Alkenones as a proxy for lake temperature?

In this edition of Papers ‘n’ Cake, we discuss an article by Jaime Toney and colleagues (Toney et al., 2010) that explores the occurrence of long-chain alkenones in lacustrine settings and their application as a paleotemperature proxy.

Alkenones in the marine realm:

During the late 1970s and early 1980s, researchers from Delft University of Technology (Boon et al, 1978; De Leeuw et al., 1980) reported the presence of very long-chain (C37 to C39) di- and triunsaturated methyl and ethyl ketones within the sedimentary record. Around the same time, John Volkman, Geoff Eglinton and others at the University of Bristol observed these ketones within cultures of the cosmopolitan haptophyte algae, Emiliani Huxleyi (Volkman et al., 1980) and later showed that the distribution of ketones varied according to temperature. They noted that as temperature increased, the relative proportion of the tri-unsaturated alkenone (C37:3) decreased. This was also evident in the marine sedimentary record and led to the development of the UK37 index as first organic paleothermometer (Brassell et al., 1986):

UK37 = 37:2 – 37:4 / (37:2 + 37:3 + 37:4)

Given that the key alkenone producers are algae, and hence need sunlight and live within the surface waters, this index was correlated with sea surface temperature. It is important to note that the UK37 index includes the C37:4 alkenone which is relatively rare within marine sediments outside of the high latitudes. As a result, the original index was later simplified to take this into account (Prahl and Wakeham, 1987):

UK37’ = 37:2 / (37:2 + 37:3)

Although there are still a number of uncertainties and limitations to UK37 this approach has been used to successfully reconstruct sea surface temperature using marine sediments throughout the Cenozoic (Brassell, 2014). Besides marine sediments, alkenones are also found in lake sediments (Zink et al., 2001); however, the application of long-chain alkenones (LCAs) within lake settings remains poorly understood.

alkeone2.png
Figure 1: Typical alkenone distribution in cultures of E. Huxleyi. Image credit: Geoff Eglinton.

Alkenones within lacustrine settings:

Long-chain alkenones (LCAs) were first reported within the Lake District by Cranwell et al (1985) and have since been discovered in lakes around the world (see Castaneda and Sinninghé Damste 2014 for more details). Within most lakes, the LCA distribution is dominated by the C37:4 compound. As the marine realm typically lacks this compound, the proportion of C37:4 has been proposed as a marker for lacustrine settings (Thiel et al., 1997). Lacustrine settings also contain an abundance of C38 alkenones (although this varies from site-to-site). In some hypersaline lakes, C41 and C42 alkenones (!) have even been reported (Zhao et al., 2014).

However, unlike the marine realm, the relationship between lacustrine LCAs and temperature remains unclear. Most studies indicate a correlation between UK37 and average lake surface temperature (e.g. Zink et al,. 2001) or mean annual air temperature (e.g. Chu et al., 2005). However, a correlation with seasonal temperature has also been noted (e.g. Pearson et al., 2008). In addition, the application of UK37 in some settings can be compromised by the absence of C37 homologues. As such, alternative indices have been proposed, based upon the distribution of C38 homologues (or a combination of the two).

The source of alkenones in lakes:

The source of alkenones within lacustrine settings remains poorly understood. Although the ratio of the C37 to C38 alkenone has been used to infer haptophyte taxonomy, the majority of our understanding is based upon genetic work. In a study of 15 geographically distinct lakes, Theroux et al (2010) indicate that all haptophyte sequences belong to the order Isochrysidales. Most importantly, they do not branch within the most common marine producers (i.e. E. Huxleyi or G. Oceanica.).  Within the same study, Theroux et al. (2010) also provide evidence for the presence of multiple haptophyte species from within a single lake sediment sample, and therefore complicates the application of the alkenone temperature proxy in lakes (see later).

hapto.png
Figure 2: The main alkenone-producing haptophytes in the mairne realm.

The big unknown: what do alkenones actually do?

Despite the widespread occurrence of alkenones within marine and lake sediments, and source organisms readily available in cultures, their function still remains largely unknown. Previous studies suggest they may function as either: 1) membrane lipids or 2) metabolic storage molecules. The former is consistent with changes in the degree of unsaturation as a function of temperature observed in culture experiments whereas an increase in alkenone concentration as a function of growth phase is suggests a metabolic role (Epstein et al., 2001).

