Aliphatic hydrocarbon concentrations in short sediment cores from the southern Okinawa Trough: Implications for lipid deposition in a complex environment

(1)

Continental Shelf Research 27 (2007) 2066–2078

Aliphatic hydrocarbon concentrations in short sediment cores

from the southern Okinawa Trough: Implications for lipid

deposition in a complex environment

Woei-Lih Jeng

Institute of Oceanography, National Taiwan University, Taipei, Taiwan, ROC Received 8 August 2006; received in revised form 18 April 2007; accepted 7 May 2007

Available online 18 May 2007

Abstract

Seven short sediment cores from the southern Okinawa Trough were collected and analyzed for the aliphatic hydrocarbon concentrations by capillary gas chromatography to explore the deposition of hydrocarbons to this area. For all cores studied, ratios of Shydrocarbons/TOC, (nC27+nC29+nC31)/TOC, terrigenous/aquatic, and diploptene/SC25–33 n-alkanes fluctuated around a mean value with coefficients of variation ranging from 9.0% to 19.7%, 4.9% to 20.0%, 27.3% to 129%, and 3.8% to 163%, respectively. For the nC31/(nC27+nC29+nC31) ratio, only station 21 showed fluctuation. Moreover, the carbon preference indexes in the C25–C33n-alkane range also exhibited fluctuating values with coefficients of variation of 1.9–14.4%. These results indicate that concentrations of hydrocarbon inputs to the sampling sites vary with time; this may result from complex current flow and sediment transport, leading to variable lipid deposition. In addition, significant correlation between diploptene (hop-22(29)-ene) and higher plant n-alkanes was found for cores 21, 42 and 46, indicating that diploptene was predominantly from higher plant sources. However, no correlation between diploptene and higher plant n-alkanes was found for cores 20, 36, 43 and 44; autochthonous sources of diploptene in these cores were quite probable.

1. Introduction

Observations from sediment traps show that the sediment ﬂuxes to the southern Okinawa Trough (SOT) are much higher than those in other marginal seas and display great temporal and spatial varia-bility (Hsu et al., 2004). High sedimentation rates are located approximately in the lower trough, deeper than 1000 m (range from 0.25 to 0.52 cm yr1 estimated from the excess 210Pb proﬁles of cores),

while low sedimentation rates are generally in the upper trough, shallower than 1000 m (Chung and Chang, 1995). Based on results from a site (24148.240_{N, 122130.00}0_{E) of ODP Leg 195 in the}

SOT, the sedimentation rate has always been high, reaching 325 cm kyr1(0.325 cm yr1) since the late Holocene (ODP, 2001). Therefore, the SOT is apparently an area of focused sedimentation along the path of the Kuroshio Current (KC). Further-more, the lower slope (41000 m) sediments consist almost entirely of silty mud, while the upper slope (o1000 m) sediments are composed of sand with little mud (Chen et al., 1995).

www.elsevier.com/locate/csr

Tel.: +886 2 23636040x301; fax: +886 2 23626092. E-mail address:wljeng@oc.ntu.edu.tw

(2)

Transport processes of particulate matter in the SOT have been studied in the KEEP (Kuroshio Edge Exchange Processes) program (Wong et al., 2000). From isotopic evidence, Kao et al. (2003)

conclude that a major fraction of the sedimentary organic matter in the SOT may originate from the inner shelf of the East China Sea (ECS). Twenty-three surface sediments from the SOT have been shown to receive the highest plant wax n-alkane contribution among the coastal marine areas surrounding Taiwan; lateral particle transport from the southern ECS shelf and river runoff from the east Taiwan coast are considered to be the major contributors (Jeng et al., 2003).

The lipid inputs by suspended sediment transport to the SOT derive from three sources (Jeng et al., 2003). One is from the northwest by an alongshore ﬂow just off northern Taiwan. Another is from the nearest river, Lanyang River (LR). The other is from south—the KC carries other river runoffs from eastern Taiwan to the SOT.

Marine sediments may preserve a record of past or on-going environmental processes and compo-nents, both natural and human induced. Records in marine sediment can provide information about past biological, chemical and physical oceano-graphic conditions. As to chemistry, important parameters may include mineral content, major elements, trace elements, total organic carbon (TOC), stable isotope geochemistry, speciﬁc organic pollutants, biochemical markers, etc. Organic geo-chemical data can be used to estimate the total mass of certain compounds in the bottom sediment, to interpret downcore environmental changes as in-dicated by the abundance and distribution of some compounds present in cores, etc. Because of their stability in the natural environment, aliphatic hydrocarbons are selected and used as a measure for indicating possible change of lipids in sediment cores, which would reﬂect changes in lipid concen-tration during sedimentation. The objective of this study was to use the variations of parameters derived from aliphatic hydrocarbon concentrations in 7 short sediment cores collected from the SOT for inferring the condition of lipid deposition in this area.

2. Experimental

In the present study, 7 short sediment cores were collected from the SOT using a box corer on board R/V Ocean Researcher I on cruise ORI-417 (April 24–

May 1, 1995). Sampling stations are given inFig. 1. Water depths for stations 20, 21, 36, 42, 43, 44 and 46 are 1096, 1115, 542, 805, 1360, 1465 and 1665 m, respectively. All sub-cores were kept frozen (20 1C) until analyzed. Cores 20, 42, 43, 44 and 46, respectively, had core lengths of 23, 24, 16, 23.2 and 26.5 cm; each core was sectioned at 4-cm intervals. Cores 21 and 36 had core lengths of 12.5 and 9.5 cm and were sectioned at 2-cm intervals with the last sections of 4.5 and 3.5 cm, respectively. Every section was freeze dried.

Following the addition of an internal standard (n-C24D50), the dried sediment was extracted with a

mixture of dichloromethane and methanol (1:1, v/v) for 24 h in a Soxhlet apparatus. The lipid extract was then saponified by reflux for 3 h with 0.5 N KOH solution in methanol. The non-saponifiable lipids were isolated by hexane extraction four times and concentrated by solvent evaporation under a N2

gas stream. The aliphatic hydrocarbon fraction was isolated from the neutral lipids by silica gel (deactivated with 5% H2O) column

chromatogra-phy using hexanes.

