R packages

AHMbook

The AHMbook package is now on CRAN and can be installed with install.packages("AHMbook") .

makeJAGSmask

You will need R version 3.3.0 or later, and the secr package. If necessary, do

install.packages("secr")

Then you can install makeJAGSmask with the githubinstall package. If necessary, do

install.packages("githubinstall")

Then do

githubinstall::githubinstall("makeJAGSmask")

If you need to install manually, you can download the files:

source (TAR.GZ 122Kb) , Windows binary (ZIP 145Kb) , Mac (Mavericks or later) binary (TGZ 141Kb)

wiqid

The wiqid package is now on CRAN and can be installed with install.packages("wiqid") .

Statistical software for wildlife and ecological research often relies on the R programming language as a flexible foundation for custom analyses. Packages hosted on public repositories allow researchers to extend base functionality with specialised tools, while mirrors and binaries make installation straightforward across operating systems. Source tarballs, Windows binaries, and Mac builds are typically distributed so that users with different setups can obtain working copies. This approach keeps analytical pipelines reproducible, since every collaborator can run the same code and arrive at comparable results regardless of their preferred platform.

Capture recapture studies sit at the heart of many animal population investigations, and dedicated packages streamline the heavy numerical work these methods demand. Functions within such packages can calculate detection probabilities, estimate density across study areas, and produce confidence intervals around derived parameters. By abstracting the underlying likelihoods and optimisation routines, they let ecologists focus on study design and interpretation rather than low level coding. Bayesian extensions further enrich this toolkit by allowing prior information to shape posterior estimates through Markov chain Monte Carlo sampling.

Spatial capture recapture adds a geographical dimension to traditional models, accounting for animal movement and the arrangement of detectors across the landscape. Tools that build detection masks from habitat layers or distance sampling inputs help researchers translate field data into formats suitable for maximum likelihood or Bayesian fitting. Outputs often include activity centres, home range centres, and spatially explicit density surfaces that can be overlaid on maps. These visual products make it easier to communicate findings to managers and stakeholders who may not be familiar with the underlying statistical machinery.

Occupancy modelling offers another lens on wildlife monitoring, particularly when species are detected imperfectly or surveys cover only part of their range. By modelling detection and occupancy as separate processes, analysts can estimate the proportion of sites used while accounting for covariates such as habitat type or sampling effort. Camera trap data, track counts, and sign surveys all feed naturally into these frameworks. Recent extensions allow multiple detection methods to be combined within a single model, improving inference when no single survey technique captures every species present.

Workshops and supporting documentation help bridge the gap between published methods and everyday practice. Tutorials often walk through installing packages, preparing data, running models, and interpreting output, sometimes contrasting frequentist and Bayesian approaches side by side. Conjugate priors, Gibbs samplers, and JAGS or Stan back ends appear in many training materials, giving participants exposure to a range of computational strategies. Such shared resources strengthen the wider analytical community by lowering the barrier for newcomers and encouraging consistent reporting standards across studies.