Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile)

G Model

ARTICLE IN PRESS

CHROMA-357751; No. of Pages 9

Journal of Chromatography A, xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile)夽 Ana Z. Miller a , José M. De la Rosa a,∗ , Nicasio T. Jiménez-Morillo a , Manuel F.C. Pereira b , José A González-Pérez a , José M. Calaforra c , Cesareo Saiz-Jimenez a a Instituto de Recursos Naturales y Agrobiología de Sevilla, Consejo Superior de Investigaciones Científicas (IRNAS-CSIC), Av. Reina Mercedes 10, 41012, Sevilla, Spain b CERENA, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais 1049-001, Lisbon, Portugal c Department of Biology and Geology. University of Almeria, La Ca˜ nada de San Urbano 04120 Almeria, Spain

a r t i c l e

i n f o

Article history: Received 2 April 2016 Received in revised form 14 July 2016 Accepted 15 July 2016 Available online xxx Keywords: Analytical pyrolysis Speleothems Lava tubes Environmental changes Easter Island

a b s t r a c t This study comprises an innovative approach based on the combination of chromatography (analytical pyrolysis and pyrolysis compound-specific isotope analysis (Py-CSIA)), light stable isotopes, microscopy and mineralogy analyses to characterize the internal layering of coralloid speleothems from the Ana Heva lava tube in Easter Island (Chile). This multidisciplinary proxy showed that the speleothems consist of banded siliceous materials of low crystallinity with different mineralogical compositions and a significant contribution of organic carbon. Opal-A constitutes the outermost grey layer of the coralloids, whereas calcite and amorphous Mg hydrate silicate are the major components of the inner whitish and honeybrown layers, respectively. The differences found in the mineralogical, elemental, molecular and isotopic composition of these distinct coloured layers are related to environmental changes during speleothem development. Stable isotopes and analytical pyrolysis suggested alterations in the water regime, pointing to wetter conditions during the formation of the Ca-rich layer and a possible increase in the amount of water dripping into the cave. The trend observed for ␦15 N values suggested an increase in the average temperature over time, which is consistent with the so-called climate warming during the Holocene. The pyrolysis compound-specific isotope analysis of each speleothem layer showed a similar trend with the bulk ␦13 C values pointing to the appropriateness of direct Py-CSIA in paleoenvironmental studies. The ␦13 C values for n-alkanes reinforced the occurrence of a drastic environmental change, indicating that the outermost Opal layer was developed under drier and more arid environmental conditions. © 2016 Elsevier B.V. All rights reserved.

1. Introduction In the last decade, revolutionary instrumentation for obtaining analytical information on the composition and chemistry of complex solid samples has emerged. Applied gas chromatography coupled to isotope ratio mass spectrometry (GC-C-IRMS) is an analytical method that allows compound-specific isotope analysis (CSIA). This technique provides valuable information on

夽 Selected paper from the XVI Latin-American Congress on Chromatography (XVICOLACRO) and the 9th National Meeting on Chromatography (9ENC), 5–9 January 2016, Lisbon, Portugal. ∗ Corresponding author. E-mail address: [email protected] (J.M. De la Rosa).

biogenetic and geographic origin of organic molecules [1,2]. Nevertheless, the application of chromatographic techniques to study cave speleothems is scarce due to their low content in organic compounds and low volatility. Pyrolysis compound-specific isotope analysis (Py-CSIA) combines the advantages of analytical pyrolysis (Py-GC–MS) with IRMS for the characterization of entrapped or low solubility organic compounds. In this case, the pyrolysate is directed to an isotope ratio mass spectrometer (Py-GC-C-IRMS) to measure stable isotope proportions i.e., ␦13 C, ␦15 N, of specific pyrolysis compounds [3]. Early work demonstrated that the pyrolysis process does not produce appreciable fractionation of stable isotopes and therefore the pyrolysis products are considered to be isotopically representative of the starting material [4–6]. Typical configurations of generic Py-CSIA devices have been previously

http://dx.doi.org/10.1016/j.chroma.2016.07.038 0021-9673/© 2016 Elsevier B.V. All rights reserved.

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9 2

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

described [7–9]. This technique has been recently used in a variety of fields from extraterrestrial OM in meteorites [7,8], soil and sediments [10–13], oil shales and petrol [14,15], polymers and food [3,16]. Therefore, Py-CSIA represents a promising technique with high potential for the study of the effect of environmental changes in subterranean environments, such as caves. Analytical pyrolysis (Py-GC–MS) is defined as the thermochemical decomposition of organic materials at elevated temperature in the absence of oxygen [17]. The products of pyrolysis (pyrolysates) are amenable to chromatographic separation which, when combined with a mass spectrometry detector, yields a valid fingerprint information about the molecular structure even of complex mixtures of macromolecular substances. This chromatographic technique has become an important tool for the analytical characterization of a wide spectrum of complex carbonaceous matrices including fossil OM, fungal, algal and higher plant materials, sediments, aged wood, soot and urban dust [18–24]. Conventional analysis used for studying complex environmental matrices, usually requires solvent extraction and a significant amount of sample material (several grams). Pyrolytic techniques, such as Py-CSIA and Py-GC–MS, have well-known advantages such as the requirement of small sample size and little to no sample preparation. Therefore, the combination of Py-GC–MS and Py-CSIA provides accurate information on the molecular and isotopic composition of organic matter from complex matrices. Speleothems, or secondary mineral deposits found in caves, are formed due to dissolution of primary minerals from the host rock and subsequent precipitation. Their formation is greatly prompted by water-rock interactions and climate conditions, which dictate how much water drip into the cave system [25]. In humid climates or during heavy precipitation a relatively rapid speleothem growth may occur, whereas in arid climates or during drought the growth is moderated or ceased. Hence, the composition, abundance and growth pattern of speleothems may be indicative of climate changes, as reported by several authors [25–29]. Both solution caves and lava tubes may contain a variety of speleothems, such as stalactites, stalagmites, flowstone and coralloids. Speleothems in lava tubes generally consist of siliceous minerals owing to the mineralogy and chemical composition of the host basalts [30,31]. Nevertheless, calcium carbonate or calcium sulfate minerals (in the form of calcite or gypsum, respectively) have been reported on the walls and ceilings of lava tubes [32–34]. For instance, the internal structure of many cave corals, or coralloids, found in lava tube caves shows deposition layers with different mineralogical compositions, alternating between siliceous and carbonate minerals [27,35]. This intercalation has been attributed to changes in the pH or chemical composition of dripping water, environmental changes of the surface area overlying caves and diagenesis [25,27,36,37]. Coralloid speleothems are commonly found on the surface of basaltic lava caves. They are defined as a variety of white to grey or dark brown nodular, globular, botryoidal or coral-like speleothems [36]. Their mineralogy has been described as amorphous silica, mainly opal-A [31,38], and their genesis has been interpreted as microbially-mediated [39,40]. The morphology, mineralogy and genesis of speleothems from lava tubes have been investigated by several researchers using conventional mineralogical and isotope analyses [25,27,32]. Analysis of ␦13 C has been successfully applied to characterize the calcite fraction of speleothems. However and taking into account that the overlying sediments are a major source of carbon for speleothem growth, the ␦13 C of the organic fraction (␦13 Corg ) can also provide useful information on the origin of the organic matter (OM) of speleothems [41,42]. Moreover, stable isotopes (␦13 Corg , ␦15 N) and C/N ratios can also be valuable indicators of OM origin and decay [42–44]. The basis for this approach is that each individ-

