Hydrobiologia 457: 17–24, 2001. © 2001 Kluwer Academic Publishers. Printed in the Netherlands.

17

Biogeochemical changes in the sediments of Lake Cantara South, a saline lake in South Australia R. L. Wang1 & W. D. Williams2 1

Biogeochemical Lab, Indiana University, Bloomington, IN 47405 and Department of Applied Science, 535A, Brookhaven National Laboratory, Upton, NY 11973, U.S.A. Email: [email protected] 2 Department of Environmental Biology, University of Adelaide, Adelaide, 5005, Australia Email: [email protected] Key words: salt lake, hypersaline/saline lakes, lake sediments, organic compounds in lake sediments, paleolimnology

Abstract Biogeochemical studies were undertaken of a 65-cm long sediment core from Lake Cantara South, South Australia. sediments had been deposited over 2000 years. Changes with sediment depth in the concentration or ratio of the following were determined: (i) total organic carbon, total carbonate (inorganic) carbon, total sulfur, total carbon, total inorganic and organic sulfur, atomic C/N, and sulfate/chloride; (ii) n-alkanes; (iii) a highly branched isoprenoid alkane, and (iv) steroids. Interpretation of the changes with sediment depth indicated the nature of changes that took place when the system changed from a protected marine lagoon to an isolated (athalassic) saline lake. This change took place about 1000 years ago.

14 C determinations indicated that the

Introduction A group of coastal athalassic saline lakes occurs in south-eastern South Australia (Fig. 1) and has been studied by several investigators (e.g., von der Borch, 1976; De Deckker & Geddes, 1980; Warren, 1988, 1990). Biological studies have focussed upon one lake, Lake Cantara South. The stratigraphy and textural composition of its sediments have been studied by Thoms & Williams (1993). To complement this study, a biogeochemical investigation of a short (65 cm) sediment core was undertaken. The results of this study form the basis of the present paper. The results also contribute to global environmental and climate change studies based on biogeochemical evidence from hypersaline lacustrine systems.

Study site and methods The core was collected from Lake Cantara South, a temporary, intermittent salt lake located ∼250 km from Adelaide, South Australia (Fig. 1). The lake is

small (144 ha or ∼1.5 km2), shallow (when water is present: Zm = 0.5 m), and is dry in summer and autumn, but regularly contains water of 0–0.5 m depth and <50 to >300 g l−1 salinity during winter and spring. Na+ and Cl− are the dominant ions (respectively, ∼80 and 90% of total ions as equivalents; Williams, 1986). The lake is presently athalassic with no direct connection to the nearby sea. It lies above Pleistocene calcrete and between Quaternary dunes running parallel to the coast. The region is characterised by slow tectonic uplift, low coastal plain gradients, and eustatic sea-level oscillations. The core, collected with a stiff plastic pipe, contained sediments to a depth of 65 cm. It was obtained in the afternoon, transported to the laboratory in Adelaide the same day, and stored in a cold room at ∼4◦ C. Subsequently, it was sent by airmail to the laboratory of the senior author in North America where it was again frozen until examination. In dividing the core, upper sections were cut at intervals of 1 cm and lower ones mostly at intervals of 2 cm. Each sample (interval) was stored in a sterile Teflon bag to minimize contamination.

18

Figure 1. Location of the study area of Lake Cantara South (LCS), South Australia (modified after Thoms & Williams, 1993).

Concentrations of total organic carbon (TOC), total carbonate carbon (TIC) and total sulfur (TS) were determined using a LECO C-S analyzer. Total carbon (TC) and TOC were determined using total sediment samples and residue after HCl treatment (0.5 N, stir, overnight), respectively. Similarly, TS and total organic sulfur were determined using total sediment samples and acid-treated residues, respectively. Core samples were freeze-dried and extracted by Soxhlet (CH2 Cl2 /CH3 OH, 1/1, 48 h). Elemental sulfur (S◦ ) was removed using activated Cu during extraction. A SiO2 column separated extracts into three fractions: F1 (aliphatic hydrocarbons, eluted with hexane), F2 (aromatic hydrocarbons, eluted with toluene (C7 H8 )), and F3 (polar fractions, eluted with CH2 Cl2 /CH3 OH, 1:1, v/v). Total nitrogen was determined by analyzing total dry sediment samples using a Perkin-Elmer elemental analyzer, chloride and sulfate, using total sediment and a Dionex Ion Chromatography facility. The saturated hydrocarbon fraction was analyzed by gas chromatogram (GC) and then by GC-MS using

