-4 CP-5 CP-6 CP-8 CP-9 CP-10 CP-11 CP-13 St.A St.B St.C St.D St.F St.G St.H St.I GT-2 GT-3 GT-4 GT-6 GT-11 GT-12 TC-1 TC-2 TC-3 CS-2 CS-3 CS-4 CS-5 CS-6 CS-8 CS-10

0.003 0.027 0.038 0.284 0.100 0.193 0.013 5.86 1.15 2.16 0.983 2.02 2.52 0.109 0.105 0.811 0.086 0.548 0.039

2.4 2.5 3.0 2.6 2.9 2.5 1.2 2.5 5.9 3.1 2.8 5.1 3.7 2.3 1.9 3.0 1.8 3.1 6.0

d n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

0.411 0.747 0.073 0.515 0.696 0.052 0.032 0.023 0.006 0.003 0.317 14.1 11.9 3.10 1.34 0.223 6.77 11.2 0.220 0.160 0.243 0.285 0.047 0.130 0.331 0.943 1.11 0.012 0.007 0.022 0.008 0.047 0.014 0.004

0.8 2.2 2.1 2.5 3.5 3.5 0.9 3.3 0.7 0.8 2.4 1.4 1.4 1.2 1.7 2.1 1.3 1.2 3.0 3.6 3.6 4.8 4.4 5.9 2.3 1.2 1.2 0.8 2.9 2.3 1.5 0.7 1.3 0.3

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

2455

Table 1 (continued) aP

P LABs bI/E ratio c TABs (mg/g-dry) (mg/g-dry)

No.

Site

35 36

CS-12 CS-14

0.020 0.007

3.2 1.1

n.d. n.d.

Malaysia 1 2 3 4 5 6 7 8 9 10 11 12 13

Kim Kim Estuary Kim Kim River Kuala Selangor Malacca Muar River Nibong Tebal Pinang Estuary Port Klang Prai River Teluk Intan MYJBPL MYKESI MYPEPB

0.006 0.122 0.023 1.08 0.032 0.168 3.00 8.59 0.025 0.004 0.066 0.013 0.055

1.2 1.8 2.0 2.0 2.8 2.1 1.5 0.7 3.4 2.8 4.8 1.2 2.1

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

Philippines 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

MB-1 MB-2 MB-3 MB-4 PR-1 PR-2 PR-3 PR-5 PR-7 PR-8 PR-9 PR-10 PR-12 PR-13 Calatagan

2.314 0.800 0.220 0.586 1.37 13.0 2.03 0.299 8.50 1.73 0.176 0.346 0.514 0.089 0.056

2.9 2.6 1.3 1.8 1.4 1.3 1.2 0.6 1.3 1.4 1.5 1.3 1.6 0.6 1.1

0.361 0.124 0.042 0.070 n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

Cambodia 1 2 3 4 5 6 7 8 9 10 11

CAKDCA CAKKDT CAKTKP CASRKC CAKHKH CAKKPK CAPPSM CAKTTC CASVTR3 CASVTR4 CAPPTT

0.003 o0.002 o0.002 0.002 n.c. n.c. 1.58 0.002 n.c. n.c. 4.20

1.7 n.c. n.c. 0.8 n.c. n.c. 1.4 0.9 n.c. n.c. 1.3

n.d. n.d. n.d. o0.001 n.d. n.d. 0.470 o0.001 n.d. n.d. 0.587

Vietnam 1 2 3 4 5 6 7 8

AG-1 AG-2 AG-3 AG-4 CT-1 CT-2 CT-3 CT-4

1.2 0.8 1.1 1.0 1.1 1.0 0.9 1.3

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

0.007 0.005 0.005 0.021 0.009 0.005 0.003 0.008

e

ARTICLE IN PRESS K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

2456 Table 1 (continued) No.

Site

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

CT-5 CT-6 CT-7 CT-9 CT-11 CT-12 CT-13 ST-1 ST-2 ST-3 TV-1 TV-2 VL-1 VL-2 VL-3 VL-4 VL-7 HP-1 HP-2 HP-3 HP-4 HP-5 HP-6 HP-7 HP-8 HP-9 HP-10

Indonesia 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 India 1 2 3 4 5 6