ual organic compound in OM inherits a specific isotopic signature resulting from its natural origin and post-depositional transformation pathway [45]. While marine and lacustrine sediments, peats, soils, pollen, glacier ice and paleosols have been extensively studied to discern OM origin and environmental changes [17,42,45,46], very few works combining stable isotopes and OM-molecular analyses have been carried out on cave speleothems [47,48]. In this study, stable isotopes, Py-GC–MS and Py-CSIA are combined for the first time to determine the origin and composition of OM in the internal layers of the coralloid speleothems from the Ana Heva lava tube in Easter Island (Chile). Miller et al. [37] previously studied the geomicrobiological interactions and mineralogical composition of these coralloid speleothems using conventional microscopy and mineralogy. However, to fully understand the internal banded structure, composition and origin of these coralloids, more comprehensive biogeochemical analyses based on isotopic signatures and molecular compositions have been incorporated. Moreover, these data have been related to past environmental changes previously documented in peat sequences and sediments from Easter Island [42,46,49,50]. The distribution of pyrolysis compounds and their stable carbon isotopic compositions were analysed and combined with morphological and mineralogical data to understand the origin and composition of the internal layers of the coralloids from the Ana Heva lava tube in Easter Island (Chile). Since the first adaptation ˜ and Eglinof the Py-CSIA hyphenated technique for ␦13 C by Goni ton [4], and to our knowledge, the present work is the first report combining light stable isotopes, Py-GC–MS and Py-CSIA for the characterization of siliceous speleothems. 2. Material and methods 2.1. Site description and sampling Easter Island, or Rapa Nui, is located in the Pacific Ocean about 3500 km west of the Chilean coast. The island has more than 2000 known lava cave entrances, being the Roiho lava field the most explored area. It derived from multiple eruptions of the volcano Maunga Hiva Hiva and has an area of about 4 km2 . The Ana Heva lava tube from the Roiho lava field (27◦ 6 47.90 ’s, 109◦ 24 49.94 W) was probably formed during the last important eruption of the Maunga Hiva Hiva volcano, which occurred between 10,000 and 12,000 years ago [51]. Globular coral-like speleothems, or coralloids, were carefully taken from the ceiling of this lava tube as described in Miller et al. [37]. 2.2. Microstructural and mineralogical characterization 2.2.1. Stereomicroscopy Several speleothem subsamples were impregnated with epoxy resin and cross sectioned for microstructural characterization of the internal structure of the coralloid stalactites. Polished crosssections of coralloid speleothems were examined using an Olympus SZ51 stereomicroscope (Olympus, Tokyo, Japan) to describe the internal structure of the samples and hand-picking visually homogeneous materials for further analyses. 2.2.2. Field emission scanning electron microscopy To appraise the morphology and chemical composition of the internal layers of the speleothems, field emission scanning electron microscopy combined with energy dispersive X-ray spectroscopy (FESEM-EDS) were performed. Polished cross-sections were directly mounted on sample stubs, sputter coated with gold/palladium, and subsequently examined on a Jeol JSM–7001F microscope (Jeol, Tokyo, Japan) equipped with an Oxford EDS detector (Oxford Microbeam, Oxford, UK). FESEM examinations were

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

operated using the secondary electron (SE) detection mode with an acceleration potential of 15 kV. 2.2.3. X-ray micro-computed tomography The X-ray micro-computed tomography (micro-CT) was performed to assess the three-dimensional (3-D) microstructural and compositional heterogeneities of the coralloids. Digital radiographs were acquired with a micro-CT SkyScan 1172 scanner (Bruker, Massachusetts, USA) using an X-ray cone incident on a rotating specimen. Due to sample variable size and composition the experimental conditions were optimized for each specimen using a constant source power (10 W). The following operating conditions were used: source voltage of 100 kV, current of 100 ␮A, a pixel size resolution ranging from 11.5 to 17.8 ␮m, and an averaging of five radiographs per each position. Downstream 0.5 mm aluminum filtration was used to increase beam penetration into the samples in order to prevent “beam hardening”, a nonlinear X-ray absorption effect. The acquisition was performed by rotating the specimen over 180◦ with variable rotational step. Slice reconstructions were obtained with the NRecon 1.6.3 routine, and volumetric visualization was achieved with DataView and CTvox programs, which integrate the instrument software packages. 2.2.4. X-ray diffraction (XRD) The mineralogical composition of the hand-picked materials (layers with different colours) was determined by XRD using a XPERT-PRO (PANalytical, Almelo, The Netherlands) diffractometer with CuK␣ radiation and a X´ıCelerator detector. The measurement conditions were: 40 kV and 35 mA, 0.002◦ 2␪ step size and 20 s of counting time. The “High Score Plus” analytical software and PDF2 database were used. 2.3. Isotopic and molecular characterization 2.3.1. Elemental analysis and stable isotope analysis (13 C and 15 N IRMS) of bulk samples Elemental (EA) analysis was carried out by dry combustion in a flash 2000 HT elemental micro-analyzer (Thermo Scientific, Bremen, Germany) at a combustion temperature of 1020 ◦ C. Total nitrogen (TN) and carbon (TC) were measured in triplicated and total organic carbon (TOC) was determined after the removal of carbonates by treating the soils samples with 1 M HCl. Bulk isotopic signature of light elements (␦13 C and ␦15 N) was analyzed using a Flash 2000 HT elemental micro-analyzer coupled via a ConFlo IV interface unit to a continuous flow Delta V Advantage isotope ratio mass spectrometer (IRMS) (Thermo Scientific, Bremen, Germany) (C/TC-IRMS). Analyses were carried out on triplicate aliquots of each coralloid layer (between 0.5 and 1 mg) that were packed into tin capsules and combusted in excess oxygen at 1020 ◦ C in a reactor packed with chromium oxide, copper oxide and silver wool. Stable isotope results are expressed in the delta notation (␦) as parts per thousand deviation (‰) from corresponding international standards of Pee Dee Belemnite (13 C/12 C, PDB) recognized by the International Atomic Energy Agency (IAEA) [52]. The standard deviation of bulk ␦13 C was typically less than ± 0.05‰ and for bulk of ␦15 N less than ±0.3‰. In addition, Carbon (13 C) isotope analyses of the organic fractions were also performed in decarbonized samples (␦13 Corg ). Approximately 20 mg of each speleothem layer was ground to a fine powder, homogenized in a ball mill and treated with 1M HCl, then washed and dried (40 ◦ C; 48 h). 2.3.2. Analytical pyrolysis (Py-GC–MS) Direct pyrolysis-gas chromatography/mass spectrometry (PyGC–MS) was performed using a double-shot pyrolyser (Frontier

