Ann Marie Prue

Ann Marie Prue

Years participated in RESESS:

Download PDF »

An Overview

Major: Geology
Academic Affiliation: University of Wisconsin - River Falls
Research Mentors: Karen Chin and Garland Upchurch
Communications Mentor: Stephanie Higgins


Ann Marie is a second-year intern who recently received her B.S. degree in Geology from University of Wisconsin-River Falls. She will be attending Texas State University in the fall for her M.S. degree in paleobotany. Ann Marie has been interested in rocks and plants since a visit with her family to the Wisconsin Dells when she was four. Her dream is to become a professor of geology to help educate the next generation of geoscientists. For her second RESESS summer, Ann Marie was able to take on a project that combines her to passions of rocks and plants. Her project looked at climate predicting plant features in dinosaur-aged leaves from the Two Medicine Formation in Montana. Through this project, she learned that leaves with toothed edges are most commonly found in cooler climates; where as smooth edges are usually found in tropical climates. These features were used to estimate the temperature for the Two Medicine Formation as possibly ~ 70°F.


A Re-Analysis of the Late Cretaceous Flora of the Two Medicine Formation in Montana Using Digital Leaf Physiognomy

Plants are directly affected by the climatic conditions in their respective environments. Leaf architectural features that are useful for climate reconstructions include leaf size, type of margins, and venation. For over one hundred years, starting with the key research of Bailey and Sinnot (1915), scientists have worked to relate these features to climate. Several key analytical methods for analyzing extant and fossilized leaves have been developed. These methods include Leaf Margin Analysis (LMA), where the percentage of entire margin leaves is correlated with Mean Annual Temperature (MAT), and Leaf Size Index (LSI), where the percentage of certain-sized species provides a general climatic description. These two methods were the standard until a new multivariate method was developed, called Digital Leaf Physiognomy (DLP); this new approach utilizes computer image editing software and algorithms to measure different leaf features and can be used to predict both temperature and precipitation (e.g., Royer et al., 2005).

This study utilizes univariate methods of LMA and Leaf area Analysis (LAA), the correlation of leaf area to precipitation, as well as the multivariate method of DLP, to reassess the paleoclimate of an Upper Cretaceous site in Montana. The specimens for this study are fossil dicotyledon (dicot), flowering-woody plants, leaves from the early Campanian Two Medicine Formation collected near Cut Bank, Montana. Twenty-eight morphotypes were described and previously studied by Crabtree (1987) to infer a megathermal (tropical) climate based off of the out dated method of LSI. In contrast Crabtree’s (1987) LMA for MAT was estimated to be 7-10 degrees Celsius with no extended cold season because of his encountering of fossil palms during excavation. Also during the time of Crabtree (1987), there was no clear method for Mean Annual Precipitation (MAP) analysis other than qualitative, which he estimated to be high and wet.

By using Digital Leaf Physiognomy and other methods, we found that Crabtree’s (1987) method of LMA yielded cooler temperatures in comparison to the DLP and univariate regression models of by Peppe et al. (2011). MAP estimates were this flora were for the first time determined using DLP. In order to evaluate the credibility of these findings, they were compared with Peppe et al.’s (2011) analysis of fossil leaves from the Fox Hills Formation. Both results suggest that DLP gives a higher and more congruent MAT than LMA. The MAP methods of DLP and LAA, were inconclusive when looking for the most credible method.