This is the GitHub hosting of the project GuadalShiftR. The paper associated to the project is published in the journal Biological Conservation. See the CITATION file for a BibTex entry to the article. This folder contains the files needed to reproduce all the results of the project, and compile the manuscript of the associated paper.
This project was conducted by:
· Pablo Almaraz (see contact below), which conceived the study, designed and conducted the analyses, and led manuscript writing.
· Andrew J. Green, which participated in study conception and participated in manuscript writing.
The major goal of the project is to explore the existence of tipping points, a catastrophic bifurcation and alternative stable states throughout a 36-year period of wintering waterfowl community dynamics in the Guadalquivir marshes, SW Spain, induced by the explosion of Mt. Pinatubo, Philippines, in 1991.
Ecological modeling has been traditionally dominated by a focus on the asymptotic behavior, but transient dynamics can have a profound effect on species and community persistence. We show a strong non-stationary coupling of ecological drivers in one of the world's major Mediterranean ecosystems, Doñana wetlands, which is currently threatened by many stressors. Recurrent changes in precipitation fluctuations triggered sudden reorganizations in community trends and population dynamics of a guild of ten wintering waterfowl species during a 36-year period. An anomalously dry and cold transient period in the Northern Hemisphere, induced by the volcanic eruption of Mt. Pinatubo in 1991, prompted an abrupt shift to an alternative regime of fluctuating species densities. Most species did not recover previous values even though local weather patterns and large-scale climatic conditions returned to normal values. Although the dynamical stability of the community is similar in both regimes, structural stability declined: the probability of feasibility dropped across time due to depressed population densities at equilibrium. A stochastic cusp catastrophe model fitted to the time series data suggests that the spatio-temporal persistence of cold and dry conditions in the wintering areas, coupled with warm and wet conditions in the breeding grounds, modulated local ecological conditions and induced hysteresis through behavioral shifts to alternative wintering sites. Our study provides empirical evidence for the existence of a catastrophic bifurcation triggered by a tipping point in the dynamics of an imperiled vertebrate community, highlighting the relevance of history and multi-stability in explaining current patterns in biological conservation.
This is a workflowr project bootstraped by a suite of open-source tools.
A suite of R packages were used in this project. I am grateful to all the people involved in the development of these open-source packages. Run the following R command from within the project for producing a reference list of the packages used:
grateful::cite_packages(include.RStudio=T, cite.tidyverse=T,
out.format = "Rmd",
out.file = "rpackages",
bib.file = "rpackages-refs",
out.dir = file.path(getwd(), "analysis"))A list of these packages is placed at the end of this document.
This section will include the problems detected in the code and, potentially, in the published paper.
-
13/02/2024
- Updated code: the most recent release of the bayestestR package, 0.13.2, broke some lines of the first version of the code when using the
map_estimate()function. This is now corrected.
- Updated code: the most recent release of the bayestestR package, 0.13.2, broke some lines of the first version of the code when using the
This section shows how to reproduce the results of the accompanying paper. The code folder has the following structure:
.
├── functions.R
├── SSRDLVR_model.JAGS
└── MAKEFILE.R
In this folder, the file ./code/functions.R contains all the functions necessary to conduct the analyses. The file ./code/SSRDLVR_model.JAGS contains the state-space regime-dependent Lotka-Volterra-Ricker model (SSRDLVR) developed in the accompanying paper written in the JAGS language.
The data folder has the following structure:
.
├── Aerial_count_data.csv
└── Environmental_data.csv
The output folder has the following structure:
.
├── BDFA_model
│ └── BDFA_results.Rdata
├── Cusp_fit.Rdata
└── SSRDLVR_model
├── plots.Rdata
├── SSRDLVR_model_results_PostPinatubo.Rdata
└── SSRDLVR_model_results_PrePinatubo.Rdata
The manuscript folder has the following structure:
.
├── biblio.bib
├── main_text.pdf
├── main_text.tex
├── processed_figs
│ ├── Common_Trends.pdf
│ ├── Cusp_diagram.pdf
│ ├── Environ_clim_ts.pdf
│ └── NOAA_CAR.pdf
├── supp_mat.pdf
└── supp_mat.tex
Prior to reproducing the results, make sure to have installed all the necessary software. In particular, you need JAGS, Stan and R. The R libraries needed to reproduce the results (see below) will be automatically installed by the package librarian.
The SSRDLVR is fitted through Bayesian MCMC methods using Gibbs sampling, and runs in JAGS: even though JAGS is written in the C++ language, the code can take several hours to run depending on the architecture used.
You can reproduce the results of the accompanying paper with three methods:
-
The first, easiest way to reproduce all the analyses in the project is to use the
Makefile. With simple GNU Make syntax, you can reproduce all the project, from statistical analyses to manuscript production. For example, in GNU/Linux based systems, you can point with the command shell to the project folder and run the following command:make all
This command will first conduct all the statistical analyses in the project, and produce all the figures. It then will assemble and compile the manuscript and associated supplementary materials with the necessary figures. Finally, it will open the files. Alternatively, note that you can run this command within RStudio from the Terminal tab.