3

Laboratories, model 2020i) attached to a GC–MS Agilent 6890N system. Samples (1–2 mg) were placed in small crucible capsules and introduced into a preheated micro-furnace at (500 ◦ C) for 1 min. The volatile pyrolysates were then directly injected into the GC–MS for analysis. The gas chromatograph was equipped with a HP-5msUI, low polar-fused silica (5%-Phenyl-methylpolysiloxane) capillary column of 30 m × 250 ␮m × 0.25 ␮m film thickness. The oven temperature was held at 50 ◦ C for 1 min and then increased to 100 ◦ C at 30 ◦ C min−1 , from 100 ◦ C to 300 ◦ C at 10 ◦ C min−1 , and stabilized at 300 ◦ C for 10 min. The carrier gas was helium at a controlled flow of 1 mL min−1 . The detector consisted of an Agilent 5973 mass selective detector, and mass spectra were acquired at 70 eV ionizing energy. The mass spectrometer scanned from m/z 40 to m/z 450. Compound assignment was achieved by taking advantage of: (i) ion chromatography via single-ion monitoring (SIM) for the major various homologous series, (ii) via low-resolution mass spectrometry, and (iii) via comparison with published data reported in the literature or stored in digital libraries (NIST and Wiley libraries). To detect the presence of long-chain fatty acids present in the different coralloid layers, the samples were also subjected to thermally assisted hydrolysis and methylation gas chromatography (THM-GC) in the presence of tetramethylammonium hydroxide (TMAH) [53]. This process incorporates a derivatization step into the pyrolysis, whereby the hydroxyl groups of carboxylic acids are converted to methyl esters by the proposed mechanism of hydrolysis at the ester bond, followed by formation of tetramethylammonium salts and finally thermal dissociation to the methylated derivatives. For each sample, approximately 5 mg of material were placed into a quartz boat, to which 2 ␮L of 25% TMAH–methanol solution was added directly on top of the sample. THM-GC pyrolysis analyses were performed on each sample at 500 ◦ C for 1 min. Each of the samples was run under splitless conditions and chromatographic conditions were settled similarly than analytical pyrolysis; however, an initial 5 min solvent delay was used at the start of each analysis to prevent exposing the MS detector to excess TMAH. The mass spectra of fatty acids (analysed as fatty acids methyl esters FAMEs) is dominated by m/z 74, thus this ion was specifically monitored to detect the presence of FAMES. 2.3.3. Pyrolysis compound-specific isotope analysis (Py-CSIA) Direct Py-CSIA of ␦13 C was carried out by coupling a double-shot pyrolyzer (Frontier Laboratories, model 3030D) attached to a GC Trace GC Ultra system. At the end of the chromatographic column the flux is divided to a GC-Isolink System equipped with a microfurnace for combustion set at 1000 ◦ C and coupled via a ConFlo IV universal interface unit to a continuous flow Delta V Advantage isotope ratio mass spectrometer (IRMS) (Thermo Scientific, Bremen, Germany) (Py-GC-C-IRMS). Samples of 1–2 mg in weight were placed in small stainless steel crucible capsules and introduced into a preheated microfurnace at (500 ◦ C) for 1 min. The evolved gases were then directly injected into the GC-C-IRMS system for analysis. The gas chromatograph was equipped with a low polar-polarity HP-5ms-UI – fused silica (5% phenyl-methylpolysiloxane) capillary column of 30 m × 250 ␮m × 0.25 ␮m film thickness. The oven temperature was held at 50 ◦ C for 1 min and then increased to 100 ◦ C at 30 ◦ C min−1 , from 100 ◦ C to 300 ◦ C at 10 ◦ C min−1 , and stabilized at 300 ◦ C for 10 min using a heating rate of 20 ◦ C min−1 . The carrier gas was helium at a controlled flow of 1 mL min−1 . Isotopic ratios are reported as parts per thousand (‰) deviations from appropriate standards recognized by the IAEA [52]. The standard deviations of compound specific ␦13 C were typically less than ±0.1‰. Structural features of specific peaks were inferred by comparing and matching the mass spectra obtained by conventional Py-GC–MS chromatograms obtained using the same column type and identical chromatographic conditions.

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

4

3. Results and discussion 3.1. Morphological, chemical and mineralogical composition of the coralloid speleothems The coralloid speleothems from the Ana Heva lava tube consists of irregular branched stalactites with varied colour (Fig. 1), as previously described by Miller et al. [37]. Fig. 1A shows a cluster of highly irregular coralloid stalactites with several broken branches revealing the honey-brown coloured layer of the inner part of the speleothems. Internally these stalactites are composed of a succession of coloured layers, ranging from white to brown, with vitreous to resinous luster (Fig. 1B). According to their colour and texture, three main layers and a dark surface coating are easily distinguishable, from the surface to the inner part of the coralloid speleothems: (a) a superficial dark brown to black coating, with approximately 0.1 mm thick; (b) a 0.1–1 mm thick semi-transparent greyish layer; (c) a 0.1–2 mm thick white layer, and (d) a heterogeneous honeybrown layer, with 1–5 mm thick and shades ranging from yellowish to honey-brown (Fig. 1B). The white layer shows typical colloidal deposition structure with occasional inclusions of brown lumps (Fig. 1B). This internal layering was clearly observed by the different opacity of the mineralogical phases to X-rays as revealed by micro-CT for all radiographed coralloid samples (Supplementary Fig. S1). FESEM-EDS microanalyses performed on a coralloid crosssection showed differences in the chemical composition of the internal layers (Figs. S2 and S3). In general, they contain variable proportions of Si, Mg and Ca (Table S1 and Fig. S2), corresponding to different minerals, such as opal-A and clay minerals, as revealed by XRD (Fig. S4). Gathering all the microstructural and mineralogical data together, the thin dark coating of the coralloid surface (layer a, Fig. 1B) is probably derived from impurities of the weathering of the overlying volcanic rock and microbial activity [37,54]; the semi-transparent grey layer (layer b, Fig. 1B) is composed of opal-A (hereinafter Opal layer), and the whitish layer (layer c, Fig. 1B) is calcite (hereinafter Ca-rich layer). In contrast, the innermost honeybrown layer (layer d, Fig. 1B) corresponds to an amorphous Mg hydrate silicate and clay minerals, probably smectite and sepiolite (hereinafter Mg-rich layer). 3.2. Elemental (C, N, TOC) and stable isotope analyses of the coralloid layers Table 1 shows the total N (TN) and total OC (TOC) contents, as well as the stable isotopic ratios of organic carbon (␦13 Corg ) and nitrogen (␦15 N) of the three coloured coralloid layers. Their TOC

Table 1 Total Nitrogen (TN), Total Organic Carbon (TOC) and bulk isotopic signatures (␦13 Corg and ␦15 N) of decarbonized samples of the coralloid layers (mean ± STD in% and ‰ respectively). Sample

TN (%)

Mg-rich layer 0.90 ± 0.01 Ca-rich layer 0.90 ± 0.01 0.96 ± 0.01 Opal layer

TOC*(%)

TOC/TN

5.1 ± 0.1 5.7 4.2 ± 0.1 4.7 5.2 ± 0.1 5.4

␦13 Corg * (‰) −26.1 ± 0.6 −27.3 ± 0.4 −24.6 ± 0.6

␦15 N (‰) 0.8 ± 1.2 1.5 ± 1.0 5.4 ± 0.9

* TOC and ␦13 Corg analyses were performed after the removal of the inorganic C by treating the samples with HCl (1 M).

contents range from 4.2 to 5.2%, which are significantly greater than the average OC content of the innermost Mg-rich layer (approx. 0.9%) reported by Miller et al. [37]. The bulk inner part of the coralloid samples from Easter Island was combusted at 1000 ◦ C prior to elemental analysis to eliminate carbonates [37]. Nevertheless, this procedure also removes the OC from the samples. In the present study, the removal of the inorganic carbon was carried out by digestion with HCl (1 M) prior to combustion. Webb and Finlayson [55] reported an OC content of 6.5% in coralloids from the Bald Rock Cave, Australia, which is similar to the OC values obtained in the present study. The C/N ratios (TOC/TN) of the three studied layers are very homogeneous, ranging from 4.7 to 5.7. C/N values from 4 to 10 are typical from organisms rich in protein (algae and microorganisms) [56–58]. Numerous authors have suggested that microorganisms may be significant in the formation of opal-A coralloids by aiding silica deposition [37,38,40]. The stable carbon isotopic ratios of organic carbon (␦13 Corg ) are not influenced by particle size but by climate conditions and microbial activity. In addition, it is well known that ␦13 Corg may be used to distinguish between remains from plants with C3 or C4 metabolic pathways of photosynthesis [58], making them useful indicators in reconstructing past sources of OM [41]. The three layers show values of ␦13 Corg ≤ −24‰, which suggests that during speleothem formation the OM was dominated by C3 plants. Margalef et al. [46], by using a peatland record from Easter Island, noticed a constant presence of C3 plants (␦13 C ≤ −24‰) after 45 cal kyr BP and the existing paleoecological data reports that the island was largely forested in prehistoric times [59]. The more depleted ␦13 Corg values in the Mg-rich (−26.1 ± 0.6‰) and Ca-rich (−27.3 ± 0.4‰) layers (Table 1), as compared with the more recent outer Opal layer (−24.6 ± 0.6‰), can be interpreted as a greater contribution of microbial and C3 OM detrital input, both favoured by wetter conditions (atmospheric precipitation and meteoric water dripping into the cave). Nonetheless, a number of

Fig. 1. Coralloid speleothems from the Ana Heva lava tube. Cross-section showing occasional inclusions embedded in the coloured layers. (A) Top view of a cluster of coralloid stalactites with varied colour. Scale bar: 5 mm. (B) Polished cross-section of a coralloid stalactite depicting four different coloured layers. Brown lumps are observed within the white layer. Scale bar: 1 mm.