For GC analysis, an HP 5890A gas chromato-graph equipped with a split/splitless injector and a flame ionization detector (FID) was used. Separa-tion of aliphatic hydrocarbons was achieved by an SPB-1 capillary column (30 m 0.25 mm i.d. 0.25 mm). Oven temperature programming was 45–90 1C at 15 1C min1 and 90–270 1C at 3 1C min1 for analyzing aliphatic hydrocarbons. Identification was made with co-injection of authen-tic standards and gas chromatography–mass spec-trometry (GC–MS). The GC–MS analyses were performed with an HP 6890 GC (HP-1MS crosslinked methyl siloxane column, 30 m 0.25 mm i.d. 0.25 mm) interfaced directly to an HP 5973 quadru-pole mass selective detector (electron ionization, electron energy 70 eV, scanned from 50 to 550 Da). An SGE (Australia) OCI-5 cool on-column injector was also fitted in the gas chromatograph for obtaining the best quantification. GC peak areas of all hydrocarbons and the internal standard were obtained using an electronic integrator (Chromato-pac C-R6A, Shimadzu, Japan). Each hydrocarbon concentration was determined using the internal standard. Based on 8 replicate analyses, the analytical imprecision (expressed as the percent coefficient of variation) of hydrocarbon abundances was calculated to be 2–8%.

TOC was quantiﬁed with a Fison NA1500 element analyzer. Carbonate carbon was removed

(3)

by adding a few drops of 1 N HCl solution to the sediment (o0.5 g) and drying it at 50 1C in an oven. The de-carbonated sample was combusted at 1050 1C in the analyzer, and CO2 produced was

determined with a TCD detector. The relative imprecision for TOC analysis was better than 2.5% (n ¼ 9).

3. Results and discussion

Before discussing the lipid data, it is important to briefly describe the flow condition in the SOT. Referring toFig. 1, the dominant ocean current in the SOT is the KC, which flows approximately northward along the east coast of Taiwan. Its subsurface water shoals up off the northeast coast of Taiwan, and the main stream turns northeast-ward along the ECS continental slope. However, part of the KC water intrudes on the shelf forming a branch current, termed the Kuroshio Branch Current (KBC). On the western side of the KBC, a cyclonic eddy is generated over the shelf-slope. Seasonal changes in the KC path, partial or extensive intrusion of the KC onto the shelf, and

seasonal existence of the eddy can affect the deposition of sedimentary lipids (Liu et al., 1998;

Tang et al., 1999). To the northwest of the study area, a southeastward filament (a small alongshore flow) occasionally appears somewhere between the eddy and northern Taiwan (Chern et al., 1990). The flow may transport sediment from the southern ECS shelf to the SOT during its existence. Sediment contribution from the nearby LR from the west is the third source. The annual sediment discharge of the river amounts to ca. 8.0 Mt yr1 (Water Resources Bureau, 1998). The sediment discharge from this river runoff varies over time (Hsu et al., 2004; Lee et al., 2004). For instance, the major transport of river sediment to the sea reaches maximums when high rainfalls cause the river to flood especially during the landfall of typhoons.

A representative GC trace for SOT sediments is shown inFig. 2, and all hydrocarbon concentrations and related parameters are given in Table 1. In all sediments, higher molecular weight (HMW, 4C23)

n-alkanes strongly outweighed lower molecular weight (LMW,oC23) ones with strong odd-to-even

preference in the HMW range, suggesting a Taiwan 20 21 36 42 43 44 46 Longitude (E) 27° 26° 25° 24° Latitude (N)

East China Sea

Okinawa Trough Lanyang River Kuroshio Filament Eddy Ilan Ridge 121° 122° 123° 124° 500m 1000m 200m 500m 1000m 200m 500m 1000m

(4)

Fig. 2. GC traces of aliphatic hydrocarbons from station 46 (12–16 cm). Numbers above peaks indicate carbon chain lengths. Pr ¼ pristane, Ph ¼ phytane, Sq ¼ Squalene, Di ¼ diploptene, UCM ¼ unresolved complex mixtures.

Table 1

Hydrocarbon concentrations (ng g1_{) and related parameters}

Compound Core 20 (0–4 cm) Core 20 (4–8 cm) Core 20 (8–12 cm) Core 20 (12–16 cm) Core 20 (16–20 cm) Core 20 (20–23 cm) Core 21 (0–2 cm) nC15 — — — — — — 119 nC16 10.6 — — 4.57 — — 104 nC17 32.4 2.68 22.5 36.9 — 4.99 222 Pristane 13.8 1.34 9.14 11.4 — — 290 nC18 28.9 21.5 48.5 55.9 5.00 32.2 116 Phytane 10.4 6.45 16.5 20.0 — 8.32 39.5 nC19 28.9 28.3 37.5 51.4 18.2 52.2 84.5 nC20 25.0 25.2 22.5 44.9 28.2 56.3 90.4 nC21 42.7 37.0 32.0 64.7 45.7 64.7 95.2 nC22 51.8 38.6 34.0 62.4 52.2 60.1 130 nC23 109 74.6 78.9 128 118 105 168 nC24 69.1 71.6 59.8 97.8 101 94.0 109 nC25 204 137 155 225 226 190 169 nC26 94.0 69.7 73.9 108 110 93.1 112 nC27 367 252 286 406 420 312 232 nC28 129 93.6 98.1 139 147 114 126 Squalene 132 28.3 74.5 89.2 29.1 38.5 186 nC29 691 565 574 773 805 679 418 nC30 124 99.7 99.7 138 150 143 171 nC31 819 638 658 910 947 699 538 nC32 13.4 74.8 68.3 92.5 94.8 66.7 96.2 Diplop. 42.0 235 13.6 24.5 24.2 17.9 433 nC33 365 288 301 404 427 377 270 nC34 20.8 18.1 17.5 23.6 24.8 43.5 36.7 nC35 101 91.9 84.2 111 130 123 168 UCMa ₁₆₉₀ ₆₇₁ ₃₇₃ ₁₂₄₀ ₄₅₀ ₁₅₀₀₀ ₄₉₇₀ TOCb 0.56 0.57 0.47 0.50 0.61 0.54 0.68 SHC 3520 2900 2870 4020 3900 3370 4520 SHC/TOCc 6290 5090 6110 8040 6390 6240 6650 Terr/TOCd 3350 2550 3230 4180 3560 3130 1750 TARe 30.6 47.0 25.3 23.7 119 29.6 2.79 nC31/Terrf 0.436 0.438 0.433 0.436 0.436 0.414 0.453

(5)

