Times are displayed in (UTC-06:00) Mountain Time (US & Canada) Change

79-8 Phylogenetic Informed Predictions of Climate from Leaf Size and Shape: Model Development and Application to Fossil Record

Session: New approaches to old fossil collections

Presenting Author:

Dana Royer

Authors:

Royer, Dana L.¹, Boyko, James D.², Peppe, Daniel J.³, Butrim, Matthew J.⁴, Hagen, Eric R.⁵, Lowe, Alexander J.⁶r> (1) Department of Earth and Environmental Sciences, Wesleyan University, Middletown, CT, USA, (2) Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI, USA, (3) Department of Geosciences, Baylor University, Waco, TX, USA, (4) Burke Museum of Natural History and Culture, University of Washington, Seattle, WA, USA, (5) Department of Ecology and Evolutionary Biology, University of Michigan, Ann Arbor, MI, USA, (6) UC Museum of Paleontology, University of California, Berkeley, Berkeley, CA, USA,

Abstract:

Paleobotanists use leaf size and shape (physiognomy) to reconstruct mean annual temperature (MAT) and precipitation (MAP), correlating a site-mean average of fossil leaf physiognomy to climate based on extant calibrations. For example, among woody non-monocot angiosperms, the percentage of species lacking toothed margins scales with MAT and leaf size scales with MAP. Current implementations of leaf-climate methods ignore taxonomic information, despite it being long known that phylogenetic signal is present in leaf-climate relationships. Here we propose a workflow for incorporating phylogenetic information into leaf-climate methods using phylogenetic generalized least squares (PGLS). We apply our workflow to Digital Leaf Physiognomy (DiLP)—a multivariate leaf-climate method based on traits that can be measured on fragmentary fossil leaves. PhyloDiLP has greater skill than standard DiLP at predicting climate at 92 globally distributed extant sites: the root mean square error for MAT is 3.4 vs 3.7 °C and for MAP is ln(0.52) vs ln(0.59) cm. We apply phyloDiLP to ten fossil sites and, promisingly, find that many of the climate predictions are more in line with independent assessments. For example, phyloDiLP predicts a warmer (16.8 vs. 10.9 °C MAT) and wetter (170 vs. 127 cm MAP) climate for the warm, everwet early Eocene Laguna del Hunco site in Argentina.

Phylogenetic Informed Predictions of Climate from Leaf Size and Shape: Model Development and Application to Fossil Record

Description

Session Format: Oral

Presentation Date: 10/11/2026

Presentation Start Time: 03:50 PM

Presentation Room: CCC, 103

Back to Session