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

studies have also documented that the ␦13 Corg values of C3 plants are sensitive to humidity or precipitation, in general, the more negative ␦13 Corg values, the more precipitation or soil water content and vice versa [60–62]. In contrast, the shift to a slightly enriched ␦13 Corg in the Opal layer (Table 1) likely indicates the decreasing of water availability and an arid climate at more recent times. This fairly agrees with the chemical and mineralogical composition of the coralloid speleothems, as the outermost Opal layer (most recent layer) is almost exclusively siliceous, which is consistent with a decrease in the element release rate during speleothem formation. The composition and weathering of the host rock is also important, as the highly fractured nature and high permeability of basalt rock leads to a more rapid process of water dripping into the cave, and thus less water-rock interactions. The biogeochemistry data from Mann et al. [49] suggest that Easter Island suffered a period of warm and dry climate episode sometime after 4090–4410 yr BP, which only finished centuries or decades before 1290 CE. These high temperatures and significant reduction in the precipitations caused severe droughts of the lake basins of the island. Other researchers have reported further periods of drought between ca 200 and 700 CE [63]. Our findings are in agreement with the hypothesis of Miller et al. [37] regarding the existence of two stages of deposition during the formation of the coralloid stalactites from the Ana Heva lava tube due to climate changes. The values of ␦15 N for the three layers of the coralloids (Table 1) are typical of N fixation (from 0 to 7‰), and 4‰ is generally considered the average of land plants [64]. Globally, soil and plant ␦15 N have been shown to decrease with decreasing mean annual temperature [65]. In cold ecosystems, soil biomass activity is reduced and little nitrogen is lost from the cycle, thus soil and plant

5

␦15 N remain low [66]. In hot and/or arid ecosystems a greater amount of nitrogen moves from the organic to the mineral nitrogen pools, which are subject to preferential loss of 14 N through leaching, denitrification and ammonia volatilization, resulting in higher soil and plant ␦15 N [66,67]. Therefore, the ␦15 N values observed in this study could indicate a trend from colder (Mg-rich layer: 0.8 ± 1.2‰) to warmer temperatures (Opal layer: 5.4 ± 0.9‰). Moreover, Kleinebecker [68] reported that an increase in the ␦15 N signal in soils and plants was consistently linked to a decrease in the plant diversity. This suggests that during the formation of the coralloid speleothems, a drastic reduction in plant diversity occurred during the final stage (Opal layer) of the coralloid formation. Mann et al. [49] stated that currently Easter Island is a treeless landscape and grass species dominate the vegetation, mainly due to deforestation by Polynesian farmers. Before their arrival around 1200 CE to Easter Island, the prehistoric land vegetation was a mesophytic forest dominated by giant palms [69].

3.3. Analytical pyrolysis (Py-GC–MS) The total ion current chromatograms (TIC) of the OM included in the three coralloid layers (Mg-rich, Ca-rich and Opal layers) are shown in Fig. 2, and the identified compounds are listed in Table 2. The three pyrograms are similar. Briefly, they reveal a diversity of entrapped organic compounds, mainly consisting of alkylbenzenes, naphthalenes (including methyl-naphthalenes), series of n-alkanes (C8 to C24 ) and a few N-containing compounds. In addition, THMGC hydrolysis in the presence of TMAH shows the presence of well resolved series of fatty acids for the Ca-rich layer (Fig. 3). The spe-

Fig. 2. Analytical pyrolysis (Py-GC–MS) of the three coralloid layers from the Ana Heva lava tube with indication of the corresponding peak labels according to Table 2.

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

6

Table 2 Products released by Py-GC–MS at 500 ◦ C of the coralloid layers with the peak labels indicated in Fig. 2. The retention time (R.T.), component series (Family), molecular weight (MW) and ␦13 C of specific compounds determined by Py-CSIA (at 500 ◦ C and the same chromatographic conditions as analytical pyrolysis) are specified. Peak label

Analytical Pyrolysis (Py-GC/MS)

Py-CSIA

Compound

R.T. (min)

Family of compounds

MW

Toluene 2-Octene 2-Furancarboxylic acid p-Xylene Benzene, x,x-dimethyl n-Decane Benzene, 1,2,3-trimethyl Benzene, 1,3,5-trimethyl Benzene, x,x,x-trimethyl Indane Benzene, 1-methyl-3 propyl Benzene,x-ethyl-x,x-dimethyl Benzene,x-ethyl-x,x-dimethyl n-Undecane Benzene, 1,2,3,4 tetramethyl 1H-Indene, 2,3-dihydro-5-methyl Naphthalene Azulene n-Dodecane C5 -Alkylbenzene 1H-Indene, 2,3-dihydro-4,7-dimethyl Undecane, 2,4-dimethyl Naphthalene, x-methyl C6 -Alkylbenzene n-Tridecane Naphthalene, x-methyl C7 -Alkylbenzene n-Tetradecane Naphthalene, 2,3-dimethyl Naphthalene, x,x-dimethyl n-Pentadecane Naphthalene, x,x,x-trimethyl Branched n-alkane Naphthalene, 2,3,6-trimethyl n-Hexadecane n-Heptadecane Pristane (2,6,10,14-Tetramethylpentadecane) Tetramethyl naphthalene n-Octadecane n-Nonadecane C8 -Alkylbenzene n-Eicosane Anthracene, x,x-dimethyl n-Heneicosane n-Docosane n-Tricosane n-Tetracosane

2.34 2.46 2.51 2.97 3.16 3.65 3.71 4.01 4.30 4.45 4.56 4.84 4.92 4.99 5.34 5.57 5.82 6.15 6.18 6.36 6.45 7.10 7.12 7.25 7.33 7.48 7.90 8.55 9.02 9.30 9.75 10.28 10.40 10.75 10.93 12.09 12.20 12.10 13.15 14.10 14.40 15.05 15.65 16.01 16.86 17.61 18.41

Ar Alk Alk Ar Ar Alk Ar Ar Ar Ar Ar Ar Ar Alk Ar Ar Ar Ar Alk Ar Ar Alk Ar Ar Alk Ar Ar Alk Ar Ar Alk Ar Alk Ar Alk Alk Alk Ar Alk Alk Ar Alk Ar Alk Alk Alk Alk

92 112 112 106 106 142 120 120 120 118 134 134 134 156 134 132 128 128 170 148 146 184 142 162 184 142 176 198 156 156 212 170

(␦13 C; ‰ vs VPDB) Mg-rich layer

1 C8 2 3 4 C10 5 6 7 8 9 10 10 C11 11 12 13 13 C12 14 15 C11-br 16 17 C13 18 19 C14 20 21 C15 22 23 24 C16 C17 Pr 25 C18 C19 26 C20 27 C21 C22 C23 C24