Table 1 (continued) P aP LABs bI/E ratio c TABs (mg/g-dry) (mg/g-dry) 8.65 0.007 0.012 0.115 0.244 0.012 0.077 0.004 0.022 0.092 0.011 0.010 0.003 0.005 0.004 0.063 0.011 0.472 0.009 0.892 0.038 0.113 0.094 0.028 0.098 0.097 0.075

1.2 1.2 1.5 1.5 0.6 1.5 1.3 0.9 2.2 1.8 0.8 1.5 1.1 1.3 0.8 2.0 0.7 1.2 1.7 1.2 1.7 1.7 1.6 1.7 1.8 1.5 1.7

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

IDJKAC IDBLMD IDBDCB IDJKCL IDGJSB IDJKKA IDJAKT IDLBPN IDMRUP IDLATH JK-1 JK-2 JK-3 JK-4 JK-5 JK-6 JK-7 JK-8 JK-9 JK-10

0.069 0.089 0.004 n.c. 0.127 0.300 0.004 o0.002 o0.002 0.003 42.6 28.2 5.29 6.38 34.0 12.3 6.42 0.273 23.8 0.284

1.7 1.0 1.4 n.c. 1.1 1.1 1.1 n.c. n.c. 1.0 0.9 1.0 2.1 1.0 1.1 1.0 1.1 1.6 0.9 1.5

0.081 0.016 o0.001 2.32 0.078 0.242 o0.001 o0.001 o0.001 o0.001 7.72 5.67 0.924 1.95 6.90 2.41 1.61 0.044 8.85 0.121

INMHMB INKRCH INWBDH INMDEN INGOGA INMDKD

0.795 0.574 o0.002 0.415 0.007 o0.002

0.5 2.1 n.c. 1.3 1.7 n.c.

n.d. n.d. n.d. n.d. n.d. n.d.

No.

Site

7 8 9 10 11 12

INVZVN HG-1 HG-2 HG-3 HG-4 HG-5

aP

P LABs bI/E ratio c TABs (mg/g-dry) (mg/g-dry) 0.002 n.c. 0.033 0.513 0.027 4.45

1.4 n.c. 1.8 1.5 1.4 1.5

n.d. n.d. n.d. n.d. n.d. n.d.

a

Sum of the 26 LAB congeners except for Cilincing. Ratio of (6-C12LAB+5-C12LAB) relative to (4C12LAB+3-C12LAB+2-C12LAB). c Sum of L, M, P, Q, R, S, T, U, V, Y, and Z. d Not detected. e No calculation was made due to overlapping TABs peaks or low concentrations. b

adsorbed onto sewage particles; sewage particles are fine, light, and readily resuspended from the bottom sediment containing large amounts of relatively heavy particles such as soil particles. Resuspension of these fine sewage particles with LABs probably resulted in higher P LABs concentrations in the suspended particles compared to those in the sediment. However, in Jakarta, P the LABs concentrations in P sediments were almost comparable to the weight-base LABs concentrations in raw sewage, i.e., STP influents in Tokyo of 7.51– 81.4 mg/g-dry (mean=42.3 mg/g-dry, n=5). This implies that sewage particles are deposited in the locations without significant dilution with soil particles. I/E ratios of LABs in sediment samples collected in Tokyo, Thailand, Malaysia, Philippines, Vietnam, Cambodia, Indonesia, and India ranged from 1.2 to 6.0, from 0.7 to 5.9, from 0.7 to 4.8, from 0.6 to 2.9, from 0.6 to 2.2, from 0.8 to 1.7, from 0.9 to 2.1, and from 0.5 to 2.1, respectively (Table 1, Fig. 5). Lower I/E ratios were observed in South and Southeast Asian sediments compared to those in Tokyo. I/E ratios in South and Southeast Asian sediments were much lower than those in secondary effluents (3–5) and close to those in raw sewage (B1). This indicates intensive input of untreated or inadequately treated sewage into the water due to deficiency of sewage treatment plants and inadequate management of the plants. Also, overflow of sewage caused by frequent and heavy rain inherent to the tropical climate contributes to the input of untreated sewage to the aquatic environments. Because sewage contains many toxic components as well as pathogenic bacteria and viruses, discharge of poorly treated sewage may endanger human health in the region. Although I/E ratios in sediments from South and Southeast Asian waters were low on the whole, relatively high I/E ratios were observed in sediments collected from >10 km off the coast e.g., 5.9, 4.8, and 4.4 in the

