Acyclic Alkanes in the Soil over Heshang Cave in Qingjiang, Hubei Province

Biomarkers have been widely used to explore global change such as paleoclimate and paleovegetation based on the analyses of marine sediments ([Volkman et al., 1995), lacustrine sediments ([Sheng et al., 1999), snow and ice ([Xie et al., 1999), peat ([Xie et al., 2004), and stalagmites ([Xie et al., 2005, 2003a). Research into biomarkers in China has been conducted in many soils. A series of biomarkers were identified from the Jiuzhoutai paleosol-loess profile in Lanzhou, and n-alkanes were proposed as a means of reconstructing the paleovegetation ([Xie et al., 2002). The Pleistocene vermicular red earth in South China contains many types of biomarkers, and the variation of a series of lipid indices is comparable with the marine oxygen isotope record ([Liang et al., 2005; [Xie et al., 2003b). In addition, the biomarkers in modern soils could discriminate among the climatic regions and the standing vegetation types ([Wang et al., 2003).

However, organic geochemical studies of modern soils in the karst area have not received much attention, and the fate and biotransformation of soil lipids are not well understood. This will greatly limit the exploitation of lipids present in stalagmites, some of which are believed to originate from the overlying soils. Our objective in this study is to investigate the distribution of common biomarkers such as acyclic alkanes in the soils directly overlying a karst cave, to see if they display trends in distribution and abundance related to paleovegetation or biogeochemical changes. This research will provide a basis for the further study of stalagmite lipids in the Qingjiang area.

SAMPLING AND EXPERI MENTAL METHODS

Samples for lipid analysis were acquired from the modern soil layer on the top of the Heshang cave in Qingjiang valley, in the Middle Yangtze River drainage area in southern China. The cave is located at an altitude of 205 m above sea level, developed in Cambrian limestone. This subtropical region is characterized by the dominance of the East Asian summer monsoon, leading to abundant precipitation. The annual average rainfall is about 1 400 mm and the annual temperature is between 16-17 ℃. The soil samples analyzed in this study were taken from an altitude of 574 m (N 30°26′44.4〞, E 110°25′10.3〞). Steep topography and heavy bushes restricted access to this site, leaving no imprints of human activity.

The samples were collected in August. After removing surface deadwood and defoliation, 7 soil samples were taken from the surface soil layer successively downward to the bottom soil layer (Fig. 1; Table 1). The samples were packed in hop-pockets, and dried in the open air immediately. After being ground to 80 meshes, the samples (70-100 g each) were soxhlet extracted with chloroform. The total lipid extracts were fractionated by flash column chromatography into saturated hydrocarbons, aromatics, and nonhydrocarbons by successively eluting with n-hexane, benzene, and methanol. The saturated hydrocarbons were directly analyzed using GC-MS.

Figure  1.  The carbon distribution of n-alkanes.

DownLoad: Full-Size Img PowerPoint

Table  1.  Soil profile from above Heshang cave

 | Show Table

DownLoad: CSV

A Hewlett-Packard 5973a mass spectrometer, interfaced directly with a 6890 gas chromatograph equipped with an HP-5MS fused silica capillary column (30 m×0.25 mm i.d; film thickness 0.25 μm) was used for molecular analyses. The operating conditions were as follows: temperature ramped from 70 to 280 ℃ at 3 ℃/min, held at 280 ℃ for 15 min, the carrier gas is helium; ionization energy of the mass spectrometer set at 70 eV; scan range from 50 to 550 u.

RESULT AND DISCUSSION n-alkanes

The content of total extractable organic matter in the soil samples ranges from 170.6 to 725.8 μg/g, and decreases with depth (Fig. 2a). This result is consistent with other studies in modern soils. It was proposed that the content of organic matter becomes lower with depth, owing to microbial activity ([Wright and Coleman, 2000; [Vinther et al., 1999; [Hendrix et al., 1998).

Figure  2.  Indices of extractable organic matter, CPI_L, CPI_h, R_h/l, ACL change with the depth. CPI_L=odd[C_15-21]/even[C_16-22], CPI_h=2×oddΣ[C_23-33]/ (even[C_22-32]+even[C_24-34]), ACL= (Σ[C_i]×i) /Σ[C_i] (15≤i≤34), R_h/l=ΣC₂₂⁺/ΣC₂₁^-.

DownLoad: Full-Size Img PowerPoint

Normal alkanes in the extracts mainly range from C₂₅ to C₃₃ (Fig. 1) with a maximum at C₃₁, showing a monomodal distribution. A distinct odd-over-even carbon predominance (CPI_h 5.92-7.80; refer to Fig. 2 for the definition) was observed above C₂₂ throughout the profile (Fig. 2a). This characteristic of n-alkane distribution is a typical model for a higher plant input, agreeing with other modern soil studies ([Baas et al., 2000; Bull et al., 2000, 1998).

In addition, there appears to be a very small proportion of lower-molecular-weight n-alkanes in all the soil samples, which is contributed by soil microbes. The CPI values of these LMW (lower molecular weight) alkanes range from 0.97 to 1.32, with a mean value of 1.10, consistent with a microbial contribution. [Cranwell et al. (1987) also considers that the LMW alkanes mainly originate from algae, fungi and other microorganisms.

