Shane Blowes
The WhittakerBetaChange repository includes R code to reproduce the analyses shown in the article:
Local changes dominate variation in biotic homogenization and differentiation
by Shane A. Blowes, Brian McGill, Viviana Brambilla, Cher F. Y. Chow, Thore Engel, Ada Fontrodona-Eslava, Inês S. Martins, Daniel McGlinn, Faye Moyes, Alban Sagouis, Hideyasu Shimadzu, Roel van Klink, Wu-Bing Xu, Nicholas J. Gotelli, Anne Magurran, Maria Dornelas, Jonathan M. Chase
Alban Sagouis created new compilations of open access data used in these analyses. Code for the compilation can be found at:
https://github.com/chase-lab/metacommunity_surveys
https://github.com/chase-lab/checklist_change
https://github.com/chase-lab/homogenisation-richness
00-fit-models-maintext-code-for-eve: This folder contains scripts that were used to fit statistical models to data shown in main text (written for EVE (HPC))
00-fit-time-series-models: This folder contains scripts that were used to fit statistical models to time series data
01-processed-data: This folder contains the processed data that models were fit to, and some additional files required to identify regions to be included/excluded from particular sensitivity analyses
Fig2-1-wrangle.R: Code to wrangle model results for Figure 2
Fig2-2-plot.R: Code to plot Figure 2
Fig3-1-wrangle.R: Code to wrangle model results for Figure 3
Fig3-2-plot.R: Code to plot Figure 3
FigS2.R: Code to plot Figure S2
FigS3-and-S5.R: Code to plot Figure S3, S5
FigS4.R: Code to plot Figure S4
FigS5-1-wrangle-anti-ts-data.R: data wrangle for Figure S5 (datasets with only two points)
FigS5-1-wrangle-ts-data.R: data wrangle for Figure S5 (time series data)
FigS6.R: Code to plot Figure S6
Each subfolder has the code used to standardise sampling effort for different sources of biodiversity data at two scales: a smaller (alpha) scale, and larger (gamma) scale. Two sets of data (each documenting the alpha and gamma scales) are compiled: (1) effect sizes of the rate of change in species richness, calculated as the log-ratio of richness at the two time points, divided by the number of years elapsed between richness observations; and, (2) time series of richness and a diversity metric related to Simpson’s concentration.