a Finnigan TSQ 700 GC-MS system, Incos XL system or HP 5973 system. A fused silica capillary column coated with Ultra-1 (30 m × 0.25 mm i.d. × 0.25 µm film thickness) was programmed from 100 to 320◦C at 10◦ C min−1 , with helium as the carrier gas. Similar GC conditions were used for GC/MS analysis, with the ion source temperature at 200◦C and the emission current and electron energy set at 200 mA and 70 eV, respectively. The GC–MS systems were operated in the full scan mode and scanned from m/z 50 to 650 per second. Absolute ages were determined by 14 C dating of organic carbon on three samples using a Wallac accelerator mass spectrophotometer (AMS).

Results and discussion Elemental composition Figure 2 shows the concentration of TOC, TIC and

19

Figure 2. Depth profiles of TOC, TIC, TS and atomic C/N ratio in core LCS-1 from Lake Cantara South, South Australia. Age interpolation based on 14 C data from three samples (see text). The age data suggest an average sedimentation rate of ∼0.7 mm yr−1 . I, II and III, sections of core. Note that the three sections show significant variation in the sediment accumulation rate of organic and carbonate carbon.

TS as percentages and the carbon/nitrogen ratio at different sediment depths. Details for individual samples are given in Table 1. Based on these data, the core falls into three sections: a bottom (III), middle (II), and top (I) section. In section II, TOC gradually increases from about 0.5% to ∼2.0% toward sections I and III. TOC in lake sediments has been used as an indicator of levels of primary production and accumulation. Its profile contrasts with that of TIC; in section III, carbonate percentages (TIC) are low, then increase (>8%), are greatest (∼10%) near 45–25 cm, and finally decrease toward the surface. TS is rather constant at all levels except in the topmost sediments where the concentration is >1%. Sulfate/chloride ratios (SSO4 /Cl; Fig. 3) peak near 45–30 and 15 cm. Both the bottom and top sections are characterized by low ratios. This variation indicates significant environmental changes, particularly in past salinities. The higher ratio in the core’s middle section indicates that salinity was then much lower, perhaps

due to the input of fresh water or hyposaline marine water. In summary, the elemental composition of the sediments suggests that during the period ∼800 – 1600 a BP, Lake Cantara South was dominated by a sulfate/carbonate type of sedimentary environment, and had low organic accumulation/production. From 800 a BP to the present, TOC increased, TIC decreased, and the C/N atomic ratio increased. The sulfate/chloride ratios indicate lower salinities occurred from ∼1600–400 a BP than at present or before 1600 a BP. Molecular characteristics of organic matter n-Alkanes Figure 4 shows three examples of reconstructed ion chromatograms which illustrate the composition of saturated hydrocarbons. They are from three samples (2, 15 and 19) from sections I, II and III, respectively. n-Alkanes in the sediment core are generally composed of nC17 –nC32 , mostly with an elevated odd-

20 Table 1. Elemental and molecular geochemical parameters of sediments from Lake Cantara South, South Australia Sample

Depth (cm)

TOC TOS TC (%) (%) (%)

TS (%)

TIC (%)

TIS (%)

N (%)

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

2.24 4.48 6.72 8.96 11.20 13.44 15.68 17.92 20.16 22.40 26.88 31.36 33.60 38.08 42.56 47.04 51.52 56.00 60.48 64.96

2.1 0.8 1.7 0.5 0.6 0.5 0.3 0.3 0.4 0.3 0.2 0.3 0.4 0.7 0.4 0.5 0.9 0.8 1.3 1.7

1.1 0.23 0.08 0.45 0.4 0.33 0.44 0.26 0.51 0.31 0.49 0.53 0.31 0.37 0.54 0.31 0.48 0.64 0.37 0.37

5.4 6.9 5.7 7.0 7.3 7.5 8.2 8.3 9.0 8.4 9.2 8.2 8.5 8.4 9.6 8.5 8.8 8.6 8.2 7.3

0.93 0.11 0 0.33 0.27 0.22 0.33 0.15 0.38 0.22 0.39 0.45 0.23 0.27 0.45 0.22 0.37 0.51 0.17 0.23