Table 1 (continued ) Compound Core 20 (0–4 cm) Core 20 (4–8 cm) Core 20 (8–12 cm) Core 20 (12–16 cm) Core 20 (16–20 cm) Core 20 (20–23 cm) Core 21 (0–2 cm) CPIg 6.06 4.94 5.23 5.07 5.03 4.66 2.81 Di/SC25–33 h 0.017 0.13 0.0069 0.0090 0.0086 0.0079 0.27 Di/TOCi 75.0 412 28.9 49.0 39.7 33.1 637 Compound Core 21 (2–4 cm) Core 21 (4–6 cm) Core 21 (6–8 cm) Core 21 (8–12.5 cm) Core 36 (0–2 cm) Core 36 (2–4 cm) Core 36 (4–6 cm) NC15 198 — 5.28 — — — — NC16 146 15.8 41.1 — 17.6 — 27.3 NC17 271 94.6 130 7.55 50.9 34.1 88.4 Pristane 343 108 146 18.7 29.2 16.5 46.2 NC18 149 111 104 19.8 43.8 60.0 88.6 Phytane 61.7 42.2 39.8 8.47 15.8 22.6 30.7 NC19 122 119 91.4 27.2 34.2 47.4 52.9 NC20 125 141 97.8 40.9 32.6 37.7 37.5 NC21 105 108 88.2 52.8 40.1 42.0 42.8 NC22 91.8 95.8 71.2 69.7 40.9 42.8 43.9 NC23 126 131 103 79.0 76.7 89.6 91.8 NC24 129 123 102 88.5 45.5 68.5 68.9 NC25 167 171 139 108 121 166 165 NC26 117 118 88.7 63.7 63.6 84.2 79.3 NC27 244 243 205 173 235 318 308 NC28 123 138 88.9 73.0 91.1 128 114 Squalene 60.8 62.1 29.8 82.3 101 — — NC29 430 469 386 369 502 679 677 NC30 158 156 95.6 92.5 81.7 126 112 NC31 481 573 463 441 693 891 876 NC32 123 126 101 64.8 55.4 88.3 93.9 Diplop. 405 442 373 353 36.5 36.3 31.3 NC33 292 285 249 211 316 417 405 NC34 56.4 32.6 34.5 27.3 18.9 25.7 23.6 NC35 133 137 139 129 80.3 102 101 UCMa ₁₇₀₀ ₇₉₈₀ ₆₄₂₀ ₃₅₅₀ _— ₁₂₇₀ ₁₅₇₀ TOCb _0.70 _0.70 _0.63 _0.66 _0.37 _0.39 _0.39 SHC 4660 4040 3410 2600 2820 3520 3610 SHC/TOCc ₆₆₆₀ ₅₇₇₀ ₅₄₁₀ ₃₉₄₀ ₇₆₂₀ ₉₀₃₀ ₉₂₆₀ Terr/TOCd ₁₆₅₀ ₁₈₄₀ ₁₆₇₀ ₁₄₉₀ ₃₈₆₀ ₄₈₄₀ ₄₇₇₀ TARe _1.95 _6.02 _4.65 _28.3 _16.8 _23.2 _13.2 NC31/Terrf 0.416 0.446 0.439 0.449 0.485 0.472 0.471 CPIg _2.64 _2.84 _3.28 _3.73 _5.76 _5.23 _5.47 Di/SC25–33h 0.25 0.25 0.26 0.27 0.020 0.015 0.013 Di/TOCi 579 631 592 535 98.6 93.1 80.3 Compound Core 36 (6–9.5 cm) Core 42 (0–4 cm) Core 42 (4–8 cm) Core 42 (8–12 cm) Core 42 (12–16 cm) Core 42 (16–20 cm) Core 42 (20–24 cm) NC15 — 8.58 — — — — — NC16 — 24.4 6.59 — — — — NC17 31.3 50.7 61.3 — 8.09 — 7.33 Pristane 15.0 34.3 37.1 — 6.15 — 3.66 NC18 63.9 41.8 89.8 8.15 33.2 — 33.6 Phytane 22.6 14.9 31.1 — 12.2 — 11.3 nC19 46.6 28.9 63.9 26.0 40.3 10.4 44.3 nC20 34.9 28.4 49.1 38.2 40.4 24.0 44.8 nC21 41.8 32.7 57.1 54.1 55.7 43.1 56.0 nC22 43.9 55.3 52.5 58.6 57.6 55.0 58.9 nC23 89.7 92.0 82.2 95.1 98.9 97.6 100

(6)

Table 1 (continued ) Compound Core 36 (6–9.5 cm) Core 42 (0–4 cm) Core 42 (4–8 cm) Core 42 (8–12 cm) Core 42 (12–16 cm) Core 42 (16–20 cm) Core 42 (20–24 cm) nC24 68.0 55.5 70.6 80.2 81.1 77.3 80.7 nC25 163 58.4 124 135 154 149 159 nC26 82.4 33.1 72.7 80.0 88.6 84.8 90.9 nC27 311 81.6 211 225 269 259 284 nC28 117 32.5 91.0 96.5 111 106 143 Squalene 5.69 65.5 94.4 24.5 65.4 70.6 113 nC29 689 154 432 444 536 511 564 nC30 115 38.7 101 105 122 130 127 nC31 899 197 487 510 636 594 679 nC32 86.1 24.4 82.3 87.5 99.6 92.1 82.3 Diplop. 26.2 32.4 90.4 90.1 108 88.4 93.9 nC33 418 90.5 233 247 304 291 331 nC34 24.9 1.88 17.0 19.3 22.3 20.2 28.3 nC35 106 33.4 71.1 77.6 94.8 87.4 98.1 UCMa 966 — 1380 1860 1860 1560 828 TOCb 0.37 0.31 0.65 0.68 0.67 0.66 0.67 SHC 3500 1310 2710 2500 3040 2790 3230 SHC/TOCc 9460 4230 4170 3680 4540 4230 4820 Terr/TOCd 5130 1400 1740 1730 2150 2070 2280 TARe _24.4 _4.91 _9.03 _45.3 _29.8 ₁₃₁ _29.6 nC31/Terrf 0.473 0.455 0.431 0.433 0.441 0.435 0.445 CPIg _5.56 _3.82 _3.83 _3.75 _4.03 _3.93 _4.07 Di/SC25–33h 0.011 0.056 0.061 0.058 0.057 0.049 0.047 Di/TOCi _70.8 ₁₀₅ ₁₃₉ ₁₃₃ ₁₆₁ ₁₃₄ ₁₄₀ Compound Core 43 (0–4 cm) Core 43 (4–8 cm) Core 43 (8–12 cm) Core 43 (12–16 cm) Core 44 (0–4 cm) Core 44 (4–8 cm) Core 44 (8–12 cm) nC15 — — — — — — — nC16 — 14.1 — — 20.8 — 18.7 nC17 7.88 64.6 3.35 7.28 84.0 — 78.1 Pristane 11.3 37.9 5.40 5.15 81.7 — 43.1 nC18 27.7 95.9 26.9 27.8 68.9 11.6 94.7 Phytane 9.93 33.2 10.2 10.4 25.7 — 33.0 nC19 39.7 73.8 40.9 41.2 63.0 32.2 78.2 nC20 45.1 64.5 46.9 42.6 64.2 41.8 72.3 nC21 62.5 63.6 60.3 54.7 80.7 65.9 83.0 nC22 89.8 59.0 58.1 55.2 112 79.0 82.6 nC23 108 78.8 76.9 74.2 130 116 130 nC24 59.4 67.4 65.7 64.8 74.0 84.3 97.5 nC25 96.0 98.4 95.5 95.5 137 170 212 nC26 58.8 62.3 59.7 59.4 81.1 90.6 113 nC27 153 157 153 155 208 297 387 nC28 64.8 65.4 63.6 64.3 105 117 154 Squalene 138 — — — 192 96.8 67.4 nC29 288 291 282 288 378 612 782 nC30 97.1 73.2 62.2 72.8 122 115 160 nC31 384 413 377 380 491 742 898 nC32 39.9 69.5 61.0 55.2 71.8 105 120 Diplop. 182 172 105 128 303 236 149 nC33 173 184 172 173 243 309 398 nC34 15.1 15.1 12.1 10.9 36.1 19.5 37.0 nC35 71.8 67.7 62.0 62.3 112 87.3 121 UCMa _— ₂₁₀₀ ₁₁₂₀ ₁₆₉₀ ₃₂₁₀ ₂₃₃₀ ₂₄₈₀ TOCb _0.56 _0.63 _0.62 _0.61 _0.74 _0.72 _0.75 SHC 2220 2320 1900 1930 3290 3430 4410 SHC/TOCc 3960 3680 3060 3160 4450 4760 5880