The ratio of LMW to HMW of n-alkanes indicates the abundance of microorganisms relative to higher plants (Fig. 2). This index has already been applied in many studies, such as snow and ice ([Xie et al., 1999), vermicular red earth ([Xie et al., 2003b), and modern soil ([Wang et al., 2003), and is believed to be sensitive to environmental conditions. In this soil profile, the R_h/l ratio is high in the surface layer, declines quickly at a depth of 5 cm, and keeps invariable below 10 cm. The surface soil is mainly dominated by humic substances formed from fresh plant and animal residues, leading to the occurrence of abundant HMWn-alkanes and thus a high R_h/l ratio. Enhanced microbial biodegradation of the HMW n-alkanes to LMW homologues induces the decrease of the R_h/l ratio in deeper samples.

Significantly, the three indices of CPI_h, ACL and R_h/l show the same trends in the soil profile, indicative of comparable controlling factors. One of the possible factors is microbial activity. Algaes are capable of producing n-alkanes ranging in chain length from C₁₄-C₃₂, usually without an odd carbon predominance ([Gelpi et al., 1970). Most photosynthetic bacteria also contain predominantly C₁₄-C₂₀ hydrocarbons, and many non-photosynthetic bacteria were reported to have C₂₆-C₃₀ hydrocarbons ([Albro, 1976). So the organic matter derived from the microorganisms will result in a lower value both in CPI_h and the R_h/l ratio.

An alternative explanation would be the contribution from vegetation change. However, the dominance of C₃₁n-alkanes throughout the interval analyzed herein rules out this possibility. Furthermore, the vegetation would not naturally be changed in such a short period in this site, as it is not easily accessible by human beings. Nevertheless, the ACL value might directly be related to a small climate change. In warmer tropical climates, land plants are postulated to biosynthesize longer chain compounds for their waxy coatings, whereas in cooler temperate regions somewhat shorter chain compounds are produced. ACL values of plants in warm periods would consequently be larger than those of plants in cool periods ([Poynter et al., 1999). So the greater ACL value in top samples may relate to the global warming. The elevated ACL value was also observed at the bottom soil, accompanied by enhanced extractable organic matter. This might also be related to the occurrence of a warm paleoclimate.

Mono methylalkanes

Seven series of monomethylalkanes (MMAs) have been identified in the samples by comparison of their mass spectral fragmentation pattern and retention times with published date ([Kenig et al., 2005; [Lu et al., 2003); they include C₁₆-C₂₂ 2-methylalkanes, C₁₆-C₂₂ 3-methylalkanes, C₁₇-C₂₂ 4-methylalkanes, C₁₇-C₂₂ 5-methylalkanes, C₁₇-C₂₂ 6-methylalkanes, C₁₇-C₂₂ 7-methylalkanes, and C₁₇-C₂₂ 8-methylalkanes (Fig. 3). There is also some 7-methyl and 8-methylheptadecane identified in the soil samples.

Figure  3.  The distribution of branched aliphatic alkanes.

DownLoad: Full-Size Img PowerPoint

The MMAs are ubiquitous components in various sediments. The short-chain MMAs are considered to be from microorganisms such as cyanobacteria. [Shiea et al. (1990) found that microbial mats without cyanobacteria lack these compounds. MMAs have been identified in studies of cyanobacteria ([Dembitsky et al., 2001; [Gernot et al., 1999; [Koster et al., 1999; [Dachs et al., 1998; [Gelpi et al., 1970; [Han et al., 1968). Short-chain homologues (C₁₄-C₂₂) have been identified in a Late Triassic oil sand sample, and compound-specific stable carbon isotopic analysis confirmed the primarily cyanobacterial origin of the MMAs ([Lu et al., 2003).

Typically, 7-methylheptadecane and 8-methylheptadecane are usually considered as biomarkers of the cyanobacteria ([Shiea et al., 1991, 1990; [Robinson and Eglinton, 1990; [Gelpi et al., 1970; [Han et al., 1968). However, MMAs found in the culture of cyanobacteria usually range from C₁₇ to C₂₁ ([Kenig et al., 1995; [Shiea et al., 1990), in striking contrast with MMAs identified in ancient sediments and oils, which often range from C₁₅ to C₃₃ ([Summons et al., 1988; [Jackson et al., 1986). This discrepancy questions cyanobacteria as the sole source of MMAs. [Thiel et al. (1999) identified homologous series of C₁₅-C₂₅ mid-chain branched alkanoic acids in eubacteria living symbiotically with demosponges. This suggests MMAs could be diagenetic products of functionalized lipid precursors in these microbes. So the MMAs identified in the soil samples mainly originate from soil microorganisms, and cyanobacteria in particular.

Abundant diploptene has been identified in the soil samples, which might also come from cyanobacteria. Some cyanobacteria were reported to contain 86% diploptene in the total lipid ([De Rosa et al., 1971). The biomarker information discussed above probably infers cyanobacteria multiply in the surface soil over the Heshang cave.

Isoprenoid

Using the fragment ion 183, we can easily identify pristine (Pr), phytane (Ph), squalane and 2, 6, 10-trimethyl penta decane on the mass chromatogram. The ratios of Pr and Ph, which range from 0.65 to 0.82, are relatively stable in the whole soil profile. This shows that the sedimentary environment has weak reducibility. The relative abundance of 2, 6, 10-trimethyl penta decane and squalane is low, and they probably originate from squalene (Gohring et al., 1964).

CONCLUSIONS

Some acyclic alkanes were identified in the overlying soil of the Heshang cave in Qingjiang. The distribution of n-alkanes is mainly from C₂₅ to C₃₃ in carbon number, with a maximum at C₃₁, indicating an input mainly from higher plants. The CPI_h, R_h/l and ACL decline with depth, mostly caused by microbial degradation as well as paleoclimate change. Seven series of MMAs were detectable in the soil, probably derived from soil microbes, particularly cyanobacteria.