0.36 6.8 0.27 3.2 0.35 5.5 0.17 3.7 0.62 1.2 0.51 1.2 0.60 0.7 0.48 0.7 0.33 1.3 0.45 0.9 0.77 0.3 0.65 0.5 0.98 0.5 1.14 0.7 0.28 1.8 0.35 1.6 0.39 2.6 0.15 6.3 0.26 6 0.25 7.7

0.17 0.12 0.09 0.12 0.13 0.11 0.11 0.11 0.12 0.09 0.10 0.08 0.09 0.11 0.09 0.09 0.11 0.13 0.21 0.14

7.5 7.6 7.3 7.5 7.9 8.0 8.5 8.6 9.4 8.7 9.4 8.5 8.9 9.1 10.0 8.9 9.7 9.4 9.5 8.9

Corg / Cino / EOM C20 HB/ NAR NAR (%) n23 24.4 33 24.4 51.6 14.9 18.4 16.6 20.8 33.2 22.7 14.2 15.2 10.6 9.3 41.8 29.8 28.9 73.3 42.7 41.7

7.4 4 4.3 3.9 2.2 2 3 2.2 2.3 2.8 2.6 2.6 2 1.9 2.2 1.7 3 3.2 3.3 3.3

0.8 0.7 0.1 0.1 0.2 0.1 0.4 0.2 0.5 0.4 0.3 2.3 2.4 4.5 6.8 4.1 5.5 4.6 0.8 3.7

Corg HAR

OEP St St C27 % C28 %

St C29 %

5.9 5 11.1 7.2 3.8 5.2 2.7 3.7 3 7.1 1.4 3.4 3.3 3 1.9 4 3.5 2.5 4.4 8.3

5.9 5 11 7.2 3.8 5.2 2.7 3.5 3 7.1 1.4 3.4 3.3 3.3 1.9 4 3.5 2.5 4.4 8.3

33 36 25 41 45 41 37 47 36 30 47 38 44 38 44 59 65 27 25 26

49 44 45 43 42 43 46 40 45 40 38 34 33 33 24 21 20 57 63 64

18 19 30 15 13 16 17 13 19 29 15 28 23 30 31 20 14 16 12 10

TOC, total organic carbon; TOS, total organic sulfur; TC, total carbon; TIC, total inorganic carbon; TIS, total inorganic sulfur; N, nitrogen; Corg /N AR, organic carbon/nitrogen atomic ratio; Cino /N AR, inorganic carbon/nitrogen atomic ratio; EOM, extractable organic matter; C20 HB/n23, C20 highly branched isoprenoid hydrocarbon /n-C23 alkane ratio; Corg /H AR, organic carbon/hydrogen atomic ratio; OEP, odd-to-even predominance; St C27 , St C28 , St C29 , C27 –C29 steroid percentage.

carbon predominance (OEP), expressed as: OEP = 3(n27 + n29)/2(n26 + n28 + n30), where n27–n30 represent the peak area of normal alkanes, respectively. The specific carbon range used in this parameter depends on the dominant peaks in the samples, varying from C25 to C29 . A particularly strong odd-carbon predominance occurs in the upper (I, OEP up to 11) and bottom (III, OEP up to 8.5) sections of the core (Fig. 3 and Table 1). Section II has significantly lower OEP values (1–3). A high OEP distribution of n-alkanes indicates that a direct biological source is the dominant input of these n-alkanes (e.g., Meyers & Ishiwatari, 1993). The greater fluctuations in OEP in section I probably reflect increased fluctuations in recent salinities and corresponding biological effects. Note that in the n-alkane series, nC23 is commonly the dominant component. The dominance of nC23 , rather than long-chain homologues (e.g., nC29 ) at all levels in the core, indicates that autochthonous sources, i.e., macroalgae, seagrasses and phytoplankton, are the most important sources. Higher plants

