Shiny

This repository is the source code for interactive maps of PLS tree composition maps on the web using Shiny. You can access the live page here:

Shiny Presettlement Vegetation

Development

• Michelle Mueller - Notre Dame
• Jody Peters - Notre Dame
• Simon Goring University of Wisconsin

Contributors

• Chris Paciorek - University of California Berkely

Program Function

The program, as installed, loads the mean posterior output from Paciorek et al. (in prep) and displays both the mean estimate and standard deviation of the posteriors for 23 tree taxa in the Northwestern United States at the time of settlement. The raw data for the posterior runs comes from Goring et al. (in review) and Charlie Cogbill (REF).

Recent Changes

• v0.1 - Plotting enabled for both means & standard deviations simultaneously, sped up the initial data loading by pre-processing data, added color scheme to plotting output

TODO

• v0.2 - Clean up the folders, remove old code & data files. Add shapefiles on top of the mapped output. Create file output for mapped layers (to PDF or PNG) as well as data output (as CSV?). [This realease goes to PalEON Team].
• v0.3 - Incorporate changes from PalEON Team. Address Issue notes for v0.3.

Repo Organization

Once the repo is cloned on your desktop and the required packages are installed (shiny, ggplot2, gridExtra, RColorBrewer), the app can be run from the file RunShinyApp.R.

Shiny-App is the folder where the raw code lives. Critically, ui.R and server.R are the files that actually run the program. load_ncdf.R is a pre-processing script that loads the ncdf posteriors from Paciorek et al. and saves them as a long data.frame in Data\all_taxa.RDS.

Shapefiles for Canada and the US are in Data as well. There are older files in ExtraFiles and elsewhere that need to be cleaned up. Shapefiles should all be converted to native R format files for faster loading.