SIBER

Ellipse and convex hull fitting package to estimate niche width for stable isotope data (and potentially other relevant types of bivariate data).

MixSIAR is intended to encompass all the mixing model functionality in the now defunct SIAR package. Additionally, we have updated the basic mixing model from SIAR and released this as a standalone package for basic mixing model fitting as simmr.

Installation

The latest stable release package is released on CRAN as v2.1.9. Type install.packages("SIBER") in the command line.

Alternatively, you can install directly from github

The stable release can be installed by

  # install.packages("devtools") # install if necessary
  devtools::install_github("andrewljackson/[email protected]", 
  build_vignettes = TRUE)
  library(SIBER)

Release notes are available for each version.

The latest development version is on the master branch. Often this includes some new things that I am working on, but I can't guarantee that the package is stable and it might not install sometimes if I have broken something; usually though I tend to break things on a separate branch and try to keep the master stable as a package even if bits and pieces of the new stuff is not working correctly.

  # install.packages("devtools") # install if necessary
  devtools::install_github("andrewljackson/SIBER@master",
  build_vignettes = TRUE)
  library(SIBER)

Tutorials

The package vignettes have been expanded to provide working examples of the two main analysis types along with common sub-analyses. Several vignettes have been developed in response to Frequently Asked Questions.

Frequently Asked Questions (FAQs)

• How do I make isotope plots in ggplot? - see the vignette "Plot-SIA-ggplot2" included in versions >2.1.4
• Why are my community-level estimates of TA are zero or NA - This will arise if you have less than three groups (e.g. species) comprising a community. A triangle with three non-collinear points is the minimum requirement to draw a polygon and so if you have only one or two groups, the area of the TA is zero at best or possibly NA.

Help, Assistance and Queries

In the first instance, queries about analyses or problems with the software can be posted here on github. Please post a minimal worked examples so that we can recreate the problem and offer solutions.

Acknowledgements

Some code and much input from my collaborator and co-author Andrew Parnell. Thanks to Alex Bond for helping identify some problems in model fitting which is now resolved by z-scoring, fitting and back-transforming. Although not affecting every analysis, the potential issue is exemplified in SIBER-sandbox. Thanks to Edward Doherty for finding the bug that turned out to be in the creating of z-scores in createSiberObject.

Citation

Jackson, A.L., Parnell, A.C., Inger R., & Bearhop, S. 2011. Comparing isotopic niche widths among and within communities: SIBER – Stable Isotope Bayesian Ellipses in R. Journal of Animal Ecology, 80, 595-602. doi