A Quick Introduction to bbsBayes
Chapter 1 Customized BBS trends, trajectories, graphs, and maps using the bbsBayes R-package, an Introduction
1.1 What is the workshop about?
This is a 2-hour introductory workshop/demonstration of the R-package bbsBayes (https://github.com/BrandonEdwards/bbsBayes). This package allows anyone to apply the hierarchical Bayesian models used to estimate status and trends from the North American Breeding Bird Survey. The package also lets the user generate a suite of alternative metrics using the existing model output from the annual CWS analyses.
1.2 Intended Audience
Everyone is welcome! Some familiarity working with R is required if you’d also like to run the code yourself during the workshop, and the relevant packages should be installed beforehand (details will be provided with a full outline to be provided closer to the date). Those who might particularly be interested in attending include:
Anyone who might wish for a BBS trend for a particular region or time-period (e.g,. I want 15-year trends not 10-year trends, I wish I had a trend for just the western portion of the Barn Swallow’s range),
Anyone who might want to analyse the BBS with a customized model (e.g., the effect of weather, land-cover, etc. on trends). Do not worry if Bayesian analyses are new to you. We will not review the basics of Bayesian statistical analyses, but the package can be very useful without understanding those basics.
1.3 What we hope you will learn here
Download the raw data
Choose and run a model, geographic stratification, and species (we will not cover formal model-selection, just explore the different model-options)
Estimate trends for any time-period (1970-2019, 2005-2015, 1980-2000, etc.)
Graph population trajectories
Generate heat maps of population trends
Create geofaceted trajectory plots
Estimate trends for customized regions (e.g., All of the eastern Boreal, Great Plains vs eastern populations of grassland birds)