Understanding climates of the past has become more and more crucial to appreciating the changes occurring on our warming planet today, and paleobotany is very important in providing baseline data to reconstruct past climates and in calibrating paleoclimate models based on physical parameters (Steppuhn et al., 2007). This area is rapidly expanding, so we will only cover a few of the many ways in which plant fossils can be used to reconstruct paleoclimate:
Data from fossil tree rings (paleodendrology) (FIG. 1.3) represent an important source of paleoclimate information, in some instances with very fine resolution, for example, major atmospheric disturbances (Miller et al., 2006). Although initially used for Holocene climate information (especially dating of archeological sites), some of the techniques used to analyze recent and subfossil wood have been extended to older material (Jefferson, 1982; Creber and Chaloner, 1984a; Creber and Francis, 1999; Taylor and Ryberg, 2007). Based on the changes in radial cell diameter within the tree rings and the variation in ring width (FIG. 1.3), it is possible to extrapolate climate information, which is especially useful when coupled with information from megafossils, microfossils, and the sedimentological record of the site. This approach has been utilized successfully by Parrish and Spicer in their work on Late Cretaceous floras from the North Slope of Alaska (Parrish and Spicer, 1988; Spicer and Parrish, 1990). More recently, Taylor and Ryberg (2007) have examined tree rings in Permian and Triassic woods from Antarctica. Based on their analysis using a variety of techniques, they suggest that the small amount of latewood indicates a very rapid transition to seasonal dormancy in response to decreasing light levels at these high polar latitudes. The mechanisms these plants evolved to cope with life in a polar light regime are of continuing interest in this and other studies based on plants that were once living at very high paleolatitudes.
The nearest living relative (NLR) method has been in use since the beginnings of paleobotany, particularly when dealing with late Mesozoic or Cenozoic floras, as these are more likely to have close living relatives. It is based on the premise that climatic tolerances of the fossils are very similar to those of their NLR. The paleobotanist compares as many fossils as possible within a flora to their most closely related extant taxa; the more species in a fossil flora that have NLRs, the more precise the paleoclimate estimate, and the more closely related a fossil taxon is to an extant one, the more precise the method. It depends, therefore, partly on the paleobotanist's ability to identify the fossils very accurately. The further back in time, the less effective this method is, as more and more extinct species or taxa which have no living relative appear. As a result, NLR has been used to best effect for Cenozoic angiosperm floras (Wolfe, 1995). This method can provide a general estimate of paleoclimate, but is limited by the fact that some fossil taxa do not have the same climatic limitations as their modern counterparts.
Leaf physiognomy analysis is a powerful technique that has been widely used in paleobotany to reconstruct Cenozoic paleoclimates. It is based only on angiosperms, however, so its applicability before the Cretaceous is uncertain (but see Glasspool et al., 2004a). Physiognomy is the general appearance of a plant, and it has long been known that plant physiognomy, especially leaf physiognomy, can be related to climate (Bailey and Sinnott, 1916). Physiognomy is primarily independent of taxonomy, for example plants with thick water-storing stems and leaves tend to grow in arid regions of the world, even though they may belong to a number of different families of plants. For fossil floras, this means that leaves do not have to be identified in order to obtain a paleoclimate signal. In his now-classic papers, Jack Wolfe (FIG. 22.276) presented the applications of leaf physiognomy to paleobotany, based on large collections of many leaves from extant floras, which he then was able to compare with Cenozoic angiosperm floras (Wolfe, 1993, 1995). Webb (1959) had previously completed a detailed physiognomic classification of Australian floras, and his definitions of leaf types are generally used in physiognomic methods today.
There are presently two methods of leaf physiognomic analysis that are in general use: leaf-area and leaf-margin analysis. Leaf area directly correlates with mean annual precipitation (MAP). CLAMP (Climate-Leaf Analysis Multivariate Program; Wolfe, 1995) measures 31 leaf character states of woody dicots (Chapter 22) and uses multivariate analysis to map leaf shape in two-dimensional space (Wolfe and Spicer, 1999). CLAMP can provide a number of climatic parameters related to precipitation, humidity, and temperature.
Leaf-margin analysis (LMA) relies on the relationship between leaf margin (toothed versus entire) and climate (Greenwood and Wing, 1995; Wilf, 1997). Specifically, the proportion of leaves in the flora with toothed margins can be correlated with mean annual temperature (MAT), as toothed leaves are more abundant in wet environments. Both CLAMP and LMA can provide quantitative reconstructions of past climates, including estimates of MAT and MAP. More recently, paleobotanists have refined physiognomic methods by using computer image analysis to analyze both leaf-shape and leaf-margin morphology (Huff et al., 2003; Royer et al., 2005). Further data on the ecophysiology of modern plants and the function of various leaf shapes (Royer and Wilf, 2006) will no doubt help to refine these methods and improve their accuracy in the fossil record. Both methods are very robust, as both rely on large databases of leaf physiognomy of living leaves from many different sites and habitats.
The stomatal index (the ratio of the number of stomata to the total number of epidermal cells plus stomata within a given leaf area expressed as percentages; see Salisbury, 1927) has been widely used in recent years to reconstruct past pCO2 levels, as the stomatal index is inversely proportional to atmospheric CO2 levels. Woodward (1987) was one of the first to demonstrate the value of this relationship for ancient climate prediction, based on comparisons of modern leaves with herbarium specimens from preindustrial times. The best results have been obtained from comparisons of the same genus and species in order to control for genetic differences, so younger fossils, such as Holocene plants, have provided reproducible results (Wagner et al., 2004). The technique has been extended further back in time, for example the Cenozoic (Royer et al., 2001), as well as to the Mesozoic and Paleozoic, although there are limitations to the technique, especially with older fossils (Roth-Nebelsick, 2005; Uhl and Kerp, 2005). For studies in deep time, researchers have coupled CO2 estimates from stomatal indices with other proxy records, such as isotope data (Beerling, 2005 and papers cited therein). A summary of the pros and cons of methods to reconstruct past levels of atmospheric CO2 can be found in Royer (2001) and Kerp (2002). Details regarding the stomatal index technique can be found in Beerling (1999) and Poole and Kürschner (1999).
Was this article helpful?