(7)

Table 1 (continued ) Compound Core 43 (0–4 cm) Core 43 (4–8 cm) Core 43 (8–12 cm) Core 43 (12–16 cm) Core 44 (0–4 cm) Core 44 (4–8 cm) Core 44 (8–12 cm) Terr/TOCd ₁₄₇₀ ₁₃₇₀ ₁₃₁₀ ₁₃₅₀ ₁₄₆₀ ₂₂₉₀ ₂₇₆₀ TARe _17.3 _6.22 _18.4 _17.0 _7.33 _51.3 _13.2 nC31/Terrf 0.465 0.480 0.464 0.462 0.456 0.449 0.434 CPIg _3.69 _3.69 _3.82 _3.81 _3.34 _4.46 _4.37 Di/SC25–33h 0.17 0.15 0.097 0.12 0.21 0.11 0.056 Di/TOCi ₃₂₅ ₂₇₃ ₁₆₉ ₂₁₀ ₄₀₉ ₃₂₈ ₁₉₉ Compound Core 44 (12–16 cm) Core 44 (16–20 cm) Core 44 (20–23.2 cm) Core 46 (0–4 cm) Core 46 (4–8 cm) Core 46 (8–12 cm) Core 46 (12–16 cm) nC15 — — — — 7.81 — — nC16 — — 19.4 9.92 52.0 — 2.11 nC17 34.9 — 80.2 58.7 129 — 22.8 Pristane 22.9 — 40.7 60.4 80.9 — 13.6 nC18 70.0 — 105 53.4 109 2.11 49.5 Phytane 24.5 — 31.4 20.9 36.5 — 18.1 nC19 64.9 — 78.5 56.2 71.2 10.8 49.3 nC20 62.9 21.7 67.5 61.3 59.7 21.4 47.9 nC21 76.1 50.9 78.2 73.0 64.6 39.1 56.4 nC22 79.8 68.7 85.7 73.9 63.0 49.5 57.0 nC23 125 116 126 98.8 87.7 73.2 76.5 nC24 92.1 87.7 94.2 67.7 74.5 64.4 66.1 nC25 190 187 180 123 109 101 97.5 nC26 102 99.4 94.8 75.2 65.3 64.3 58.3 nC27 340 337 318 183 172 170 158 nC28 138 134 128 84.0 73.5 74.7 63.2 Squalene 198 27.6 74.0 430 380 78.7 41.9 nC29 676 678 645 335 322 340 314 nC30 143 137 130 86.7 82.5 79.3 68.4 nC31 795 816 805 444 424 449 407 nC32 112 116 107 52.0 52.0 76.5 65.6 Diplop. 101 109 64.3 533 325 403 323 nC33 374 387 368 216 203 194 185 nC34 22.8 24.3 22.9 19.2 14.6 13.2 12.1 nC35 110 115 109 90.0 86.6 81.6 73.7 UCMa ₃₂₀₀ ₁₆₁₀ ₃₁₀₀ _— ₆₇₀ ₁₃₃₀ ₉₇₃ TOCb _0.70 _0.70 _0.69 _0.68 _0.63 _0.64 _0.61 SHC 3950 3510 3850 3310 3150 2390 2330 SHC/TOCc 5640 5010 5580 4870 5000 3730 3820 Terr/TOCd 2590 2620 2560 1410 1460 1500 1440 TARe 18.1 — 11.1 8.37 4.41 88.8 12.2 nC31/Terr f 0.439 0.446 0.455 0.462 0.462 0.468 0.463 CPIg 4.32 4.45 4.49 3.83 3.91 3.78 3.98 Di/SC25–33h 0.043 0.045 0.028 0.41 0.26 0.32 0.28 Di/TOCi 144 156 93.2 784 516 630 530

Compound Core 46 (16–20 cm) Core 46 (20–24 cm) Core 46 (24–26.5 cm)

nC15 — — — nC16 — — — nC17 38.4 9.27 2.88 Pristane 25.8 4.63 1.44 nC18 64.5 33.8 22.9 Phytane 24.1 12.2 8.33 nC19 54.4 42.4 41.6 nC20 45.3 44.9 54.0 nC21 54.0 55.7 67.1

(8)

predominant contribution from higher plant sources (Rieley et al., 1991; Hedges and Prahl, 1993) although shorter-chain n-alkanes were more prone to degradation than longer-chain ones (Meyers et al., 1984; Gagosian and Peltzer, 1986). The odd-carbon predominance in the C25–C33 range was

strongest among the coastal marine sediments surrounding Taiwan. In addition, most samples showed a small baseline elevation, the so-called ‘‘unresolved complex mixtures’’ (UCM). UCM is known to consist of cyclic and branched alkanes and to resist microbial degradation more effectively than n-alkanes, and thus has a greater tendency to remain in the environment after n-alkanes have degraded (Gough and Rowland, 1990;Bouloubassi

and Saliot, 1993). This does not necessarily reﬂect the content of degraded petroleum in the sediments since in some cases concentrations of UCM o10 mg g1

are common in coastal marine environ-ments far from petrogenic hydrocarbon sources (Matsumoto, 1983; Tolosa et al., 1996). Only one core section (bottom of station 20) gave a UCM value 410 mg g1. In addition, n-alkanes were prominent, indicating little weathering.