ARTICLE IN PRESS K.O. Isobe et al. / Water Research 38 (2004) 2449–2459 ΣLABs concn. Japan Thailand Malaysia Philippines Vietnam Cambodia Indonesia India 0

10 (µg/g-dry)

5

(a)

15

30 40

I/E ratio Japan Thailand Malaysia Philippines Vietnam Cambodia Indonesia India 0

(b)

1

2

3

4

5

6

7

(I/E ratio)

P Fig. 5. LABs concentration (a) and I/E ratio (b) in sediment samples collected from Japan, Thailand, Malaysia, Philippines, Vietnam, Cambodia, Indonesia, and India. Solid circles indicate riverine sediments and sediments collected from within B10 km from the coast while open circles indicate sediments collected P from more than 10 km away from the coast. [Note: LABs: sum of 26 LAB congeners; I/E ratio: a ratio of (6-C12LAB+5C12LAB) relative to (4-C12LAB+3-C12LAB+2-C12LAB).]

Gulf of Thailand (GT-12, GT-6, GT-11, respectively). This could be explained by degradation of LABs during lateral transport from the terrestrial source [14]. Even excluding I/E ratios of the sediments collected from offshore sites, relatively high I/E ratios were observed in sediments from Thailand and Malaysia. Installation of sewage treatment plants was reflected in an increase in I/ E ratio in the sediments. In Thailand, the average I/E ratio in canal sediments was 1.6 while much higher I/E ratios were found in the Chao Phraya River sediments (3.5, 3.5, and 3.3 in CP-6, CP-8, and CP-10, respectively). Presently, there are 6 sewage treatment plants covering 192 km2 in Bangkok city, and approx. 1 million m3/day of secondary effluent is discharged into the Chao Phraya River, whereas the canals receive exclusively untreated sewage. This could contribute to the higher I/E ratios in the Chao Phraya River than in the canals. On the other hand, in Cambodia and Indonesia, where sediments were highly contaminated by LABs and TABs, the isomeric compositions of LABs were almost the same as raw sewage. This observation emphasizes the needs for an adequate sewage system as well as further studies on human health risks in the region. High concentrations of tetrapropylene-based alkylbenzenes (TABs) were also detected at several locations in Indonesia, Cambodia, and the Philippines (Table 1). An example of the chromatogram containing both

2457

LABs and TABs is shown in Fig. 2b. In many cases, LABs prevailed over TABs, while TABs are predominant in the sediment collected at IDJKCL in Jakarta. TABs were significantly detected at 15 out of 20 locations in Jakarta (Indonesia), and the TABs concentrations in Jakarta ranged from 0.044 to 8.85 mg/gdry. TABs were also found in sediments from Cambodia at 2 out of 11 locations (0.470 and 0.587 mg/g-dry), although at lower concentrations. In the Philippines, only sediments from Manila Bay contained significant concentrations of TABs (0.042–0.361 mg/g-dry). Highly branched alkylbenzenesulfonate surfactants (ABS) were used in detergents in industrialized countries in the past. ABS were replaced by LAS in industrialized countries in the 1960s because they were found to have poor biodegradability. Nevertheless, TABs, the precursors for the ABS, are found in sediments [6,25,28] as well as in marine organisms [22]. This is partly because TABs are recalcitrant to degradation and well preserved in sediments. Detection of TABs in surface sediments can be ascribed to mixing of sediments deposited in 1960s into surface sediments through vertical mixing and remobilization of sediments in the coastal zone. This can also partially explain the detection of TABs in sediments in the present study, especially in coastal sediments such as those collected from Manila Bay. However, Tsutsumi et al. [22] reported that the presence of TABs in green mussels collected from South and Southeast Asian coasts indicates that non-degradable surfactants are still being used in some Asian countries. In fact, they confirmed that the commercially available synthetic detergents in Jakarta contained ABS [22]. The fact that the TABs concentrations in sediments from Jakarta and Phnom Penh City were comparable to or even exceeded LABs concentrations further supports the current usage of ABS detergents in the region.

