Nussaïbah B. Raja, Andreas Lauchstedt, John M. Pandolfi, Sun W. Kim, Ann F. Budd & Wolfgang Kiessling
This repository contains all the data (in /data) and R scripts
(in /scripts) necessary to develop and evaluate a model to predict
extinction risk of corals in modern ocean using fossil data. The outputs
of the scripts are provided in the /output and figures in the /figs
folder. Please cite the study as:
Raja, NB, Lauchstedt, A, Pandolfi, JM, Kim, SW, Budd, AF, Kiessling, W. Morphological traits of reef corals predict extinction risk but not conservation status. Global Ecol Biogeogr. 2021; 00: 1– 12. https://doi.org/10.1111/geb.13321
This code was developed in R 4.0.0. It is therefore recommended to use
the same or any more up-to-date version of R for reproducing the
analyses in this study.
You will need to either use the Rstudio project environment or set your working directory to the root of this folder.
To install all required depdendencies (packages), run:
source(file.path("inst","dependencies"))The scripts/ folder contains all the code generated for the above
mentioned study. The folder contains 3 folders for each of the
models trained in the study and one additional script functions.R
containing custom functions used in this study.
-
01-model_fmm: This folder contains all required scripts to train and evaluate the fossil morphology model (trained using fossil data).
-
02-model_mmm: This folder contains all required scripts to train and evaluate the modern morphology model (trained using modern morphology data) and the predict the extinction risk of modern corals (using both the fmm and mmm).
-
03-model_mdmm: This folder contains all required scripts to train and evaluate the modern morphology and distributional model (trained using modern data).
The issue tracker is the preferred channel for bug reports. You may also contact me by email.