For comparing all sections in a core, total hydrocarbon concentrations were normalized to TOC for eliminating the grain size effect. As indicated in Table 2, cores 20, 21, 36, 42, 43, 44 and 46, respectively, gave the following averages and standard deviations: 63607951 107

g (gC)1

Table 1 (continued )

Compound Core 46 (16–20 cm) Core 46 (20–24 cm) Core 46 (24–26.5 cm)

nC22 51.4 56.5 62.8 nC23 72.1 74.6 77.2 nC24 65.1 64.2 66.1 nC25 97.3 96.8 93.2 nC26 53.7 59.0 53.8 nC27 165 157 149 nC28 61.5 69.6 60.1 Squalene 43.0 126 42.7 nC29 340 308 293 nC30 72.8 65.8 65.0 nC31 463 425 392 nC32 79.5 59.9 68.2 Diplop. 430 253 306 nC33 203 180 170 nC34 11.4 13.0 11.6 nC35 76.9 63.5 66.3 UCMa 1070 1910 2740 TOCb 0.63 0.58 0.63 SHC 2590 2270 2180 SHC/TOCc 4110 3910 3460 Terr/TOCd ₁₅₄₀ ₁₅₃₀ ₁₃₂₀ TARe _10.4 _17.2 _18.8 nC31/Terrf 0.478 0.478 0.470 CPIg _4.18 _4.02 _3.87 Di/SC25–33h 0.34 0.22 0.28 Di/TOCi ₆₈₃ ₄₃₆ ₄₈₆ a

UCM ¼ unresolved complex mixtures.

b

TOC in unit of gC (100 g)1.

c

In unit of 107g (gC)1.

d

Terr/TOC ¼ (nC27+nC29+nC31)/TOC, in unit of 107g (gC)1. e

TAR ¼ terrigenous/aquatic ratio ¼ (nC27+nC29+nC31)/(nC15+nC17+nC19). f_nC

31/Terr ¼ nC31/(nC27+nC29+nC31). g_{CPI ¼ carbon preference index ¼}1

2 nC25þnC27þnC29þnC31þnC33nC24þnC26þnC28þnC30þnC32þnC25þnC27þnC29þnC31þnC33nC26þnC28þnC30þnC32þnC34

.

h_Di/SC

25–33¼diploptene/(nC25+nC27+nC29+nC31+nC33). i_{Di/TOC ¼ diploptene/TOC, in unit of 10}7_{g (gC)}1_.

(9)

(C.V. ¼ 15.0%), 568671119 107g (gC)1(C.V. ¼ 19.7%), 88437834 107g (gC)1 (C.V. ¼ 9.4%), 42787384 107g (gC)1 (C.V. ¼ 9.0%), 34657 428 107g (gC)1 (C.V. ¼ 12.4%), 52207564 107g (gC)1 (C.V. ¼ 10.8%) and 41297586 107g (gC)1(C.V. ¼ 14.2%). Total aliphatic

hydro-carbon concentrations normalized to TOC (SHC/ TOC) among cores spanned a wide range. Referring toFig. 1, high averages of SHC/TOC were found at stations 20, 21 and 36, closer to the shore; low averages were found at stations 42, 43, 44 and 46, farther from the shore. This result reﬂects that

Table 2

Averages and coefﬁcients of variance for parameters

Core Parameter Average71s.d. (%) Coefﬁcient of variance (C.V.)

20 SHC/TOCa 63607951 15.0 (nC27+nC29+nC31)/TOC b 33337536 16.1 TAR 45.9736.8 80.2 nC31/(nC27+nC29+nC31) 0.43270.009 2.1 CPI 5.1770.48 9.3 Di/SC25–33 0.03070.049 163 21 SHC/TOC 5686₇₁₁₁₉ 19.7 (nC27+nC29+nC31)/TOC 16807130 7.7 TAR 8.74711.0 126 nC31/(nC27+nC29+nC31) 0.44170.015 3.4 CPI 3.0670.44 14.4 Di/SC25–33 0.2670.01 3.8 36 SHC/TOC 88437834 9.4 (nC27+nC29+nC31)/TOC 46507549 11.8 TAR 19.475.3 27.3 nC31/(nC27+nC29+nC31) 0.47570.007 1.5 CPI 5.5170.22 4.0 Di/SC25–33 0.01570.004 26.7 42 SHC/TOC 42787384 9.0 (nC27+nC29+nC31)/TOC 18957329 17.4 TAR 41.6746.3 111 nC31/(nC27+nC29+nC31) 0.44070.009 2.0 CPI 3.91_70.13 3.3 Di/SC25–33 0.05570.005 9.1 43 SHC/TOC 34657428 12.4 (nC27+nC29+nC31)/TOC 1375768 4.9 TAR 14.775.7 38.8 nC31/(nC27+nC29+nC31) 0.46870.008 1.7 CPI 3.7570.07 1.9 Di/SC25–33 0.1370.03 23.1 44 SHC/TOC 52207564 10.8 (nC27+nC29+nC31)/TOC 23807476 20.0 TAR 20.2717.8 88.1 nC31/(nC27+nC29+nC31) 0.44770.009 2.0 CPI 4.2470.44 10.4 Di/SC25–33 0.08270.069 84.1 46 SHC/TOC 41297586 14.2 (nC27+nC29+nC31)/TOC 1457777 5.3 TAR 22.9_729.5 129 nC31/(nC27+nC29+nC31) 0.46970.007 1.5 CPI 3.9470.13 3.3 Di/SC25–33 0.3070.06 20.0 a,b In unit of 107g (gC)1.

(10)

stations close to land receive more contribution from higher plants since HMW n-alkanes are the predominant components in the sediments. Hydro-carbon concentration profiles in a core can be affected by hydrocarbon degradation after burial and hydrocarbon input from the water column to sediment. Compared with other compound classes, aliphatic hydrocarbons especially n-alkanes are relatively unreactive and unlikely to be affected by any mechanism that would be selective for alter-nating homologs in the sequence (Hedges and Prahl, 1993). This means that aliphatic hydrocarbon concentrations are expected to remain virtually constant with core depth as long as input concen-trations do not change. To obtain a high precision for lipid analysis, on-column injection in GC has been used routinely in this lab for the last 15 years. The overall measurement imprecision (i.e., reprodu-cibility—from the weighing of sediment samples to the final GC determination) is expressed as the percent coefficient of variation (%C.V. or 1 standard deviation expressed as percentage of the mean concentration). Analytical imprecision varied nonlinearly over a range of analyte concentrations and generally gave a low % C.V. for high analyte concentrations (e.g. hundreds of ng g1) and a high %C.V. for low analyte concentrations (e.g., multi-ng g1). If hydrocarbon inputs to the sediment remain constant over time, total hydrocarbon concentrations in a core are estimated to have an overall measurement imprecision of 8% at most, considering the worst case. In the present result, great variations (C.V. range 9.0–19.7%) in the ratios of SHC/TOC for the 7 cores exceeded the estimated imprecision (8.4% calculated from hydro-carbon imprecision 8% and TOC imprecision 2.5%), suggesting that hydrocarbon inputs to each core varied with time. It is noted that station 21 closest to the LR exhibited a relatively high C.V. of 19.7%, partially attributable to episodic high sediment discharges of the river such as heavy rainfall after typhoons make landfall (Hsu et al., 2004; Lee et al., 2004). Relatively high C.V.s for SHC/TOC values were found for stations 20 (15%) and 21 (19.7), and low C.V.s were found for stations 36 (9.4%), 42 (9.0%), 43 (12.4%), 44 (10.8) and 46 (14.2%) (Table 2). This result suggests that stations 20 and 21 received irregular inputs from the LR and filament and that station 36 did not, probably because of the northward flow of the KC.