in the lake’s catchment would have deposited organic matter dominated by longer chain compounds (Cranwell, 1973; Meyers & Ishiwatari, 1993). In section II, OEP values dropped to ∼1.5–1.8, indicating smaller inputs from higher plants during the deposition of this section of the core. Highly branched isoprenoid alkane A significant alkane eluted between the positions of isoC19 (pristane) and nC17 (Fig. 4). In many samples (e.g., 14–17; Figure 3), this alkane comprized up to 60% of the total hydrocarbon fraction in the organic extract. Gas chromatograph–mass spectrometric data showed that it had a molecular ion at m/z 282 (C20 H42 ), and three major characteristic ions at m/z 168, 197 and 211 (Fig. 5). According to GC retention time and mass spectra, this alkane is tentatively identified as 2,6,10-trimethyl-7(3-methyl-butyl)dodecane (C20 highly branched isoprenoid or C20 HBI, hereafter). Also shown in Fig. 5 is the possible fragmentation of major ions. To determine the quantitative distribution of C20 HBI, its molecular ratio with nC23 alkane, i.e., C20 HBI/nC23 , was calculated and plot-

21

Figure 3. Depth profiles of molecular and elemental composition of sediments from core LCS-1 from Lake Cantara South, South Australia. Panel 1, ratio of C20 highly branched isoprenoid alkane/n-C20 alkane; panel 2, C20 HBI concentration in dry sediment; panel 3, sulfate/chloride ratio of dry sediments; panel 4, odd-over-even predominance of normal alkanes; panel 5, ratio of C29 /(C27 + C28 + C29 ) sterenes. I, II and III, sections of sediment core. a. B.P., years before present.

ted against core depth (Fig. 3). The highest ratio occurred at ∼42.5 cm. In this section of the core, C20 HBI is the major peak in the whole chromatogram with a ratio approaching 7 (Table 1). Its relative abundance dropped markedly at ∼35 cm and upward (Fig. 3). The depth profile of C20 HBI/nC23 is similar to that of the sulfate/chloride ratio (Fig. 3), suggesting a relationship between this alkane and sulfate deposition. Highly branched isoprenoids, commonly C20 – C25 , have previously been recognized as potential hypersaline biological markers. They have been found in various marine or lacustrine hypersaline settings (Robson & Rowland, 1988), including Shark Bay, Australia (a carbonate hypersaline environment with C21 , C22 , Dunlop & Jeffries, 1985), Rozel Point, USA, a playa deposit of the Miocene Salt-Lake group (Meissner et al., 1984), and hypersaline carbonate lagoon and sabkha environments at Abu Dhabi, United Arab Emirates (C20 –C22 , Kenig et al., 1990). Abundant C20 HBI has also been found in coastal surface sediments (Gearing et al., 1976) and identified as

7-isopranyfarnenes (Robson, 1987; Rowland et al., 1985; Yon et al., 1982). In many reports, highly branched alkanes and alkenes have been noted as the most abundant hydrocarbons. This group of organic compounds probably originated from algae, as suggested by Rowland et al. (1985), who found that Enteromorpha prolifera from low tide sand banks at Sandyhaven, Wales, contained 2,6,10-trimethyl-7(3-methylbutyl)dodecane and a related monoene and pseudohomologue C25 diene. However, in modern lagoon and sabkha environments at Abu Dhabi, abundant HBI was found associated with microbial mats and seagrass (Kenig et al., 1990). As well as hydrocarbon HBI, S-bound species, including C20 , C21 and C25 HBI, have recently been found in a Messinian evaporitic sequence in Vena del Gesso, Italy (Kenig et al., 1995). This occurrence is possibly due to the incorporation of S during sedimentary diagenesis. Based on the enrichment of 13 C in these compounds, it has been concluded that the carbon skeleton of S-bound HBI also originates from

22

Figure 4. Reconstructed ion chromatograms of saturated hydrocarbons from three sediment samples from Lake Cantara South, South Australia. Upper panel, sample from section I (sample LCS-2); central panel, sample from section II (sample LCS-15); lower panel, sample from section III (sample LCS-19). Numbers on peaks indicate carbon numbers of n-alkanes; C20 HBI, 2, 6,10-trimethyl-7-(3-methyl-butyl) dodecane; OEP, odd-over-even predominance (see text for details); H/23, C20 HBI/nC23 alkane.

