Skip to content
Modified HMM fish geolocation toolbox originally created by Pedersen
MATLAB R Shell
Branch: dev
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.

README.md

hmm_smast

hmm_smast is a MATLAB package for geolocating New England groundfish using Data Storage Tag (DST) data. This package is a fork of Martin W Pedersen's MATLAB geolocation toolbox, available at http://mwpedersen.dk/tracking.html (The link is dead as of August, 2017. See mwp_original branch for the original geolocation toolbox code).

Dependencies

This code requires a recent MATLAB installation (tested on R2015a and newer) with the Curve Fitting Toolbox.

This code requires t_tide to perform tidal elevation prediction and the Matlab Google Earth Toolbox to plot the most probable track in .kml format. These packages are included in the dependencies directory.

Documentation

View the documentation generated by m2html here.

Preprocessing

As an alternative to the preprocessing code in MATLAB provided in [test/preprocessing], an R code package for preprocessing raw DST data for the use of hmm_smast is available at https://github.com/cliu3/R_HMM_Preprocessing.

Publications

  • Liu, C., Cowles, G., Zemeckis, D.R., Cadrin, S.X, and Dean, M.J. (2017), Validation of a hidden Markov model for the geolocation of Atlantic cod. Canadian Journal of Fisheries and Aquatic Sciences, 74(11): 1862–1877. doi:10.1139/cjfas-2016-0376.
  • Pedersen, M.W., Righton, D., Thygesen, U.H., Andersen, K.H., Madsen, H., 2008. Geolocation of North Sea cod (Gadus morhua) using hidden Markov models and behavioural switching. Canadian Journal of Fisheries and Aquatic Sciences, 65(11), 2367–2377.
You can’t perform that action at this time.