Historical Geology/Varves

In this article we shall examine what a varve is, how they can be used for absolute dating, and when they cannot.

The reader may find it useful to go back and re-read the article on glaciers before continuing with this article.

What is a varve?
In its original definition, a varve was a sedimentary feature in a proglacial lake, consisting of a couplet of coarse and fine sediment. Such varves are deposited in proglacial lakes annually because of the seasonal changes in the ablation of the glacier and the amount of meltwater feeding the lake.

Since then the definition of a varve has been extended so that it can be used to describe any layer which is deposited annually, the varves in proglacial lakes being only one example.

In this article we shall discuss varves in the wider sense, since they are equally good for absolute dating whatever the origin of the sediment.

Varves and absolute dating
Given such a situation, there is no difficulty in principle in finding the age of any varve; we just start from the one that was deposited this year and count backwards. (In practice there may be technical difficulties, but the principle is straightforward enough.)

Of course, this only works if there is still a source of sediment, so that we can identify this year's varve and know which year we're counting from. Once the source of sediment is cut off, the link with the present is severed, and unless we could find some other method to place an absolute date on one of the varves, the only thing we can tell from them is the difference in age between two varves, but not how old either of them is.

This limits the use of varves for absolute dating. We can count back thousands or tens of thousands of years, but as we shall see in later articles this is only a short span of time in proportion to the much longer history of the Earth.

Varves and cross-dating
We introduced the idea of cross-dating in the article on dendrochronology. The same principle can sometimes be applied to varves. In varves in proglacial lakes, for example, the thickness of the layers will vary yearly, as a hotter year will result in more melting, more outwash, and more deposition in the coarse-grained part of the varve.

However, it is difficult to use this principle as it is used in dendrochronology, to link together varve sequences in the present and in the past, and to associate varves in a lake currently being fed by glacial meltwater with a different lake in a different location where the glacier has melted entirely and the source of sediment has been removed.

Rather, the technique is more usually used on core samples taken from different locations in the same lake, in order to reduce error: if a minor disturbance has taken place in one location within the lake, removing one or more varves, then it is possible to use cross-dating with a sample taken from a different point in the lake to determine that this has taken place and to correct the chronology.

Varves: how do we know?
We can look in a lake or other environment of deposition, and see that varves are deposited on an annual basis; then we can take a core sample and see that the layer deposited in the past look just like those that are being deposited on an annual basis in modern times. It is an obvious conclusion that layers that look just like annually deposited layers were in fact annually deposited.

But what if the deposition has long ceased, and a succession of varves has been buried and lithified? How can we recognize them then? This question has various answers, depending on what sort of varves we're looking at. In classic varves, i.e. those found in a proglacial lake, the appearance of the sedimentary couplets is quite distinctive, and we may be fortunate enough to find striated dropstones, unambiguous evidence of a glacial depositional environment. In other cases it might not be so easy.

However, as we have pointed out, such ancient varves would usually not allow us to carry out absolute dating in any case, because they would have lost their link with the present: by counting varves we could determine the time interval between any two of them, but not the age of either of them.