algal compounds (e.g., Nichols et al., 1988; Volkman et al., 1994). In some sediments, alkene concentrations appear to be greatest a few centimeters below the surface (Requejo et al., 1984), suggesting that they may be in situ bacterial products. Nevertheless, although the only confirmed occurrence of 7-isopranylfarnesenes

so far is in Enteromorpha, algae seem to be the most likely source of the high concentrations of C20 HBI in hypersaline environments. So far as Lake Cantara South is concerned, the high concentrations of C20 HBI in its sediments indicate that macroalgae and/or diatoms were the major source of sedimentary organic matter from about 1600–1300

23

Figure 5. Mass spectrum and the possible structure of 2,6,10-trimethyl-7-(3-methyl-butyl) dodecane (peak HBI in Figure 4) (El, 70 eV).

a BP. In the topmost sediments (from ∼35cm to the surface), C20 HBI concentrations fell significantly, suggesting the development of an isolated, supratidal lake environment characterized by lower concentrations of sulfate and higher ones of chloride. Low concentrations of C20 HBI also occurred below 60 cm (Fig. 3). Steroids Steroid hydrocarbons in the LCS sediments are mainly unsaturated sterenes, characterized by the dominance of m/z 215 fragmental ions. Their survival provides strong evidence of the preservative powers of saline lake sediments for organic deposits. Percentage profiles of C29 /(C27 + C 28 + C29 ) sterenes, calculated on the basis of m/z 215 ion chromatograms versus sediment burial depth (cm), are illustrated in Fig. 3. Between about 55 and 35 cm, the sterenes are dominated by C29 homologues, with C27 and C28 moieties relatively lower. In sediments above and below these depths, C27 and C28 are significantly higher and C29 lower. The origin of C29 steroids was thought to be mainly from higher plants (Huang & Meinshein, 1979). However, subsequent studies have shown that certain algae, e.g diatoms, may have abundant C29 steroids (Volkman, 1986; Volkman et al., 1994). C27 derives mainly

from marine phytoplankton or zooplankton. Thus, the steroids in the sediments of section II may derive from higher plants or diatom lipids (which contain more C29 sterenes). Note in this connection that C27 steroids are less affected by biodegradation than C28 and C29 homologues (Seifert & Moldowan, 1981) if microbial activity is important - after regression and aeration of sediments for example.

Conclusions Our biogeochemical studies of the sediments of Lake Cantara South do not provide an unequivocal paleolimnological history of the lake. Even so, they do broadly accord with previous paleolimnological studies which indicate that the system arose first as a protected marine embayment, became a restricted marine lagoon, and finally developed into an athalassic lake. The period covered by the core (0–2000 a BP) relates to the time when the system changed from a restricted marine lagoon to an athalassic lake. Isolation appears to have taken place about 1000 a BP. Once isolated, the lake became more saline, intermittently dried, biological production/accumulation increased, and organic deposits came largely from in situ plants (macroalgae, seagrasses, phytoplankton). Before isolation, the

24 system was less saline, permanently contained water, had lower biological production/accumulation, and a sulfate/carbonate depositional environment dominated.

Acknowledgments Sincere thanks go to Prof. Simon Brassell (Indiana University) for support, encouragement and expert help. We also thank Prof. Stuart Hurlbert (San Diego State University) for help in the field. Finally, we thank Dr J. Volkman (CSIRO, Australia) and Dr E. V. de Santos Neto (PETROBRAS, Brazil) for critical comments on a draft manuscript.

References Cranwell, P. A., 1973. Chain-length distribution of n-alkanes from lake sediments in relation to post-glacial environmental change. Fresh. Biol. 3: 259–265. De Deckker, P. & M. C. Geddes, 1980. The seasonal fauna of saline lakes near the Coorong Lagoon, South Australia. Aust. J. mar. fresh. Res. 31: 667–699. Dunlop, R. W. & P. R. Jeffries, 1985. Hydrocarbons of the hypersaline basins of Shark Bay, Western Australia. Org. Geochem. 8: 313–320. Gearing, P. J., J. N. Gearing, T. F. Lytle & J. S. Lytle, 1976. Hydrocarbons in 60 north east Gulf of Mexico Shelf sediments: a preliminary study . Geochim. Cosmochim. Acta, 40: 1005–1017. Huang, W. Y. & W. G. Meinshein, 1979. Sterols as ecological indicators. Geochim. Cosmochim. Acta, 43: 739–745. Kenig, F., A. Y. Huc, B. H. Purser & J. L. Oudin, 1990. Sedimentation, distribution and diagenesis of organic matter in a recent carbonate environment, Abu Dhabi, U.A.E. In Baker, E. W. & A. G. Douglas (eds), Advances in Organic Geochemistry 1989. 735-747. Org. Geochem. 16: 735–747. Kenig, F., J. S. Sinninghe Damste, N. L. Frewin, J. M. Hayes & J. W. de Leeuw, 1995. Molecular indicators for palaeoenvironmental change in a Messinian evaporitic sequence (vana del Gesso, Italy). II: High-resolution variations in abundances and 13 C contents of free and sulphur-bound carbon skeletons in a single marl bed. Org. Geochem. 23 (6): 485–526. Meissner, F. F., J. Woodward & J. L. Clayton, 1984. Stratigraphic relationship and distribution of source rocks in the Greater Rocky Mountain Region. In: Woodward J., F. F. Meissner & J. L.