-
From within
R, simplysourcethe filecode/MAKEFILE.R. This will perform all the analyses of the paper in the required order. -
The final method is to open the R Markdown file
analysis/Reproduce.Rmdto interactively execute the workflow.
In all cases, read the WARNING at the beginning of code/MAKEFILE.R or the analysis/Reproduce.Rmd file!
Distributed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. See LICENSE for more information.
Pablo Almaraz - @palmaraz_Eco - pablo.almaraz@csic.es
Personal webpage: https://palmaraz.github.io/
Project Link: https://github.com/palmaraz/GuadalShiftR
We used R version 4.3.2 (R Core Team 2023a) and the following R packages: bayesplot v. 1.10.0 (Gabry et al. 2019; Gabry and Mahr 2022), bayestestR v. 0.13.1 (Makowski, Ben-Shachar, and Lüdecke 2019), librarian v. 1.8.1 (Quintans 2021), MASS v. 7.3.60.0.1 (Venables and Ripley 2002), numDeriv v. 2016.8.1.1 (Gilbert and Varadhan 2019), parallel v. 4.3.2 (R Core Team 2023b), reshape2 v. 1.4.4 (Wickham 2007), rmarkdown v. 2.25 (Xie, Allaire, and Grolemund 2018; Xie, Dervieux, and Riederer 2020; Allaire et al. 2023), rstan v. 2.32.5 (Stan Development Team 2024), runjags v. 2.2.2.1.1 (Denwood 2016), tidyverse v. 2.0.0 (Wickham et al. 2019), workflowr v. 1.7.1 (Blischak, Carbonetto, and Stephens 2019), running in RStudio v. 2023.6.1.524 (Posit team 2023).
Allaire, JJ, Yihui Xie, Christophe Dervieux, Jonathan McPherson, Javier Luraschi, Kevin Ushey, Aron Atkins, et al. 2023. rmarkdown: Dynamic Documents for r. https://github.com/rstudio/rmarkdown.
Blischak, John D, Peter Carbonetto, and Matthew Stephens. 2019. “Creating and Sharing Reproducible Research Code the Workflowr Way.” F1000Research 8 (1749). https://doi.org/10.12688/f1000research.20843.1.
Denwood, Matthew J. 2016. “runjags: An R Package Providing Interface Utilities, Model Templates, Parallel Computing Methods and Additional Distributions for MCMC Models in JAGS.” Journal of Statistical Software 71 (9): 1–25. https://doi.org/10.18637/jss.v071.i09.
Gabry, Jonah, and Tristan Mahr. 2022. “bayesplot: Plotting for Bayesian Models.” https://mc-stan.org/bayesplot/.
Gabry, Jonah, Daniel Simpson, Aki Vehtari, Michael Betancourt, and Andrew Gelman. 2019. “Visualization in Bayesian Workflow.” J. R. Stat. Soc. A 182: 389–402. https://doi.org/10.1111/rssa.12378.
Gilbert, Paul, and Ravi Varadhan. 2019. numDeriv: Accurate Numerical Derivatives. https://CRAN.R-project.org/package=numDeriv.
Makowski, Dominique, Mattan S. Ben-Shachar, and Daniel Lüdecke. 2019. “bayestestR: Describing Effects and Their Uncertainty, Existence and Significance Within the Bayesian Framework.” Journal of Open Source Software 4 (40): 1541. https://doi.org/10.21105/joss.01541.
Posit team. 2023. RStudio: Integrated Development Environment for r. Boston, MA: Posit Software, PBC. http://www.posit.co/.
Quintans, Desi. 2021. librarian: Install, Update, Load Packages from CRAN, “GitHub,” and “Bioconductor” in One Step. https://CRAN.R-project.org/package=librarian.
R Core Team. 2023a. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/.
———. 2023b. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. https://www.R-project.org/.
Stan Development Team. 2024. “RStan: The R Interface to Stan.” https://mc-stan.org/.
Venables, W. N., and B. D. Ripley. 2002. Modern Applied Statistics with s. Fourth. New York: Springer. https://www.stats.ox.ac.uk/pub/MASS4/.
Wickham, Hadley. 2007. “Reshaping Data with the reshape Package.” Journal of Statistical Software 21 (12): 1–20. http://www.jstatsoft.org/v21/i12/.
Wickham, Hadley, Mara Averick, Jennifer Bryan, Winston Chang, Lucy D’Agostino McGowan, Romain François, Garrett Grolemund, et al. 2019. “Welcome to the tidyverse.” Journal of Open Source Software 4 (43): 1686. https://doi.org/10.21105/joss.01686.
Xie, Yihui, J. J. Allaire, and Garrett Grolemund. 2018. R Markdown: The Definitive Guide. Boca Raton, Florida: Chapman; Hall/CRC. https://bookdown.org/yihui/rmarkdown.
Xie, Yihui, Christophe Dervieux, and Emily Riederer. 2020. R Markdown Cookbook. Boca Raton, Florida: Chapman; Hall/CRC. https://bookdown.org/yihui/rmarkdown-cookbook.