Summary of ␦13 C data

Average Total (Py-CSIA) Average aromatic Average Alkane

170 226 240 268 184 256 270 190 284 206 298 312 326 340

Ca-rich layer

Opal layer

−27.8 −24.9 −24.9

−29.3 −29.1

−25.1 −24.8

−23.9 −25.1

−29.4 −29.5

−23.6 −24.7

−23.3 −23.6 −23.5

−27.8 −28.9 −28.5

−22.5 −23.2 −22.6

−23.2 −22.7 −22.5

−29.8 −26.5 −28.1

−22.6 −21.6 −21.5

−26.2

−28.5

−26.7

−28.4 −28.3

−28.8 −28.9

−27.7 −27.5

−29.0

−29.8

−26.8

Py-CSIA (␦13 C; ‰ vs VPDB) Mg-rich layer

Ca-rich layer

−25.0 ± 2.2 −23.8 ± 0.9 −28.0 ± 1.2

−28.7 ± 0.9 −28.6 ± 1.0 −29.0 ± 0.6

cific differences observed between the three layers of the coralloids are discussed in the next paragraphs. 3.4. Alkylbenzenes This series of compounds is the most abundant in the three speleothem layers, being toluene, xylene and trimethylbenzene (peaks 1, 3 and 5) the most abundant compounds. Toluene is a ubiquitous pyrolysis product. Low molecular weight aromatic compounds, i.e. alkylbenzenes, indanes and naphthalenes, are commonly observed in the pyrolysis products of natural macromolecular OM. The pyrogram of the Ca-rich layer is particularly

Opal layer −24.4 ± 2.2 −23.2 ± 1.3 −27.2 ± 0.5

enriched in C4 to C8 -alkylbenzenes. Although previous pyrolysis studies of soil OM linked the presence of Cn -alkylbenzene series to the degradation of lignin-derived structures [70], it is also known that alkylbenzenes can be formed directly through cyclization of aliphatic chains upon pyrolysis or biomass burning [71]. 3.5. n-Alkanes n-Alkane homologous were identified in all the pyrolysates. They ranged from C8 to C20 for the Mg-rich layer, C8 –C24 for the Ca-rich layer and C8 –C21 for the Opal layer, with the most abundant homologs in the C14 –C16 range. A similar pattern has

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

7

3.7. Fatty acid methyl esters (FAMEs) It is known that analytical pyrolysis presents some limitations in the detection of FAs due to the incompatibility with the apolar columns used in GC [75] and to secondary reactions that occurs at elevated temperatures [76], and thus a derivatization step into the pyrolysis was carried out (TMAH-Py-GC–MS). Fatty acid methyl esters (FAMEs C6 to C18 ) were identified in the pyrogram of the Carich layer (Fig. 3, m/z = 74), with palmitic acid methyl ester (C16 ) as the major compound; traces of FAMEs were detected in the Mg-rich layer. These FAMEs derive from recent biogenic sources, microorganisms and plant waxes [77,78]. The complete absence of FAMEs in the amorphous Opal layer points out to different speleothem formation conditions (e.g. water shortage and consequently low microbial activity). This reinforces the mineralogical, isotopic and elemental data, discussed in the Supplementary data and in Section 3.2. 3.8. Pyrolysis compound-specific isotope analyses (Py-CSIA)

Fig. 3. Single ion monitoring (SIM) chromatogram trace of mass fragments m/z 74 from the pyrograms of each coralloid layer. Numbers on the peaks refer to carbon chain lengths of n-fatty acid methyl esters.

been attributed to the pyrolysis products of aliphatic biomolecules found in phototrophic microorganisms and plant roots [72]. The n-alkanes with C27 to C33 carbon atoms originated from waxes typical of higher land plants were absent in the coralloid layers. The Carbon Preference Index (CPI) is the sum of the odd-carbonnumber homologs over a specified range divided by the sum of the even-carbon-number homologs over the same range. It is usually significantly greater than 1 for recent OM from unaltered terrestrial vegetation. Our samples resulted in CPI close to 1, which could suggest the alteration (biological and/or physical) of the OM. The detection of larger n-alkane series (C8 –C24 ) in the Ca-rich layer can be caused by a higher abundance of OM in this layer, probably related to an increase in the dripping water from the overlying andosol, which is also consistent with the ␦13 Corg data obtained for this coralloid layer. In addition to the linear n-alkanes, the isoprenoid pristane was found in the Ca-rich layer. It is primarily derived from the phytol side chain of the chlorophyll molecule [73].

3.6. Polycyclic aromatic hydrocarbons (PAHs) Several polycyclic aromatic hydrocarbons were identified in the three samples. They consist of naphthalenes (including mono-, di- and tri-methylnaphthalenes) and a minor peak corresponding to dimethylanthracene (peak 27 in Table 2). These products may derive from incomplete combustion of OM and have been identified, among other sources, in particles from plant burning [17,74].

It was possible to obtain isotopic composition readings for several compounds common to the three coralloid layers and to assign ␦13 C values confidently to different chemical structures (Fig. 2 and Table 2). Among these, series of low molecular weight aromatic compounds (LMWA), such as alkylbenzene series from C2 to C4 , indane (peak 8) and methylindane (peak 12), and the polycyclic aromatic hydrocarbon naphthalene (peak 13) were detected. In addition, series of n-alkane in the range of C15 to C18 were also found. In general, the differences among specific compounds within each layer showed a similar trend with the bulk ␦13 C values, being the ␦13 C of the specific compounds from the Ca-rich layer < Mgrich layer < Opal layer. This correspondence between the values of the individual compounds and the bulk ␦13 C (Fig. 2 and Table 2) is an indication that no preferential C isotopic fractionation occurs during the pyrolysis process. Significant 13 C depleted signatures (−28.6 ± 1.0‰) were found for the Ca-rich layer. The LMWA may derive from the pyrolysis/combustion of biomass, i.e. wildfires [17]. In addition, the ␦13 C values of aromatic compounds were similar to that of burnt fuel wood [79]. For the other speleothem layers, significantly more 13 C enriched signatures were found (−23.8 ± 0.9‰ for the inner Mg layer and −23.2 ± 1.3‰ for the outer Opal layer), which indicates a possible contribution of microbial activity or a mixture of both burning biomass and microbial reworked OM. Easter Island was an unaffected fire ecosystem prior to the first human arrivals that began ca. 1200 CE [80]. Therefore, the presence of a conspicuous light ␦13 C signature for LMWA in the Ca-rich layer could indicate that its formation and development may correspond to the period of the Polynesian settlement and the use of fire that ended with the complete deforestation of the island ca. 1200-1650 CE. Taking into consideration that n-alkanes (Table 2) are known to be isotopically 13 C depleted in comparison with the bulk SOM [81] and that ␦13 C values decrease with increasing chain length [82], the average values obtained for the n-alkanes series from C15 to C18 in each layer (−28.0 ± 1.2‰ for the Mg-rich layer, −29.0 ± 0.6‰ for the Ca-rich layer and −27.2 ± 0.5‰ for the outermost Opal layer) show good correlation (R2 = 0.977) with the bulk ␦13 C of the OM contained in the corresponding speleothem layer (Table 1). Therefore this finding points to the appropriateness of direct Py-CSIA as an accurate technique for measuring ␦13 C isotopic signatures of n-alkanes. Several recent studies have demonstrated that leaf wax biomarker ␦13 C vary considerably among different plant species growing at the same location [[83] and references therein]. For the