Higher plant n-alkanes were the predominant components in the sediments from the SOT; they

were at the concentration of hundreds of ng g1 with a low measurement imprecision of 2%. To correct for the grain size effect, (nC27+nC29+nC31)/TOC was used to estimate the

input of higher plant n-alkanes to SOT sediments. In general, (nC27+nC29+nC31)/TOC ratios in

Table 2were higher at stations close to shore than those away from shore with the exception of station 21. The low ratio at station 21 is considered to be inﬂuenced by the direct input from the LR, which is known to contain relatively low proportions of terrigenous sources (Jeng and Huh, 2006). Com-pared with the calculated imprecision of (nC27+nC29+nC31)/TOC (3.2%), the 7 cores

ex-hibited great variations (C.V. range 4.9–20.0%), attributable to varied inputs from higher plant sources. Stations 20, 42 and 44 with relatively high C.V.s of 16.1%, 17.4% and 20.0%, respectively, are located in the north of the study area and approximately along the path where the KC turns. This could be explained as a result of episodic inputs of HMW n-alkanes from the ﬁlament and LR to the SOT as well as deposition of these alkanes along the path of the KC.

The ratio of terrigenous-to-aquatic n-alkanes were estimated as terrigenous/aquatic ratio (TAR) ¼ (nC27+nC29+nC31)/(nC15+nC17+nC19).

This ratio is valuable for determining changes in relative contributions of organic matter from land and aquatic ﬂora although it may over-represent the absolute amounts from terrigenous sources (Meyers, 1997). This ratio was adopted in the present study to estimate the relative contribution of terrigenous and marine sources to the cored sediments. Stations 20 and 42 showed compara-tively high TAR values (45.9 and 41.6, respeccompara-tively) among the 7 cores, which could be caused by input of sediment with high TAR values from the ECS shelf known to contain high proportions of relict sediments (Niino and Emery, 1961; Chen et al., 1995). The relict sediments presumably have high TAR ratios based on the fact that aquatic n-alkanes degrade faster than terrigenous n-alkanes. Based on the calculated imprecision of TAR, 8.2% (nC27+nC29+nC31 imprecision 2% and

nC15+nC17+nC19 imprecision 8%), far greater

variations of TAR (C.V. range 27.3–129%) were found for all cores, probably attributing to varied inputs of terrigenous and marine hydrocarbon sources. Again a minimum C.V. was found for station 36, indicating that the input of hydrocarbon sources was relatively stable.

(11)

In the HMW region, the C27 and C29 n-alkanes

are diagnostic of waxes from trees and shrubs, and the n-C31 alkane is diagnostic of grass waxes. The

nC31/(nC27+nC29+nC31) ratio was employed to see

if any difference existed among the 7 cores. As listed in Table 2, low ratios were found at stations 20 (0.432), 21 (0.441), 42 (0.440), and 44 (0.447); high ratios were found at stations 36 (0.475), 43 (0.468) and 46 (0.469). The results might reveal that the sediments with high ratios may receive more contribution from grass waxes than from tree and shrub waxes. Alternatively, the low ratios may be caused by the input from the LR, which has n-alkane distribution maximizing at nC29 (7 out of

9 sediments,Jeng and Huh, 2006). It is quite likely that suspended sediment with low nC31/

(nC27+nC29+nC31) from the river enters the sea,

merges into the KC, and ﬂows along the continental slope of the ECS, resulting in those stations with low ratios (Fig. 1). All stations but station 21 showed constant nC31/(nC27+nC29+nC31) ratios in

a core, judging from the result that the C.V. of each core was smaller than the calculated imprecision of 2.8 (measurement imprecisions of C27, C29and C31,

2%). The result appears to mean that only station 21 is inﬂuenced. Note that the nC31/

(nC27+nC29+nC31) ratio is solely sensitive in

responding to the great change of C31 n-alkane

since it is the most abundant component of the aliphatic hydrocarbons.

Owing to the CPIs of terrestrial higher plant waxes yielding high values, usually 5–10, higher CPI values found in sediment or soil show greater contribution from vascular plants (Rieley et al., 1991;Hedges and Prahl, 1993); CPI values close to unity are thought to indicate greater input from microorganisms, recycled organic matter, and/or petroleum (Bray and Evans, 1961; Farrington and Tripp, 1977; Kennicutt et al., 1987). In the present study, the average CPIs ranged from 3.06 (core 21) to 5.51 (core 36) (Table 2). This reﬂects that most marine sediments are relatively free of contamina-tion of petrogenic hydrocarbons based on CPI values o3 indicating oiled sediments (Farrington and Tripp, 1977) and no known oil pollution in the area for a long period before sampling. It should be noted that lower CPI values of 2.81, 2.64 and 2.84 (Table 1) were found in the upper 3 sections of station 21; this could be partially attributed to a greater inﬂuence from the nearby LR, which exported suspended matter derived from relatively old carbon—the bedrock of argillite-slate and

meta-sandstone—due to massive road construction in 1975–1980 (Kao and Liu, 1996; Jeng and Kao, 2002). On the other hand, the lower 2 sections exhibited higher CPI values of 3.28 and 3.73 (Table 2), presumably representing the values before humans disturbed the river.