4. Conclusion The analytical results of LABs found in suspended particles in water and sediments presented in this paper were obtained as part of a comprehensive study on sewage pollution in South and Southeast Asia. The high P LABs concentrations and low I/E ratios found in the sediments from populous cities in tropical Asia clearly indicated high contamination by untreated domestic waste. This paper demonstrated that it is advantageous to use LABs as molecular tracers of sewage contamination because their I/E ratio is indicative of types of sewage discharged into the water. LABs enabled to understand both quantity- and quality-based assessment of sewage pollution. Sewage will continue to be a problem, as long as there is a human population to produce it. In the coming years, the total amount of sewage discharged into the

ARTICLE IN PRESS 2458

K.O. Isobe et al. / Water Research 38 (2004) 2449–2459

rivers and coastal waters in tropical Asian countries will progressively increase and add to the pollution already affecting parts of the region. Clearly in view of the current data and evidence of the implications of sewage pollution, this paper highlights the necessity of the continuation of water treatment system improvement. Continuous assessment of the extent of sewage pollution in rivers and coastal waters can indicate the level of improvement brought about by the upgraded sewage systems. Therefore, further research concerning sewage and other anthropogenic pollutants is critical to reduction of health risks in tropical Asia.

Acknowledgements The authors wish to thank Dr. Shinsuke Tanabe and scientists who joined ‘‘Asian Mussel Watch Project’’ for kindly providing sediment samples from Cambodia, Indonesia, and India, Dr. Shin Takahashi and Dr. Daisuke Ueno for their managing the samples and Dr. Pravakar Mishra for his assistance on collecting Kolkata samples. This work was financially supported by the Ministry of Education, Culture, Sports, Science and Technology, Japan, under ‘‘Research Project for Sustainable Coexistence of Humans, Nature and the Earth’’. Our field trips were supported by Japanese International Cooperation Agency (JICA) under ‘‘JICA/UPM Technical Collaborative Project on Malacca Straits (MASDEC)’’ and ‘‘JICA/Can Tho University Project for Improvement of Environmental Education in Agricultural Sciences’’. Assistance by several graduates and undergraduates in our laboratories and fieldwork are kindly acknowledged.

[5] [6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

References [1] International Division for Infrastructure, Policy Bureau, Ministry of Land, Infrastructure and Transport, Government of Japan. General Information on Present Situation and Issues of Infrastructure, Tokyo, 1999. [2] Takada H, Eganhouse RP. Molecular markers of anthropogenic waste: their use in determining sources, transport pathways and fate of wastes in the environment. In: Meyers R, editor. Encyclopedia of environmental analysis and remediation. New York: Wiley; 1998. p. 2883–940. [3] Isobe KO, Tarao M, Zakaria MP, Chiem NH, Minh LY, Takada H. Quantitative application of fecal sterols using gas chromatography-mass spectrometry to investigate fecal pollution in tropical waters: western Malaysia and mekong delta, Vietnam. Environ Sci Technol 2002;36:4497–507. [4] Isobe KO, Tarao M, Chiem NH, Minh LY, Takada H. Effect of environmental factors on the relationship between concentrations of coprostanol and fecal indicator bacteria in tropical (Mekong Delta) and temperate

[16]

[17]

[18]

[19]

[20]