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

8

outermost Opal layer, ␦13 C values for alkanes were significantly 13 C enriched than the Ca-rich layer. This may indicate that the organisms that produced the original biomass in the outermost Opal layer grew under drier and more arid environmental conditions, reinforcing the occurrence of a drastic environmental change. 4. Conclusions The microstructural and mineralogical approach applied for characterizing the internal structure of the coralloid speleothems from the Ana Heva lava tube revealed that they consist of three major coloured layers with different mineralogical composition and OM content. Biogeochemical analyses based on isotopic signatures and molecular characterization confirmed that the genesis of the three coloured layers of the coralloid was related to two different stages of speleothem development caused by environmental changes on Easter Island. Analytical pyrolysis data reflected the contributions of plant remains and microorganisms in the OM entrapped in the coralloid speleothems. Bulk and compoundspecific isotope analyses suggested different water regime during speleothem development, as well as the presence of 13 C depleted series of low molecular weight aromatic compounds with a probable pyrogenic origin in the Opal layer. Variations in ␦13 Corg values pointed to wetter conditions during the formation of the innermost Mg- rich layer and a water shortage during the latest stage of speleothem formation (Opal layer) producing n-alkane series with heavier ␦13 C signature. In addition, the trend observed for the ␦15 N values could also indicate an increase on the average temperature, which is consistent with the so-called climate warming during the Holocene. The combination of analytical pyrolysis and mass spectrometric isotope analysis creates possibilities for successfully analysing organic compounds present in speleothems and reveal their origin. Acknowledgements A.Z. Miller acknowledges the support from the Marie Curie Intra-European Fellowship of the European Commission’s 7th Framework Programme (PIEF-GA-2012-328689). The Spanish Ministry of Economy and Competitiveness (MINECO, project CGL2013-41674-P) and the FEDER Funds are acknowledged for the financial support. N.T. Jiménez-Morillo thanks the MINECO for funding his pre-doctoral FPI fellowship (BES-2013-062573). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma.2016.07. 038. References [1] W. Meier-Augenstein, Applied gas chromatography coupled to isotope ratio mass spectrometry, J. Chromatogr. A 842 (1999) 351–371. [2] W. Meier-Augenstein, Stable isotope analysis of fatty acids by gas chromatography–isotope ratio mass spectrometry, Anal. Chim. Acta 465 (2002) 63–79. [3] J.A. González-Pérez, N.T. Jiménez-Morillo, J.M. de la Rosa, G. Almendros, F.J. González-Vila, Pyrolysis-gas chromatography–isotope ratio mass spectrometry of polyethylene, J. Chromatogr. A 1388 (2015) 234–236. ˜ T.I. Eglinton, Analysis of kerogens and kerogen precursors by flash [4] M.A. Goni, pyrolysis in combination with isotope-ratio-monitoring gas chromatography–mass spectrometry (IRM-GC–MS), J. High Resolut. Chrom. 17 (1994) 476–488. [5] T.N. Corso, J.T. Brenna, High-precision position-specific isotope analysis, Proc. Natl. Acad. Sci. U. S. A. 94 (1997) 1049–1053. [6] S. Steinbeiss, C.M. Schmidt, K. Heide, G. Gleixner, 13C values of pyrolysis prod-ucts from cellulose and lignin represent the isotope content of their precursors, J. Anal. Appl. Pyrol. 75 (2006) 19–26.

[7] M.A. Sephton, I. Gilmour, Pyrolysis–gas chromatography–isotope ratio mass spectrometry of macromolecular material in meteorites, Planet. Space Sci. 49 (2001) 465–471. [8] M.A. Sephton, I. Gilmour, Compound-specific isotope analysis of the organic constituents in carbonaceous chondrites, Mass Spectrom. Rev. 20 (2001) 111–120. [9] B. Glaser, Compound-specific stable-isotope (13 C) analysis in soil science, J. Plant Nutr. Soil Sci. 168 (2005) 633–648. [10] A. Ekberg, N. Buchmann, G. Gleixner, Rhizospheric influence on soil respiration and decomposition in a temperate Norway spruce stand, Soil Biol. Biochem. 29 (2007) 2103–2110. [11] W. Liu, Y. Huang, Reconstructing in-situ vegetation dynamics using carbon isotopic composition of biopolymeric residues in the central Chinese Loess Plateau, Chem. Geol. 249 (2008) 348–356. [12] R. Bol, N. Poirier, J. Belesdent, G. Gleixner, Molecular turnover time of soil organic matter in particle-size fractions of an arable soil, Rapid Comm. Mass Spectrom. 23 (2009) 2551–2558. [13] S. Mambelli, J.A. Bird, G. Gleixner, T.E. Dawson, M.S. Torn, Relative contribution of foliar and fine root pine litter to the molecular composition of soil organic matter after in situ degradation, Org. Geochem. 42 (2011) 1099–1108. [14] W. Jia, Z. Xiao, C. Yu, P. Peng, Molecular and isotopic compositions of bitumens in Silurian tar sands from the Tarim Basin, NW China: characterizing biodegradation and hydrocarbon charging in an old composite basin, Mar. Petrol. Geol. 27 (2010) 13–25. [15] S. Tulipani, K. Grice, P. Greenwood, L. Schwark, A pyrolysis and stable isotopic approach to investigate the origin of methyltrimethyltridecylchromans (MTTCs), Org. Geochem. 61 (2013) 1–5. [16] J.A. González-Pérez, N.T. Jiménez-Morillo, J.M. de La Rosa Arranz, G. Almendros, F.J. González-Vila, Compound specific isotopic signature of carbohydrate pyrolysis products from C3 and C4 plants (Py-CSIA), J. Sci. Food Agric. 96 (2016) 948–953. [17] J.M. De la Rosa, E. Lopez Capel, F.J. Gonzalez-Vila, J.A. Gonzalez-Perez, D.A.C. Manning, Direct detection of black carbon in soils by Py-GC/MS, 13 C NMR spectroscopy and thermogravimetric techniques, Soil Sci. Soc. Am. J. 72 (2008) 258–268. [18] C. Saiz-Jimenez, J.J. Boon, J.L. Hedges, J.K.C. Hessels, J.W. de Leeuw, Chemical characterization of recent and buried woods by analytical pyrolysis Comparison of pyrolysis data with 13C NMR and wet chemical data, J. Anal. Appl. Pyrol. 11 (1987) 437–450. [19] C. Saiz-Jimenez, J.J. Ortega-Calvo, J.W. de Leeuw, The chemical structure of fungal melanins and their possible contribution to black stains in stone monuments, Sci. Total Environ. 167 (1995) 305–314. [20] J.G. Lee, C.G. Lee, J.J. Kwag, A.J. Buglass, G.H. Lee, Determination of optimum conditions for the analysis of volatile components in pine needles by double-shot pyrolysis-gas chromatography–mass spectrometry, J. Chromatogr. A 1089 (2005) 227–234. [21] A. Camean, I. Moreno, T. Verdejo, F.J. González-Vila, J.A. González-Pérez, Pyrolytic behaviour of microcystins and microcystin-spiked algal blooms, J. Anal. Appl. Pyrol. 74 (2005) 19–25. [22] B. Pereira de Oliveira, J.M. De la Rosa, A.Z. Miller, C. Saiz-Jimenez, A. Gómez-Bolea, M.A. Sequeira Braga, A. Dionísio, An integrated approach to assess the origins of black films on a granite monument, Environ. Earth Sci. 63 (2011) 1677–1690. [23] S. Braovaca, D. Tamburinib, J. Łucejko, C. McQueena, H. Kutzkea, M.P. Colombi, Chemical analyses of extremely degraded wood using analytical pyrolysis and inductively coupled plasma atomic emission spectroscopy, Microchem. J. 124 (2016) 368–379. [24] J.M. De la Rosa, J.A. Gonzalez-Perez, P.G. Hatcher, H. Knicker, F.J. Gonzalez-Vila, Determination of refractory organic matter in marine sediments by chemical oxidation, analytical pyrolysis and solid-state 13 C nuclear magnetic resonance spectroscopy, Eur. J. Soil Sci. Soc. 59 (2008) 430–438. [25] J.M. Calaforra, P. Forti, A. Fernándes-Cortés, Speleothems in gypsum caves and their paleoclimatological significance, Environ. Geol. 53 (2008) 1099–1105. [26] D. Genty, A. Baker, B. Vokal, Intra- and inter-annual growth rate of modern stalagmites, Chem. Geol. 176 (2001) 191–212. [27] K.S. Woo, D.W. Choi, K.C. Lee, Silicification of cave corals from some lava tube caves in the Jeju Island, Korea: implications for speleogenesis and a proxy for paleoenvironmental change during the late quaternary, Quat. Int. 176–177 (2008) 82–95. [28] A. Moreno, H.M. Stoll, M. Jiménez-Sánchez, I. Cacho, B. Valero-Garcés, E. Ito, R.L. Edwards, A speleothem record of glacial (25-11.6 kyr BP) rapid climatic changes from northern Iberian Peninsula, Global Planet. Change 71 (2010) 218–231. [29] H.M. Stoll, A. Moreno, A. Mendez-Vicente, S. González-Lemos, M. Jiménez-Sánchez, M.J. Dominguez-Cuesta, R.L. Edwards, H. Cheng, X. Wang, Paleoclimate and growth rates of speleothems in the northwestern Iberian Peninsula over the last two glacial cycles, Quat. Res. 80 (2013) 284–290. [30] J.A. Webb, B.L. Finlayson, Incorporation of Al, Mg, and water in opal-A: evidence from speleothems, Am. Mineral. 72 (1987) 1204–1210. [31] R.A.L. Wray, Opal and chalcedony speleothems on quartz sandstones in the Sydney region, southeastern Australia, Aust. J. Earth Sci. 46 (1999) 623–632. [32] K.S. Woo, J.C. Kim, D.W. Choi, J.K. Kim, R. Kim, O. Nehza, The origin of erratic calcite speleothems in the Dangcheomul cave (lava tube cave), Jeju Island, Korea, Quat. Int. 176–177 (2008) 70–81.