Referring toTable 2, variations (mean71 s.d.) of

CPIs in the 7 cores are as follows: core 20, 5.1770.48 (C.V. ¼ 9.3%); core 21, 3.0670.44 (C.V. ¼ 14.4%); core 36, 5.5170.22 (C.V. ¼ 4.0%); core 42, 3.9170.13 (C.V. ¼ 3.3%); core 43, 3.7570.07 (C.V. ¼ 1.9%); core 44, 4.2470.44 (C.V. ¼ 10.4%); 3.9470.13 (C.V. ¼ 3.3%). It should be mentioned that core 21 showed the largest C.V. for CPI, which could be partially attributed to receiving more input of lower CPIs from the LR, resulting in large change in CPI. All 7 cores studied were relatively short (9.5–26.5 cm) and were collected mainly from the Trough with high sedimentation. Little or no diagenetic change in the concentration of aliphatic hydrocarbons is expected. Moreover, the CPI is a diagenetically insensitive ratio because it is based on the relative abundances of chemically similar compounds within a limited molecular weight range (Hedges and Prahl, 1993). CPI’s of all cores fluctuated with the core depth. This result indicates that n-alkane inputs to the study area vary with time. C.V.s for CPI are relatively high for stations 20 (9.3%), 21 (14.4%), and 44 (10.4%) and comparatively low for stations 36 (4.0%), 42 (3.3%), 43 (1.9%) and 46 (3.3%). It appears that stations with high C.V.s are located close to the LR and filament and hence are under the direct influence of them. It is interesting that station 21 showed the highest C.V. due to the direct input from the LR exporting old hydrocarbons.

Diploptene (hop-22(29)-ene) is derived from terrestrial higher plants and is also formed by bacteria (Rohmer et al., 1984). Prahl (1985)

observes an offshore increase in the ratio of diploptene to combined C25, C27, C29, C31, C33

plant wax n-alkanes (Di/SC2533) in sediments

throughout the southern Washington continental shelf and slope region, and the ratio is used to compare marine sediments and soils (Prahl et al., 1992). As displayed inTable 2, high average ratios of Di/SC2533were found for cores 21 (0.2670.01,

C.V. ¼ 3.8%), 43 (0.1370.03, C.V. ¼ 23.1%) and 46 (0.3070.06, C.V. ¼ 20.0%), and low average ratios were found for cores 20 (0.03070.049, C.V. ¼ 163%), 36 (0.01570.004, C.V. ¼ 26.7%), 42 (0.05570.005, C.V. ¼ 9.1%), and 44 (0.0827

(12)

0.069, C.V. ¼ 84.1%). Compared with the average Di/SC2533 ratio (0.1670.05, C.V. ¼ 32%)

(calcu-lated from Jeng and Huh, 2004) of 9 surface sediments from the ECS covering a relatively larger area (41 latitude 1.51 longitude), the present Di/ SC2533 ratios have a wider range for a relatively

smaller area (11 latitude 500_{longitude). The result}

may suggest that the input of diploptene and higher plant n-alkanes to the study area is more complex than that to the ECS. Moreover, a comparison with the nearby LR with high sediment discharge is worthwhile. The average Di/SC2533 ratio of 7

surface sediments from this river is 0.3070.14 (C.V. ¼ 47%) (calculated from Jeng and Huh, 2006), which is comparable to those of cores 21, 43 and 46. The river samples cover a large area with different environments, but for a core site each section is supposedly deposited in the same sedi-mentary environment. Therefore, the C.V.s of these three cores are considered fairly large, implying variable inputs. Besides, the average Di/SC2533

ratio of river samples is closer to those of cores 21 and 46 (which and river mouth lie on a straight line), likely implying that the river might be a signiﬁcant contributor to the two stations. The large differences in Di/SC2533 ratios between cores can

be attributed to (1) varied input of diploptene from higher plant sources and (2) possible input of bacterial diploptene to the study area based on the fact that submarine hydrothermal activities have been observed (Lee et al., 1998;Hsu et al., 2003). If higher plants are the sole source of diploptene, a strong correlation between diploptene and terres-trial higher plant n-alkanes has been demonstrated (Prahl et al., 1992). For instance, positive correla-tions between diploptene and higher plant n-alkanes were found for cores 21 (SC2533¼ 215+4.39 Di,

r ¼ 0.97), 42 (SC2533¼1.02+18.6 Di, r ¼ 0.94)

and 46 (SC2533¼980+0.628 Di, r ¼ 0.82).

How-ever, no such correlation was found for cores 20, 36, 43 and 44; contributions of diploptene mainly from bacteria were unlikely because series of long-chain odd-predominant n-alkanes were typically absent (Kolattukudy, 1976) and because CPI values of the 4 cores were high (Table 2). Venkatesan (1988)

reviews reports of the occurrence of diploptene in coastal marine sediments on a very broad geo-graphic basis and concludes that this compound derives largely from autochthonous sources. It is thought that diploptene in the 4 cores might be generated in situ analogous to that of perylene (Silliman et al., 2000). If this is the case, an

increasing trend for the diploptene/TOC ratio would be expected. However, no such trend was observed for the 4 cores probably because they were too short for diagenetically produced diploptene to be clearly observed.

In summary, the 7 sediment cores have been found to show great variations in SHC/TOC, (nC27+nC29+nC31)/TOC, TAR and Di/SC2533

ratios as well as CPI values. The results can be attributed to hydrocarbon inputs to the study area fluctuating over time; this may be caused by complex current flow and sediment transport, resulting in a complicated environment for lipid deposition. This is supported by the observation from sediment traps that sediment fluxes to the SOT display great temporal and spatial variability (Hsu et al., 2004).

Acknowledgments

My thanks go to the captain, crew and techni-cians of the R/V Ocean Researcher I for help with sediment collection. I am grateful to Prof. Philip A Meyers (University of Michigan) and one anon-ymous reviewer for constructive comments and suggestions. This study was ﬁnancially supported by the National Science Council, Republic of China, Grant no. NSC93-2611-M-002-016.

References

Bouloubassi, I., Saliot, A., 1993. Investigation of anthropogenic and natural organic inputs in estuarine sediments using hydrocarbon markers (NAH, LAB, PAH). Oceanologica Acta 16, 145–161.

Bray, E.E., Evans, E.D., 1961. Distribution of n-parafﬁns as a clue to recognition of source beds. Geochimca et Cosmochi-mica Acta 22, 2–15.

Chen, M.P., Lin, K.L., Huang, C.K., 1995. Sedimentary structure and texture of surﬁcial sediment off northeastern Taiwan. Ti-Chih (ROC) 15, 15–47 (in Chinese with English abstract).

Chern, C.S., Wang, J., Wang, D.P., 1990. The exchange of Kuroshio and East China Sea shelf water. Journal of Geophysical Research 95, 16017–16023.

Chung, Y., Chang, W.C., 1995. Pb-210 ﬂuxes and sedimentation rates on the lower continental slope between Taiwan and the South Okinawa Trough. Continental Shelf Research 15, 149–164.

Farrington, J.W., Tripp, B.W., 1977. Hydrocarbons in western North Atlantic surface sediments. Geochimica et Cosmochi-mica Acta 41, 1627–1641.

Gagosian, R.B., Peltzer, E.T., 1986. The importance of atmo-spheric input of terrestrial organic material to deep sea sediments. Organic Geochemistry 10, 661–669.

