Interactive, reproducible, and accessible species distribution modeling for conservation with Wallace
- Andrew Gaier, Grad Center-CUNY ([email protected])
- Daniel F. López-Lozano, CBC-AMNH ([email protected])
- Bethany A. Johnson, CBC-AMNH ([email protected])
- Date: Wednesday, October 9rd
- Time: 1:00 - 3:00 PM EDT
- Location: Meet at Linder Theater (Please arrive 15 minutes early).
Species distribution modeling (SDM) is an important tool for conservation scientists, as it enables us to estimate present species range limits and make predictions about ranges for other areas and time periods. Advances in model-building and evaluation theory are common in the ecology and evolution literature. However, most cutting-edge methods are only accessible to those scientists who can read and write computer code, which results in a 'barrier to use' for many potential users. The Wallace ecological modeling application, implemented in the R programming language as the CRAN package wallace, provides a graphical user interface allowing any user to implement advanced SDM methods. Additionally, Wallace provides extensive guidance text and references key papers from the literature to help both new and experienced users learn best practices. Each model-building session can be exported as a fully documented R Markdown script file, to ensure reproducibility, ease of reporting, and for more advanced users, an easily modifiable code that extends Wallace's functionality. In this workshop, we will go through the basics of SDM using Wallace. We will demonstrate the key features of the modular software, show applications to conservation science, and specifically the utility of SDM outputs for formal IUCN assessments.
- 1:00 -- 1:10: Who we are & why we are here
- 1:10 -- 1:40: Introduction to species distribution modeling
- 1:40 -- 2:00: Introduction to R and Wallace
- 2:00 -- 2:10: Break / troubleshooting
- 2:10 -- 2:30: Wallace v2 live demo and walkthrough
- 2:30 -- 2:45: Using Wallace v2 with your own data
- 2:45 -- 3:00: Preview of new v3 features for conservation
Please take the pre- and post-workshop surveys. They really help us!
-
Before the workshop: Pre-workshop survey
-
After the workshop: Post-workshop survey
Please install R and Wallace prior to this Workshop. Downloading Wallace in R can take a while depending on your computer.
-
Install R (version ≥ v.3.5.0) Download for Windows or Mac We also recommend downloading Rstudio, which makes running R visually easier.
Alternatively, you can use Rstudio cloud with a Google or Github account. -
Install the 'wallace' package from CRAN or Github.
NOTE: Due to a temporary issue with CRAN (R's official package repository), if you have a Mac with an M1, M2, or M3 processor (you can find this out by clicking on the Apple logo in the upper left-hand corner, clicking on "About This Mac," and looking at the Processor information), you should install it from Github to get the latest version of Wallace.
- Open Rstudio.
- Enter the following code into the terminal and hit Run. This may take several minutes.
From CRAN...
install.packages("wallace")
library(wallace)
run_wallace()
From Github...
install.packages("devtools")
devtools::install_github("wallaceecomod/wallace@master")
library(wallace)
run_wallace()
During the workshop, we will allow you to choose your own species to work with during the demo portion. However, in the event of internet connectivity issues or issues with GBIF during the workshop, we have included a dataset you can work with here.
Canned Data
Includes:
- Occurrence data for two species
- Environmmental data at 2.5arcmin resolution
- a binary SDM
- a shapefile of Colombia
If you encounter a bug while using Wallace, you can report it here Bug Reporting Form