j-cleeland / sgat

Solar/Satellite Geolocation for Animal Tracking

Solar/Satellite Geolocation for Animal Tracking

SGAT (pronounced "tags backwards") provides facilities for estimating broadscale animal motions from archival or satellite tag data.

Installing

The package is easily built with RStudio

1. Install R

2. Install RStudio

3. Install Rtools or equivalent for your platform

4. Install devtools and roxygen2 packages and their dependencies in R.

5. Install sp, raster, coda, and maptools packages and their dependencies in R.

6. Clone the repository from GitHub (https://github.com/SWotherspoon/SGAT).

7. Create an Rstudio project in the folder containing this README file.

8. In the build tab, choose More/Configure Build Tools... and

   1. Select Use devtools if available, and

   2. Click Generate documentation with Roxygen, select Configure and select all six teckboxes.

9. Choose Build & Reload to make the package immediately available to R, or choose More/Build source package More/Build binary package from the Build tab to make source or binary packages.

TODO

• Alternate Coordinates. The Metropolis sampler may be more efficient if locations are represented in geocentric ecliptic coordinates. But it is unclear whether the gains in efficiency due to higher acceptance rates would outweigh the costs of the coordinate transformation and the added code complexity. This is a low priority.

• Parallelisation. At this point, the Metropolis samplers are only capable of utilizing a single core on a multicore machine. It would be relatively simple to introduce coarse grain parallelism by having the samplers draw multiple chains in parallel, using something like the multicore facility in the parallel package. Unfortunately, at the time of writing there does not seem to be a good parallization solution that works equally well on all platforms. This is a low priority.