(13)

Gough, M.A., Rowland, S.J., 1990. Characterization of unre-solved complex mixtures of hydrocarbons in petroleum. Nature 344, 648–650.

Hedges, J.I., Prahl, F.G., 1993. Early diagenesis: consequences for applications of molecular biomarkers. In: Engel, M.H., Macko, S.A. (Eds.), Organic Geochemistry: Principles and Applications. Plenum Press, New York, pp. 237–253. Hsu, S.C., Lin, F.J., Jeng, W.L., Chung, Y.C., Shaw, L.M., 2003.

Hydrothermal signatures in the southern Okinawa Trough detected by the sequential extraction of settling particles. Marine Chemistry 84, 49–66.

Hsu, S.C., Lin, F.J., Jeng, W.L., Chung, Y.C., Shaw, L.M., Hung, K.W., 2004. Observed sediment ﬂuxes in the south-westernmost Okinawa Trough enhanced by episodic events: ﬂood runoff from Taiwan rivers and large earthquakes. Deep-Sea Research I 51, 979–997.

Jeng, W.L., Kao, S.J., 2002. Lipids in suspended matter from the human-disturbed Lanyang River, northeastern Taiwan. En-vironmental Geology 43, 138–144.

Jeng, W.L., Huh, C.A., 2004. Lipids in suspended matter and sediments from the East China Sea Shelf. Organic Geochem-istry 35, 647–660.

Jeng, W.L., Huh, C.A., 2006. A comparison of sedimentary aliphatic hydrocarbon distribution between the southern Okinawa Trough and a nearby river with high sediment discharge. Estuarine, Coastal and Shelf Science 66, 217–224.

Jeng, W.L., Lin, S., Kao, S.J., 2003. Distribution of terrigenous lipids in marine sediments off northeastern Taiwan. Deep-Sea Research II 50, 1179–1201.

Kao, S.J., Liu, K.K., 1996. Particulate organic carbon export from a subtropical mountainous river (Lanyang Hsi) in Taiwan. Limnology and Oceanography 41, 1749–1757. Kao, S.J., Lin, F.J., Liu, K.K., 2003. Organic carbon and

nitrogen contents and their isotopic compositions in surﬁcial sediments from the East China Sea shelf and the southern Okinawa Trough. Deep-Sea Research II 50, 1203–1217.

Kennicutt II, M.C., Barker, C., Brooks, J.M., DeFreitas, D.A., Zhu, G.H., 1987. Selected organic matter source indicators in the Orinoco, Nile and Changjiang deltas. Organic Geochem-istry 11, 41–51.

Kolattukudy, P.E., 1976. Chemistry and Biochemistry of Natural Waxes. Elsevier, Amsterdam.

Lee, C.S., Chung, S.L., SPOT Members, 1998. Southernmost part of the Okinawa Trough (SPOT): an active extension/ collision/subduction area. EOS Transaction of American Geophysical Union 79, W109.

Lee, S.Y., Huh, C.A., Su, C.C., You, C.F., 2004. Sedimentation in the Southern Okinawa Trough: enhanced particle scaven-ging and teleconnection between the Equatorial Paciﬁc and western Paciﬁc margins. Deep-Sea Research I 51, 1769–1780.

Liu, C.T., Cheng, S.P., Chuang, W.S., Yang, Y., Lee, T.N., Johns, W.E., Li, H.W., 1998. Mean structure and transport of

Taiwan Current (Kuroshio). Acta Oceanographica Taiwanica 36, 159–176.

Matsumoto, G., 1983. Comparative study on organic constitu-ents in polluted and unpolluted inland aquatic environ-ments—V. Organic carbons and hydrocarbons in sediments. Water Research 17, 823–830.

Meyers, P.A., 1997. Organic geochemical proxies of paleoceano-graphic, paleolimnologic, and paleoclimatic processes. Or-ganic Geochemistry 27, 213–250.

Meyers, P.A., Leenheer, M.J., Eadie, B.J., Maule, S.J., 1984. Organic geochemistry of suspended and settling particulate matter in Lake Michigan. Geochimica et Cosmochimica Acta 48, 443–452.

Niino, H., Emery, K.O., 1961. Sediments of shallow portions of East China Sea and South China Sea. Geological Society of America Bulletin 72, 731–762.

ODP, 2001. Seaﬂoor Observatories and the Kuroshio Current. Ocean Drilling Program, Leg 195 Preliminary Report. Available at: /http://www-odp.tamu.edu/publications/prelim/ 195_prel/195toc.htmlS.

Prahl, F.G., 1985. Chemical evidence of differential particle dispersal in the southern Washington coastal environment. Geochimica et Cosmochimica Acta 49, 2533–2539.

Prahl, F.G., Hayes, J.M., Xie, T.M., 1992. Diploptene: an indicator of terrigenous organic carbon in Washington coastal sediments. Limnology and Oceanography 37, 1290–1300.

Rieley, G., Collier, R.J., Jones, D.M., Eglinton, G., 1991. The biogeochemistry of Ellesmere Lake, UK—I: source correla-tion of leaf wax inputs to the sedimentary lipid record. Organic Geochemistry 17, 901–912.

Rohmer, M., Bouvier-Nave, P., Ourisson, G., 1984. Distribution of hopanoid triterpenes in prokaryotes. Journal of General Microbiology 130, 1137–1150.

Silliman, J.E., Meyers, P.A., Ostrom, P.H., Ostrom, N.E., Eadie, B.J., 2000. Insights into the origin of perylene from isotopic analyses of sediments from Saanich Inlet, British Columbia. Organic Geochemistry 31, 1133–1142.

Tang, T.Y., Hsueh, Y., Yang, Y.J., Ma, J.C., 1999. Continental slope ﬂow northeast of Taiwan. Journal of Physical Oceano-graphy 29, 1353–1362.

Tolosa, I., Bayona, J.M., Albaige´s, J., 1996. Aliphatic and polycyclic aromatic hydrocarbons and sulfur/oxygen deriva-tives in northwestern Mediterranean sediments: spatial and temporal variability, ﬂuxes and budgets. Environmental Science and Technology 30, 2495–2503.

Venkatesan, M.I., 1988. Diploptene in Antarctic sediments. Geochimica et Cosmochimica Acta 52, 217–222.

Water Resources Bureau, 1998. Hydrological Year Book of Taiwan, Republic of China 1997. Ministry of Economic Affairs, Taipei, Taiwan, Republic of China, p. 23.

Wong, G.T.F., Chao, S.Y., Li, Y.H., Shiah, F.K., 2000. The Kuroshio edge exchange processes (KEEP) study—an intro-duction to hypotheses and highlights. Continental Shelf Research 20, 335–347.