The paper: Climatic and environmental controls on the occurrence and distributions of long chain alkenones in lakes of the interior United States (Toney et al., 2010).

As alkenones have the potential for quantitative reconstruction of past continental climate, Toney et al. (2010) surveyed 55 lakes in the North American interior for long chain alkenones (LCAs). In particular, they focus upon a suite of lakes from the NE Sandhills (41-42°N) and the Northern Great Plains (44-48°N). In order to constrain the environmental conditions that favour the occurrence of LCA-producing haptophytes, they also reconstruct water chemistry (major cation and anion concentrations) and other geochemical and physical parameters (surface water temperature, conductivity, salinity, pH, dissolved oxygen).

Alkenone distributions in the North American Interior:

Only 13 of  the 55 lake surface sediments examined contain detectable concentrations of LCAs. The chain length ranges from C37-C39 and contains di-, tri- and tetra-unsaturated alkenones. Intriguingly, 8 of these sites do not contain the C37:4 alkenone, previously considered a marker for lacustrine settings. Instead, these lakes are dominated by the C37:3 alkenone. In a number of lakes, there are also a series of compounds that elute closely with the alkenones. These compounds were present in both the alkenone-containing and alkenone-free lakes and were investigated further via GC-MS. Toney et al. (2010) first identified a C36 compound, which elutes immediately after the C37:3 alkenone. This compound exhibits a distinct m/z 152 base peak and m/z 516 molecular ion. They also identify a C36:1 compound which likely indicates the occurrence of a double bond. The mass chromatogram of m/z 152 reveals a series of related compounds . which range in mass from m/z 348 to 544. As some of these compounds elute at the same time as LCAs, it is important to run samples via GC-MS to ensure proper identification and quantification.

The role of groundwater, carbonate and sulphate chemistry:

A first order difference in the distribution in LCAs between the two major regions is attributed to changes in groundwater geochemistry. In the north, the underlying lithology is shale which contributes high dissolved mineral contents including sulphate due to the oxidation of pyrite. Interestingly, these sites are often associated with the presence of the C37:4 alkenone. In contrast, the south is associated with  a different underlying lithology (sand) and a different LCA distribution.

Changes in carbonate and sulphate content can also provide insights into LCA distributions. In the southern region, the majority of lakes are carbonate dominated (n = 7/8) and lack the C37:4 alkenone. In the north, the lakes are sulphate-dominated. Of these lakes, the majority contain the C37:4 alkenone. When comparing these two regions, the ratio between sulphate and carbonate is significantly different and may be an important factor controlling the occurrence of the C37:4 alkenone. Toney et al. (2010) also note that when sodium content is high, alkenones are usually present. Collectively, this suggests a number of complex environmental parameters control the occurrence of alkenones in lakes.

The C37:4 alkenone as marker for salinity and/or lacustrine settings?

Although previous studies have reported a higher percentage of C37:4 in lower salinity environments, Toney et al. (2010) suggest that there is no clear relationship between salinity and C37:4 abundance. Instead, the presence of the C37:4 is geographically restricted to sites in the north which are associated with the dominance of sodium and sulphate (see above). As such, the C37:4 alkenone should not be used as a proxy for lacustrine settings (c.f. Thiel et al. 1997).

Alkenones as a temperature proxy in the North American Interior

In order to investigate the impact of temperature upon LCA alkenone distributions, the UK37, UK3738, UK38 and UK38 indices were regressed against seasonal temperatures and mean annual air temperature. No clear relationship was observed between these variables suggesting that temperature may not be the primary factor influencing these indices. This contrasts with previous studies that have suggested a significant relationship between UK37 and summer surface water temperatures in German lakes (Zink et al., 2001).

