The Cenozoic history of palms: Global diversification, biogeography, and the decline of megathermal forests

  • Jun Ying Lim (Contributor)
  • Huasheng Huang (Contributor)
  • Alexander J Farnsworth (Contributor)
  • Dan Lunt (Contributor)
  • William Baker (Contributor)
  • Robert Morley (Contributor)
  • W. Daniel Kissling (Contributor)
  • Carina Hoorn (Contributor)



Aim: Megathermal rainforests and mangroves are much smaller in extent today than in the early Cenozoic, primarily due to global cooling and drying trends since the Eocene--Oligocene Transition (~ 34 Ma). The general reduction of these biomes is hypothesized to shape the diversity and biogeographic history of tropical plant clades. However, this has rarely been examined due to a paucity of good fossil records of tropical taxa and the difficulty in assigning them to modern clades. Here, we evaluate the role that Cenozoic climate change may have played in shaping the diversity and biogeography of tropical plants through time. Location: Global. Time period: Cenozoic, 66 million years ago to present. Major taxa studied: Four palm clades (Calaminae, Eugeissoneae, Mauritiinae, and Nypoideae), and their fossil pollen record. Methods: We compiled fossil pollen occurrence records for each focal palm lineage to reconstruct their diversity and biogeographic distribution throughout the Cenozoic. We use climatic niche models to project the distribution of climatically suitable areas for each lineage in the past, using paleoclimate data for the Cenozoic. Results: For most palm lineages examined, global pollen taxonomic diversity declined throughout the Cenozoic. Geographic ranges for each focal lineage contracted globally and experienced regional-scale extinctions (e.g., Afrotropics), particularly after the Miocene. However, climatic niche models trained on extant species of these focal lineages often predict the presence of climatically suitable habitat in areas where these lineages went extinct. Main conclusions: Globally, the decline in megathermal rainforest and mangrove extent may have led to diversity declines and range contractions in some megathermal plant taxa throughout the Cenozoic. However, while global climatic trends are an important backdrop for the biogeography and diversity of tropical groups at global scales, their continental or regional-scale biogeographic trajectories may be more dependent on regional abiotic and biotic contexts.,Please refer to the Methods section and Electronic Supplementary Material of the published article.,Reference: Lim, J.Y., Huang, H., Farnsworth, A., Lunt, D.J., Baker, W.J., Morley, R.J., Kissling, W.D. & Hoorn, C. (2022) The Cenozoic history of palms: Global diversification, biogeography, and the decline of megathermal forests. Contact: Jun Ying Lim (, Huasheng Huang ( Contents This data repository consists of the following files: This "Read Me" Word document containing details on each file ("ReadMe.docx") Fossil occurrence dataset ("PPRD_v4.3_28April2021.xlsx") R script to infer diversity through time ("pollenDivDyn.R") R script to plot diversity through time ("plotDivDyn.R") R script for data processing and cleaning of extant species occurrence data ("cleanGBIFocc.R") R script to fit climatic niche models to extant species occurrences for each focal lineage ("fitENM.R") R script containing utility functions that are called by the other R scripts ("utils.R") Palm checklist (presence-absence) data at scale of botanical countries ("palms_in_tdwg3.csv") Palm trait data ("PalmTraits_10.xlsx") Standardized dataset of geologic time intervals ("geologicScale.csv") To run the R script on your local machine, you will first need to specify your local working directory by modifying the "main.dir" variable in the R script. This will be the main folder. All R scripts should be put in a subfolder in the main folder called "src". Raw data will be need to be in a subfolder named "data" in the main folder. Subfolders for results ("results") and figures ("figures") will also need to be created for the results and figures to be exported correctly. Running the code To run this code you will need a current installation of R and the following R packages: `sp`, `rgdal`, `raster`, `maptools`, `ENMeval`, `ggplot2`, `cowplot`, `RColorBrewer`, `ggnewscale`, `deeptime`, `rgbif`, `CoordinateCleaner`, `divDyn`, `readxl`, `plyr`, `rnaturalearth`, `stringr`, and `spThin`. Some R package may require additional dependencies (e.g., `ENMeval`, `rgdal`); please refer to documentation for those R packages directly. Please feel free to get in contact if you have trouble running the script. Metadata Fossil occurrence dataset ("PPRD_v4.3_28April2021.xlsx ") The dataset contains information for each fossil occurrence and local stratigraphic range compiled from the literature, including metadata on source, geographic coordinates etc. A description of each of the columns is available in the spreadsheet. Palm checklist data ("palms_in_tdwg3.csv") This dataset was obtained from the Kew World Checklist of Selected Plant Species (; downloaded June 2015) and contains the following columns: Area_code_L3: TDWG level 3 (botanical country) unit (Brummitt et al. 2001 World geographic scheme for recording plant distributions. Intenational Working Group on Taxonomic Databases for Plant Sciences, Pittsburg, PA, U.S.A.; SpecName: Palm species name Palm trait data ("PalmTraits_10.xlsx") This dataset is the Palm Traits database v.1 (Kissling, W.D. et al. PalmTraits 1.0, a species-level functional trait database of palms worldwide. Sci Data 6, 178 (2019). An up-to-date version of this dataset and associated metadata may be obtained from This dataset was used to get a list of extant species for each of the focal lineages. Standardized dataset of geologic time intervals ("geologicScale.csv") Dataset of standardized geologic time intervals. This file is required to reproduce plots in the paper that include a geologic scale. Other required files but not in this repository Due to restrictions for open distribution, the following files are available on request. Some are readily downloadable using the links provided. Botanical country polygon shapefiles ("level3.shp" in R scripts) A shape file of TDWG level 3 / botanical country polygons (downloaded from: These polygons are used in conjunction with the palm checklist data to identify species occurrence records outside of the range of a particular taxon. GBIF occurrence data (referred to as "gbif_occurrences.txt" in R scripts) A download of species occurrence records from the Global Biodiversity Information Facility (GBIF; Metadata on the search parameters can be accessed using this link ( Kew occurrence data (referred to as "kew_occurrences.txt" in R scripts) A download of species occurrence records from the specimen database of the Royal Botanic Gardens, Kew.,
Date made available30 Dec 2021

Cite this