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038

G Model CHROMA-357751; No. of Pages 9

ARTICLE IN PRESS A.Z. Miller et al. / J. Chromatogr. A xxx (2016) xxx–xxx

[33] W.B. White, Secondary minerals in volcanic caves: data from Hawai’i, J. Cave Karst Stud. 72 (2010) 75–85. [34] D.E. Northup, L.A. Melim, M.N. Spilde, J.J.M. Hathaway, M.G. Garcia, M. Moya, F.D. Stone, P.J. Boston, M.L. Dapkevicius, C. Riquelme, Lava cave microbial communities within mats and secondary mineral deposits: implications for life detection on other planets, Astrobiology 11 (2011) 1–18. [35] P. Zawidzki, F. Urbani, B. Koisar, Preliminary notes on the geology of the Sarisarinama plateau, Venezuela, and the origin of its caves, Bol. Soc. Venez. Espeleol. 7 (1976) 29–37. [36] C.A. Hill, P. Forti, Cave Minerals of the World, second ed., National Speleological Society, Huntsville, Ala, 1997, pp. 463. [37] A.Z. Miller, M.F.C. Pereira, J.M. Calaforra, P. Forti, A. Dionísio, C. Saiz-Jimenez, Siliceous speleothems and associated microbe-mineral interactions from Ana Heva lava tube in Easter Island (Chile), Geomicrobiol. J. 31 (2014) 236–245. [38] J.R. Vidal Romaní, J. Sanjurjo Sánchez, M. Vaqueiro Rodríguez, D. Fernández Mosquera, Speleothem development and biological activity in granite cavities, Geomorphologie 4 (2010) 337–346. [39] F. Urbani, P. Compère, L. Willems, Opal-A speleothems of Wei-Assipu-tepui, Roraima province, Brazil, Bol. Soc. Venez. Espeleol. 39 (2005) 21–26. [40] R. Aubrecht, Ch. Brewer-Carías, B. Smída, M. Audy, L. Kovácik, Anatomy of biologically mediated opal speleothems in the world’s largest sandstone cave: Cueva Charles Brewer, Chimantá Plateau, Venezuela, Sediment. Geol. 203 (2008) 181–195. [41] P.A. Meyers, Organic geochemical proxies of paleoceanographic paleolimnologic, and paleoclimatic processes, Org. Geochem. 27 (1997) 213–250. ˜ [42] O. Margalef, A.Martínez Cortizas, M. Kylander, S. Pla-Rabes, N. Canellas Boltà, J.J. Pueyo, A. Sáez, B.L. Valero-Garcés, S. Giralt, Environmental processes in Rano Aroi (Easter Island) peat geochemistry forced by climate variability during the last 70 kyr, Palaeogeogr. Palaeoclimatol. Palaeoecol. 414 (2014) 438–450. [43] M.R. Talbot, T. Johannessen, A high resolution palaeoclimatic record for the last 27,500 years in tropical West Africa from the carbon and nitrogen isotopic composition of lacustrine organic matter, Earth Planet Sci. Lett. 110 (1992) 23–37. [44] A.M. Aucour, R. Bonnefille, C. Hillaire-Marcel, Sources and accumulation rates of organic carbon in an equatorial peat bog (Burundi, East Africa) during the Holocene: carbon isotope constraints, Palaeogeogr. Palaeoclimatol. Palaeoecol. 150 (1999) 179–189. [45] P.A. Meyers, Applications of organic geochemistry to paleolimnological reconstructions: a summary of examples from the Laurentian Great Lakes, Org. Geochem. 34 (2003) 261–289. ˜ [46] O. Margalef, N. Canellas-Boltà, S. Pla-Rabes, S. Giralt, J.J. Pueyo, H. Joosten, V. Rull, T. Buchaca, A. Hernández, B.L. Valero-Garcés, A. Moreno, A. Sáez, A 70,000 year multiproxy record of climatic and environmental change from Rano Aroi peatland (Easter Island), Global Planet. Change 108 (2013) 72–84. [47] A.I. Rushdi, P.U. Clark, A.C. Mix, V. Ersek, B.R.T. Simoneit, H. Cheng, R.L. Edwards, Composition and sources of lipid compounds in speleothem calcite from southwestern Oregon and their paleoenvironmental implications, Environ. Earth Sci. 62 (2011) 1245–1261. [48] J.S. Polk, P. van Beynen, Y. Asmerom, V.J. Polyak, Reconstructing past climates using carbon isotopes from fulvic acids in cave sediments, Chem. Geol. 360–361 (2013) 1–9. [49] D. Mann, J. Edwards, J. Chase, W. Beck, R. Reanier, M. Mass, B. Finney, J. Loret, Drought vegetation change, and human history on rapa nui (Isla de Pascua, Easter Island), Quat. Res. 69 (2008) 16–28. ˜ [50] N. Canellas-Boltà, V. Rull, A. Sáez, O. Margalef, R. Bao, S. Pla-Rabes, M. Blaauw, B. Valero-Garcéset, S. Giralt, Vegetation changes and human settlement of Easter Island during the last millennia: a multiproxy study of the Lake Raraku sediments, Quat. Sci. Rev. 72 (2013) 36–48. [51] O. González-Ferrán, Evolución geológica de las Islas Chilenas en el Océano Pacífico, in: J.C. Castilla (Ed.), Islas Oceánicas Chilenas: Conocimiento Científico Y Necesidades De Investigaciones, Ediciones Universidad Católica de Chile, Santiago, 1987, pp. 37–54. [52] S. Valkiers, M. Varlam, K. Ru␤e, M. Berglund, P. Taylor, J. Wang, M. Milton, P. De Bièvre, Quantification of the degree-of-isotopic-equilibrium of carbon and oxygen isotopes in mixtures of CO2 gases, Int. J. Mass Spectrom. 263 (2007) 195–203. [53] H. Otani, Y. Ishida, M. Ushiba, S. Tsuge, Thermally assisted hydrolysis and methylation–gas chromatography of poly(aryl ether sulfone)s in the presence of tetramethylammonium hydroxide, J. Anal. Appl Pyrol 61 (2001) 35–44. [54] A. Miller, M.F.C. Pereira, J.M. Calaforra, P. Forti, A. Dionísio, C. Saiz-Jimenez, Ana Heva lava tube (Easter Island, Chile): preliminary characterization of the internal layers of coralloid-type speleothems, Microsc. Microanal. 21 (2015) 68–69. [55] J.A. Webb, B.L. Finlayson, Allophane and opal speleothems from granite caves in south-east Queensland, Aust. J. Earth Sci. 31 (1984) 341–349. [56] S.G. Hollugan, J.P. Montoya, J.L. Nevins, J.J. McCarthy, Vertical distribution and partitioning of organic carbon in mixed, frontal and stratified waters of the English Channel, Mar. Ecol. Prog. Ser. 14 (1984) 111–127. [57] J.I. Hedges, D.C. Mann, The characterization of plant tissues by their lignin oxidation products, Geochim. Cosmochim. Acta 43 (1979) 1803–1807.