(Tokyo) freshwaters. Appl Environ Microbiol 2004;70: 814–21. Myers D. Surfactant science and technology. New York: VCH Publishers, Inc.; 1992. Eganhouse RP, Blumfield DL, Kaplan IR. Long-chain alkylbenzenes as molecular tracers of domestic wastes in the marine environment. Environ Sci Technol 1983;17: 523–30. Takada H, Ishiwatari R. Linear alkylbenzenes in urban riverine environments in Tokyo: distribution, source, and behavior. Environ Sci Technol 1987;21:875–83. Eganhouse RP, Ruth EC, Kaplan IR. Determination of long-chain alkylbenzenes in environmental samples by argentation thin-layer chromatography/high resolution gas-chromatography and gas chromatography/mass spectrometry. Anal Chem 1983;55:2120–6. Eganhouse RP, Olaguer DP, Gould BR, Phinney CP. Use of molecular markers for the detection of municipal sewage sludge at sea. Mar Environ Res 1988;25:1–22. Takada H, Ishiwatari R. Linear alkylbenzenes (LABs) in urban riverine and coastal sediments and their usefulness as a molecular indicator of domestic wastes. Water Sci Technol 1991;23:437–46. Sherblom PM, Gschwend PM, Eganhouse RP. Aqueous solubility, vapor pressure, and n-octanol-water partition coefficients for C9-C14 linear alkylbenzenes. J Chem Eng Data 1992;37:394–9. Crisp PT, Brenner S, Venkatesan MI, Ruth E, Kaplan IR. Organic chemical characterization of sediment-trap particles from San Nicolas, Santa Barbara, Santa Monica and San Pedro Basins, California. Geochem Cosmochim 1979;43:1791–801. Chalaux N, Bayona JM, Venkatesan MI, Albaiges J. Distribution of surfactant markers in sediments from Santa Monica Basin, Southern California. Mar Pollut Bull 1992;24:403–7. Chalaux N, Takada H, Bayona JM. Molecular markers in Tokyo Bay sediments: sources and distribution. Mar Environ Res 1995;40:77–92. Eganhouse RP, Kaplan IR. Depositional history of recent sediments from San Pedro Shelf, California: reconstruction using elemental abundance, isotopic composition and molecular markers. Mar Chem 1988;24:163–91. Ishiwatari R, Takada H, Yun SJ, Matsumoto E. Alkylbenzene pollution of Tokyo Bay sediments. Nature 1983;301:599–600. Murray AP, Gibbs CF, Kavanagh PE. Linear alkylbenzenes (LABs) in sediments of Port Phillip Bay (Australia). Mar Environ Res 1987;23:65–76. Raymundo CC, Preston MR. The distribution of linear alkylbenzenes in coastal and sediments of the Western North Sea. Mar Pollut Bull 1992;24:138–46. Takada H, Ogura N, Ishiwatari R. Distribution of Linear alkylbenzenes (LABs) and linear alkylbenzene sulfonates (LAS) in Tokyo Bay sediments. Estuar Coast Shelf Sci 1992;35:141–56. Albaiges J, Farran A, Soler M, Gallifa A, Martin P. Accumulation and distribution of biogenic and pollutant hydrocarbons, PCBs and DDT in tissues of western Mediterranean fishes. Mar Environ Res 1987;15: 595–8.

ARTICLE IN PRESS K.O. Isobe et al. / Water Research 38 (2004) 2449–2459 [21] Serrazanetti GP, Pagnucco C, Conte LS, Artusi R. Aliphatic hydrocarbons and linear alkylbenzenes in zooplankton from the Gulf of Trieste. Chemosphere 1994;28:1119–26. [22] Tsutsumi S, Yamaguchi Y, Nishida I, Akiyama K, Zakaria MP, Takada H. Alkylbenzenes in mussels from South and South East Asian coasts as a molecular tool to assess sewage impact. Mar Pollut Bull 2002;45:325–31. [23] Takada H, Ishiwatari R. Biodegradation experiments of linear alkylbenzenes (LABs): isomeric composition of C12 LABs as an indicator of the degree of LAB degradation in the aquatic environment. Environ Sci Technol 1990;24: 86–91. [24] Hartmann PC, Quinn J, King JW, Tsutsumi S, Takada H. Intercalibration of LABs in marine sediment SRM1941a

[25]

[26]

[27]

[28]

2459

and their application as a molecular marker in Narragansett Bay sediments. Environ Sci Technol 2000;34:900–6. Takada H, Ishiwatari R. Quantitation of long-chain alkylbenzenes in environmental samples by silica gel column chromatography and high-resolution gas chromatography. J Chromatogr 1985;346:281–90. Gledhill W, Saeger VW, Trehy ML. An aquatic environmental safety assessment of linear alkylbenzene. Environ Toxicol Chem 1991;10:169–78. Bayona JM, Albaiges J, Solanas AM, Grifoll M. Selective aerobic degradation of linear alkylbenzenes by pure microbial cultures. Chemosphere 1986;15:595–8. Eganhouse RP, Pontolillo J. Depositional history of organic contaminants on the Palos Verdes Shelf, California. Mar Chem 2000;70:317–38.