In this article we shall look at alkenones and how they can be used to construct the temperature proxy known by the somewhat cryptic name of Uk'37.
Alkenones are organic molecules having a chain-like structure, as organic molecules so often do. The length of the chain is given by the number of carbon atoms in it, so that we can speak of the alkenones C37, C38, etc. Note that these are not chemical formulas for the alkenones, which contain other atoms besides carbon; they merely record the length of the alkenone.
A further variation in the structure of alkenones is that they can be either doubly or triply unsaturated, a detail depending on the nature of their carbon bonds. So we can speak of the alkenones C37:2 and C37:3.
It is not really necessary for the reader to understand the organic chemistry of alkenones in any detail, or to know what "unsaturated" means; the important thing is that they exist and occur in different varieties.
Alkenones in nature
Only a small number of species are known to produce alkenones, all lying within the group of planktonic organisms known as haptophytes: the species Gephyrocapsa oceanica, the genus Chrysotila, and most importantly the incredibly numerous coccolithophore Emiliania huxleyi.
These alkenones have two useful properties. First of all, they survive conditions that would destroy most organic molecules. Consequently, alkenones have been found in marine sediment as much as 120 million years old. Second, the different varieties of alkenone are produced in different quantities at different temperatures. The relationship between temperature and the proportions of C37 alkenones is given by:
- T = 29.41 × Uk'37 - 1.15
where Uk'37 is the proportion of C37:2 and C37:3 which is C37:2, and T is the water temperature in °C; specifically, since all the organisms that produce alkenones are planktonic, T gives us the surface temperature of the water.
The longevity of alkenones in the sediment means that we can use them as a paleoclimatic proxy.
Limitations of the method
There are two main limitations on the method. Firstly, the relevant organisms don't grow in polar waters, and so obviously can't be used to indicate their temperature. Secondly, the formula breaks down for very warm water. For the proportion of C37 which is C37:2 can never be more than 100%, which means that the formula can never yield a T value greater than 28.3°C no matter what the actual temperature of the water is.
We should also mention a couple of potential confounding factors which have been discovered experimentally (see Prahl, Wolfe & Sparrow, Physiological impacts on alkenone paleothermometry, Paleoceanography, 18(2)). First, U37k' is reduced by conditions of low nutrition; second, it is increased by conditions of prolonged darkness.
How do we know?
We can measure the relationship between temperature and Uk'37 in oceanic plankton; we can also grow plankton in the laboratory at controlled temperatures (see, for example Prahl and Wakeham, Calibration of unsaturation patterns in long-chain ketone compositions for palaeotemperature assessment, Nature, 330, 367-369).
Can we be sure that this relationship held good in the past? It seems likely: if there is some reason why C37:2 is preferable in warm waters now, then why should it have been different in the past? But we could say this with more confidence if why knew why C37:2 is preferentially produced in warmer waters. As it is, biologists don't yet know why these haptophytes produce alkenones in the first place. Under these circumstances, perhaps we should be a little more skeptical of this biochemical proxy than of those based on known chemical and physical mechanisms.
Note on terminology
The subscript and superscript in Uk'37 can also be written as U37k', and when superscripts and subscripts are not available, people will write either Uk'37 or U37k'. The reader wishing to research the subject further by performing an internet search should be aware of this variation.