BBMM (v3.0) is an R package for modeling the movement path of an animal or object whenever continuous observation is impossible. The BBMM package implements a Brownian bridge movement model (BBMM) using discrete spatiotemporal data to obtain a probabilistic estimate of an animal’s movement path. The BBMM is commonly used to identify animal home ranges and migration routes. You can download the BBMM package for a 32-bit Windows OS below and install by opening R and selecting ‘Packages’ and ‘Install package(s) from local zip files…’. A re-start of R and/or a reboot of your computer may be required before using ‘library(BBMM)’ to load the BBMM package for use in the current R working environment. BBMM is available for both 32- and 64-bit machines on the CRAN website at www.r-project.org.
The software routines referred to in Chapter 9 of Handbook for Capture-Recapture Analysis are now an official package for R. These routines are available from here.
There are 2 ways to install:
– Download the “windows binary” zip file from CRAN, then from within R, choose “Packages” – “Install from local zip file”. When prompted, navigate to the downloaded zip and select okay.
– Start R. Assuming you have a live internet connection, select ‘Packages’ ‘Install packages’ from the R menus. A dialog box asking you to choose a nearby CRAN site may appear. If so, select one. Another dialog box should appear listing all the packages available for R. Scroll down to ‘mra’ (list is alphabetical, and long), select ‘mra’ and click okay. The binary zip will then download and install automatically.
After installing MRA using either of the above methods, you must load the library prior to accessing the routines. Load the library by issuing the command ‘library(mra)” at the R command prompt. If you do not take this step, R will not be able to find the routines. After installation and loading, type “help(mra)” to view detailed documentation.
This intuition builder is designed to let users play with the Evidence of Absence statistical model to build an understanding of how carcass counts, detection probabilities, and the credible bound work together to arrive at a fatality estimate. The inputs are simple and intuitive and assume no prior knowledge of the EoA statistical model.
escapeMR estimates total salmon carcasses produced in a stream (i.e., escapement) after
weekly searches to tag carcasses. Carcass characteristics such as length and sex are measured and can be used
to improve estimates.
This zip file contains Windows programs for the iterative Mayfields (MAYITER.EXE) and maximum likelihood estimation (BROODMLE.EXE) of survival rates as used for the papers below. When the programs are started the format for the data is described and two example sets of data MAYITER.DAT and BROODMLE.DAT are also in this zip file.
Schmutz, J.A., Manly, B.F.J. and Dau, C.P. (2001). Effects of gull predation and weather on survival of emperor goose goslings. J. Wildl. Manage. 65: 248-57.
Manly, B.F.J. and Schmutz, J.A. (2001). Estimation of brood and nest survival: comparative methods in the presence of heterogeneity. J. Wildl. Manage. 65: 258-70.
GenEst is an R software package for estimating bird and bat fatalities at wind and solar power facilities. The graphical user interface available here is identical to the one that ships with the GenEst R package (available at CRAN), and is suitable for production use in developing bird and bat fatality estimates.
The GenEst and Evidence of Absence fatality estimators both include a detection reduction factor (k) that describes how searcher efficiency changes through successive searches. Under some circumstances, the overall detection probability depends strongly on the value of k, and it may be worth estimating k in the field, and under other circumstances, overall detection probability does not depend strongly on the value of k, and it is likely feasible to make an assumption about the value of k. This app helps users explore the dependence of overall detection probability on the parameter, k.
Estimating probability of occupancy has recently gained popularity as wildlife conservationists and managers have started using occupancy based summaries to indicate species well-being and distribution. However, most estimates of occupancy probability are naïve because they do not account for imperfect detection probability. Patch occupancy modeling will account for imperfect detection probability and hence more accurately predicts probability of occupancy. pom (v 1.1) is a R package that allows users to fit a patch occupancy model. This R package, co-developed by Fawn Hornsby, Ryan Nielson, and Trent McDonald at Western EcoSystems Technology, Inc., allows the user to specify covariates for probability of occupancy as well as probability of detection. The pom package also has the ability to fit a beta-binomial mixture model, which allows the probability of detection to vary across sites and visits.
RRS is defined as the ratio of offspring per hatchery parent spawning naturally to offspring per wild (natural spawned) parent spawning naturally (Hinrichsen 2003, Araki et al. 2008). Computer programs written in the statistical computing language R are provided to estimate RRS and the precision of the estimate. One of the programs is designed to correct for bias in the estimate of RRS when the data contain a large number of sampling zeros. The other two programs provide guidance on sample size and statistical power when the null hypothesis is RRS is equal to one.
Documents included in the download provide background information, instructions on how to run the programs, and sample data.
The package RT (Randomization Testing) was originally written by Dr Bryan Manly to accompany the first and second editions of his book Randomization, Bootstrap and Monte Carlo Methods in Biology that was published by Chapman and Hall in 1991 and 1997. A third edition of the book is available but the original RT package is no longer available for purchase and some parts of the package do not work on modern computers. For this reason most of the important programs in the package have now been compiled to work with 32 bit and 64 bit Windows operating systems and these are now freely available for downloads from this website.
Make a directory with a suitable name on your computer.
Download the file Read-Me.pdf into the directory. This contains information about the 11 RT programs now available and 13 data files that can be used to test the programs.
Download the file RT-HProg.zip into the directory and unzip it into thhe same directory. This will provide you with all of the RT heritage programs and data sets.
To start a program just click on the name.
The SDraw Shiny app draws spatially balanced samples from uploaded shapefiles. SDraw implements the following:
Halton Iterative Partition (HIP)
(Robertson et al., 2018; ) samples,
Balanced Acceptance Samples (BAS)
(Robertson et al., 2013; ) samples,
Generalized Random Tessellation Stratified (GRTS)
(Stevens and Olsen, 2004; ) samples,
Simple Systematic Samples (SSS) samples, and
Simple Random Samples (SRS) samples from point, line, and polygon resources.