Clayton (eds), Hydrocarbon Source Rocks of Greater Rocky Mountain Region. Rocky Mountain Association of Geology, Denver, CO.: 1–34. Meyers, P. A. & R. Ishiwatari, 1993, Lacustrine organic geochemistry – an overview of indicators of organic matter sources and diagenesis in lake sediments. In: Meyers, P. A. & R. Ishiwatari (eds), Lacustrine Organic Geochemistry. Org. Geochem. 20: 867–900. Nichols, P. D., Volkman, J. K., Palmisano, A. K., Smith, J. A. & D. C. White, 1988. Occurrence of an isoprenoid C25 diunsaturated alkene and high neutral lipid content in Antarctic sea-ice diatom communities. J. Phycol. 24: 90–96. Requejo, A. G., J. G. Quinn, J. N. Gearing & P. J. Gearing, 1984. C25 and C30 biogenic alkenes in sediment from upper anoxic basin of the Pettaquamscutt River (Rhode Island, U.S.A.). Org. Geochem. 7: 1–10. Robson, J. N., 1987. Synthetic and biodegradation studies of some novel isoprenoid hydrocarbons. Ph.D. Thesis, Plymouth Polytechnic, Plymouth, 242 pp. Robson, J. N. & S. J. Rowland, 1988. Biodegradation of highly branched isoprenoid hydrocarbons: a possible explanation of sedimentary abundance, In: Mattavelli, L. & L. Novelli (eds), Advances in Organic Geochemistry Part II Analytical Geochemistry. Org. Geochem. 13: 691–695. Rowland, S. J., D. A. Yon, C. A. Lewis, & J. R. Maxwell, 1985. Occurrence of 2,6,10-trimethyl-7-(3-methylbutyl)dodecane and related hydrocarbons in the green alga Enteromorpha prolifera and sediments. Org. Geochem. 8: 207–213. Seifert, W. K. & J. M. Moldowan, 1981. Paleoreconstruction by biological markers. Geochim. Cosmochim. Acta 45: 783–794. Thoms, M. C. & W. D. Williams, 1993. The siliceous sediments of Lake Cantara South, a saline lake in South Australia. Int. J. Salt Lake Res. 2(1): 29–40. Volkman, J. K., 1986. A review of sterol markers for marine and terrigenous organic matter. Org. Geochem. 9: 83–99. Volkman, J. K., S. M. Barrett & G. A. Dunstan, 1994. C25 and C30 highly branched isoprenoid alkenes in laboratory cultures of two marine diatoms. Org. Geochem. 21 (3/4): 407–413. von der Borch, C. C., 1976. Stratigraphy and formation of Holocene dolomitic carbonate sediments of the Coorong area, South Australia. J. sediment. Petrol. 46: 952–966. Warren, J. K., 1988. Sedimentology of Coorong dolomite in the Salt Creek region, South Australia. Carbonates Evaporites 3 (2): 175– 199. Warren, J. K., 1990. Sedimentology and mineralogy of dolomitic Coorong lakes, South Australia. J. sediment. Petrol. 60 (6): 843– 858. Williams, W. D., 1986. Conductivity and salinity of Australian salt lakes. Aust. J. mar. freshwat. Res. 37: 177–182. Yon, D. A., G. Ryback & J. R. Maxwell, 1982. 2,6,10-trimethyl-7(3-methylbutyl)dodecane, a novel sedimentary biological marker compound. Tetrahedron Lett. 23: 2143–2146.