9

[58] P.A. Meyers, Preservation of elemental and isotopic source identification of sedimentary organic matter, Chem. Geol. 144 (1994) 289–302. [59] C. Orliac, The woody vegetation of Easter Island between the early 14th and the mid-17th centuries AD, in: C.M. Stevenson, W.S. Ayres (Eds.), Easter Island Archaeology: Research on Early Rapanui Culture, Easter Island Foundation, Los Osos, 2000, pp. 211–220. [60] C. Hatté, P. Antoine, M.R. Fontugne, A. Lang, D.D. Rousseau, L. Zöller, ␦13 C variation of loess organic matter as a potential proxy for paleoprecipitation, Quat. Res. 55 (2001) 33–38. [61] Y.T. Hong, Z.G. Wang, H.B. Jiang, Q.H. Lin, B. Hong, Y.X. Zhu, Y. Wang, L.S. Xu, X.T. Leng, H.D. Li, A 6000-year record of changes in drought and precipitation in northeastern China based on a ␦13 C time series from peat cellulose, Earth Planet Sci. Lett. 185 (2001) 111–119. [62] X. Chen, J.Y. Rong, J.X. Fan, R.B. Zhan, C.E. Mitchell, D.A.T. Harper, M.J. Melchin, P.A. Peng, S.C. Finney, X.F. Wang, The global stratotype section and point (GSSP) for the base of the hirnantian stage (the uppermost of the ordovician system), Episodes 29 (2006) 183–196. [63] A. Maldonado, C. Villagrán, Paleoenvironmental changes in the semiarid coast of Chile (∼32 ◦ S) during the last 6200 cal years inferred from a swamp-forest pollen record, Quat. Res. 58 (2002) 130–138. [64] D. Maksymowska, P. Richard, H. Piekarek-Janowska, P. Riera, Chemical and isotopic composition of the organic matter sources in the Gulf of Gdansk (Southern Baltic Sea), Estuar. Coast. Shelf Sci. 51 (2000) 585–598. [65] R. Amundson, A.T. Austin, E.A.G. Schuur, K. Yoo, V. Matzek, C. Kendall, A. Uebersax, D. Brenner, T. Baisden, Global patterns of the isotopic composition of soil and plant nitrogen, Global Biogeochem. Cycles 17 (2003) 1031. [66] L.A. Handley, A. Austin, D. Robinson, C. Scrimgeour, J. Raven, T. Heaton, S. Schmidt, G. Stewart, The 15-N natural abundance (␦15 N) of ecosystem samples reflects measures of water availability, Aust. J. Plant Phys. 26 (1999) 185–199. [67] A. Austin, P.M. Vitousek, Nutrient dynamics on a precipitation gradient, Oecologia 113 (1998) 519–529. [68] T. Kleinebecker, N. Hölzel, D. Prati, B. Schmitt, M. Fischer, V.H. Klaus, Evidence from the real world: 15 N natural abundances reveal enhanced nitrogen use at high plant diversity in Central European grasslands, J. Ecol. 102 (2014) 456–465. [69] T.L. Hunt, Rethinking easter island’s ecological catastrophe, J. Archaeol. Sci. 34 (2007) 485–502. [70] H.R. Schulten, G. Gleixner, Analytical pyrolysis of humic substances and dissolved organic matter in aquatic systems: structure and origin, Water Res. 33 (1999) 2489–2498. [71] C. Saiz-Jimenez, The chemical structure of humic substances: recent advances, in: A. Piccolo (Ed.), Humic Substances in Terrestrial Ecosystems, Elsevier, Amsterdam, 1996, pp. 1–44. [72] S. Derenne, C. Largeau, A. Brukner-Wein, M. Hetenyi, G. Bardoux, A. Mariotti, Origin of variations in organic matter abundance and composition in a lithologically homogeneous maar-type oil shale deposit (Gérce, Pliocene Hungary), Org. Geochem. 31 (2000) 787–798. [73] J.K. Volkman, J.R. Maxwell, Acyclic isoprenoids as biological markers, in: R.B. Johns (Ed.), Biological Markers in the Sedimentary Record, Elsevier, New York, 1986, pp. 1–42. [74] L.J. Standley, B.R.T. Simoneit, Characterization of extractable plant wax: resin and thermally matured components in smoke particles from prescribed burns, Environ. Sci. Technol. 21 (1987) 163–169. [75] M.F. Dignac, S. Houot, S. Derenne, How the polarity of the separation column may influence the characterization of compost organic matter by pyrolysis-GC/MS, J. Anal. Appl. Pyrol. 75 (2006) 128–139. [76] C. Saiz-Jimenez, Production of alkylbenzenes and alkylnaphthalenes upon pyrolysis of unsaturated fatty acids. A model reaction to understand the origin of some pyrolysis products from humic substances, Naturwissenschaften 81 (1994) 451–453. [77] J.D. Weete, Fungal Lipid Biochemistry, Plenum Press, New York, 1974. [78] M.A. Mazurek, G.R. Cass, B.R.T. Simoneit, Biological input to visibility-reducing aerosol particles in the remote arid Southwestern United States, Environ. Sci. Technol. 25 (1991) 684–694. [79] K.V. Czapiewski, E. Czuba, L. Huang, D. Ernst, A.L. Norman, R. Koppmann, J. Rudolph, Isotopic composition of non-methane hydrocarbons in emissions from biomass burning, J. Atmos. Sci. 43 (2002) 45–60. [80] J.R. Flenley, S.M. King, J. Jackson, C. Chew, The late quaternary vegetational and climatic history of easter island, J. Quat. Sci. 6 (1991) 85–115. [81] Y. Chikaraishi, H. Naraoka, S.R. Poulson, Hydrogen and carbon isotopic fractionations of lipid biosynthesis among terrestrial (C3, C4 and CAM) and aquatic plants, Phytochemistry 65 (2004) 1369–1381. [82] Y. Huang, M.J. Lockheart, J.W. Collister, G. Eglinton, Molecular and isotopic biogeochemistry of the Miocene Clarkia formation: hydrocarbons and alcohols, Org. Geochem. 23 (1995) 785–801. [83] Y. Eley, L. Dawson, N. Pedentchouk, Investigating the carbon isotope composition and leaf wax n-alkane concentration of C3 and C4 plants in Stiffkey saltmarsh, Norfolk UK, Org. Geochem. 96 (2016) 28–42.

Please cite this article in press as: A.Z. Miller, et al., Analytical pyrolysis and stable isotope analyses reveal past environmental changes in coralloid speleothems from Easter Island (Chile), J. Chromatogr. A (2016), http://dx.doi.org/10.1016/j.chroma.2016.07.038