PALEOCLIMATE ESTIMATION FROM DIGITAL LEAF PHYSIOGNOMY OF FOSSIL LEAF FLORAS: PRELIMINARY RESULTS

Paleoclimate studies based on physiognomic analysis of fossil leaves have been widely used for several decades. Nonetheless, there have been few technical improvements to the principal methods of leaf-margin analysis, which utilizes only one character (presence/absence of teeth), and CLAMP, which includes 31 discrete characters. Huff et al. (2003, Palaios) developed a method that uses continuous, reproducible physiognomic variables. This technique, digital leaf physiognomy (DLP), has the potential to provide more accurate paleoclimate estimates because subjective and irreproducible discrete characters are replaced by continuous variables, generated mostly by image analysis using fixed algorithms. Recently, Royer et al. (2005, Am. J. Bot.) calibrated this technique using 17 extant floras along a transect through the eastern USA and Central America, and they also demonstrated potential for correlations of DLP to leaf ecological variables such as leaf mass per area and nitrogen content. Here, we show the first application of DLP to a fossil flora, the exquisitely preserved 52 Ma assemblage from Laguna del Hunco, in Patagonia (Chubut), Argentina. The aim of this study is to test whether the DLP technique can be used on fossil floras, since even the best-preserved assemblages are imperfect. Typical problems encountered include partial leaf and margin preservation and missing or damaged teeth. We present procedures for handling these issues quickly and reproducibly, along with preliminary paleotemperature results from DLP for the Laguna del Hunco paleoflora.