In this article we shall discuss how the analysis of the shapes of leaves can be used to reconstruct past climates.
== Leaf shape ==
In the previous article, we saw that if we know the environment preferred by a species, we can use it as a climatic indicator. We also noted that this is difficult when the species is extinct, since then we can no longer find by direct observation what climate is congenial to it.
But what if the morphology of a species was an indicator of the climate it prefers? In that case, we would be able to tell what climate it inhabited just by looking at its remains in the fossil record. One morphological feature commonly used for this purpose is the leaf shapes of plants.
Plants in temperate climates tend to have leaves with serrated margins, i.e. they have jagged edges; plants in warmer and more humid climates tend to have what are known in botanical jargon as entire margins, that is, smooth and unserrated. The difference is illustrated in the photographs to the right.
Rather than there being a sharp cut-off between the temperate and tropical styles of leaves, there is a continuous relationship between the climate and the mix of leaf types found in it: that is, as the climate gets a little hotter and wetter, the proportion of entire margins increases a little. This means that looking at a single leaf doesn't tell us that we are looking at a temperate or tropical climate; but looking at a whole lot of species will allow us to do something a whole lot better than simply dividing climates into tropical or temperate: we can actually estimate the average annual temperature.
So once we have established the relationship between climate and the proportions of leaf shapes in the flora, we can use this relationship to find out about past climates where leaves have been preserved in the fossil record. Although leaves are not the most easily preserved material, there are some environments, such as lacustrine sediment, where they are preserved and can be used as a paleoclimatic indicator.
== Other leaf characteristics ==
Using leaf shape as a climatic indicator began with counting the proportions of serrated to entire margins, but it didn't end there. Other characteristics such as leaf length, leaf width, leaf area, the presence or absence of a pointed tip also have a relationship to the climate.
The CLAMP (Climate Leaf Analysis Multivariate Program) group has measured many more characteristics of leaves and have used sophisticated methods to produce a formula relating these characteristics to climate; their website can be found here.
== How do we know? ==
We can establish by observation that the ratio of temperate to tropical leaf styles is a surprisingly good indicator of average annual temperature, as illustrated by the graph to the right, showing the relationship between floras and temperature in the forests of East Asia (figure from J. A. Wolfe, 1979, Temperature parameters of humid to mesic forests of eastern Asia and relation to forests of other regions of the Northern Hemisphere and Australasia, USGS Professional Paper, 1106).
This means that if we look at the ratio of leaf types in the fossil record, we can use it to produce an estimate of the temperature at the time when the plants grew.
Now, you might wonder if this relationship between leaf types and temperature really held in the past. Well, the adaptation of fauna and flora to their environment does seem to be one of the great constants of nature, so in principle we would expect the relationship to have held in the past much as it does in the present. It is after all extremely unlikely on the face of it that serrated margins would suit temperate climates in the present and yet have been more suitable to tropical climates in the past.
However, it should be noted that the exact relationship between leaf types and climate varies from continent to continent: although more serrated leaves always correspond to cooler climates on every continent, the exact numerical relationship is not the same. So a graph similar to the one above but based on Australian data rather than Asian data would still be a good approximation to a straight line sloping downwards to the right, but the exact angle of slope would be different. Now, what can vary from place to place could also in principle vary from time to time. This means that although leaf shape is a useful indicator of past climate, it would be naïve to treat it as though the relationship between climate and leaf shape was quite so exact as the relationship between temperature and a thermometer.