Given these discrepancies. a new approach is to develop a calibration based upon in-situ water samples instead. Here, they reconstruct LCA distributions from a single site at 0m, 5m and 10m depth during a number of months. During this interval, temperatures span a wide range (2 to 22°C) and intriguingly, the UK37 index is significantly correlated with lake water temperature (r2 = 0.75). The application of UK37(which removes the C37:4 alkenone) yields a lower correlation (r2 = 0.14). Although the slope of the Zink et al. 2001 core-top and Toney et al. 2010 in-situ calibration are similar, the absolute values are very different. Despite this, the in-situ approach appears a promising approach and may resolve some of the uncertainties in the core-top calibration related to taphonomy, growth depth and/or seasonal biases.

tonney2.png
Figure 3: UK37 vs temperature for NGP (red) and NE Sandhill (blue) lakes (Toney et al,. 2010). This data is compared to the calibration of Muller et al (1998; marine) and Zink et al., (2001; lacustrine). Also shown in green are lacustrine datapoints published in Zink et al,. 2001. Image redrawn from Toney et al., 2010.

 

Conclusion:

Within the North American Interior, alkenones were found in ~25% of samples lakes and appear to cluster into two distinct distribution based upon their region. Using a variety of parameters, the authors conclude that alkenones predominate in sites which are relatively cold, brackish-to-mesosaline and alkaline and where sulphate and sodium ions are abundant. Intriguingly, the C37:4 alkenone – previously a marker for lacustrine systems – was absent from several lakes. As the ecology of haptophytes in lakes complicates our application of traditional core-top temperature proxies, Toney et al. (2010) propose a promising new in-situ temperature calibration based on LCA in lakes, but which now requires further testing by the community.

References:

Boon, J.J., et al. 1978. Organic geochemical analyses of core samples from Site 362, Walvis Ridge, DSDP Leg 40. In: Initial Reports of the Deep-Sea Drilling Project.

Brassell, S. C., et al. 1986. Molecular stratigraphy: a new tool for climatic assessment Nature. 320. 129-133.

Brassell, S. C. 2014. Climatic influences on the Paleogene evolution of alkenones.Paleoceanography 29. 255-272.

Castañeda, I. S., and Schouten. S. 2011. A review of molecular organic proxies for examining modern and ancient lacustrine environments. Quaternary Science Reviews 30. 2851-2891.

Cranwell, P. A. 1985. Long-chain unsaturated ketones in recent lacustrine sediments. Geochimica et Cosmochimica Acta 49. 1545-1551.

De Leeuw, J. W., et al. 1980. On the occurrence and structural identification of long chain unsaturated ketones and hydrocarbons in sediments. Physics and Chemistry of the Earth. 12. 211-217.

Epstein, B.L., et al. 2001. The possible metabolic role of C37 alkenones in Emiliania huxleyi. Organic Geochemistry. 32. 867-875

Pearson, E, et al. Distribution and significance of long-chain alkenones as salinity and temperature indicators in Spanish saline lake sediments. Geochimica et Cosmochimica Acta. 72. 4035-4046.

Prahl, F. G., and Wakeham, S.G. 1987. Calibration of unsaturation patterns in long-chain ketone compositions for palaeotemperature assessment. Nature. 330. 367-369.

Theroux, S, et al. 2010. Phylogenetic diversity and evolutionary relatedness of alkenone-producing haptophyte algae in lakes: implications for continental paleotemperature reconstructions. Earth and Planetary Science Letters 300. 311-320.

Thiel, V, et al. 1997. Unusual distributions of long-chain alkenones and tetrahymanol from the highly alkaline Lake Van, Turkey. Geochimica et Cosmochimica Acta 61 2053-2064.

Toney, J. L., et al. 2010. Climatic and environmental controls on the occurrence and distributions of long chain alkenones in lakes of the interior United States.Geochimica et Cosmochimica Acta 74. 1563-1578.

Volkman, J. K., et al. 1980. Novel unsaturated straight-chain C 37-C 39 methyl and ethyl ketones in marine sediments and a coccolithophore Emiliania huxleyi. Physics and Chemistry of the Earth. 12. 19-227.

Volkman, J. K., et al. 1980. Long-chain alkenes and alkenones in the marine coccolithophorid Emiliania huxleyi. Phytochemistry 19. 2619-2622.

Zink, K-G., et al. 2001. Temperature dependency of long-chain alkenone distributions in recent to fossil limnic sediments and in lake waters.Geochimica et Cosmochimica Acta. 65. 253-265.

Zhao, J., et al. 2014. Occurrence of extended chain length C 41 and C 42 alkenones in hypersaline lakes. Organic Geochemistry. 75. 48-53.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s