R devel - Feb 2009 - [SoC09-Idea] Movement Ecology add-ons for adehabitat package

Hi all,
  here's a proposal for a SoC project on wildlife movement patterns
analysis.
A MSc student is available to anwser the call, of course. The idea has already 
been briefly discussed with the mantainer of the adehabitat package and the 
AniMov project community.


Short description
Develop some add-on functions to use the adehabitat package to perform basic 
movement ecology analysis, in particular analysis of fractal dimension D. 
Developed functions could possibly be included in a future release of the 
adehabitat package itself.

Detailed description
Fractal dimension analysis is an useful tool to better understand not just 
the 'path tortuosity', in an animal's trajectory, but also the scale
level at
which 'search patterns' occur, i.e. an indirect index of the scale at
which
habtat resources are 'perceived'.
It has already been shown using simulations (Nams 2005; Oecologia 143: 
179-188) that the fractal dimension D of a trajectory changes within scales, 
and that the fractal dimension could be a valuable tool in multiscale 
analysis of wildlife movement patterns.
The adehabitat package, increasingly used as a free-open source alternative in 
radiiotracking and animal habitat selection studies, already has base 
functions to deal with trajectories in time and space, both as regular 
samples (e.g. data from GPS radio tags) and as irregular ones (as 
in 'classic' VHF radiotraking). The implementation of the algorithms
devised
by Nams could widen the range of analytical instruments that already make 
adehabitat an almost complete tool.
Moreover, the author of the sole existing software to do fractal analysis (V. 
O. Nams [0]) has expressed no objection towards such a porting (as reported 
by Paolo Cavallini in the AniMov mailing list [1]), suggesting that a 
complete rewrite could be a better solution.
In detail, the project objectives will be to port (or rewrite) into R 
the 'standard' methods offered by the FRACTAL program:
- fractal dimension estimation (D) using the basic divider method
- estimation of D the resampling divider method (Nams 2006: Acta Biotheoretica 
54:1-11)
- estimation of D using the VFractal estimator (Nams, 1996: Landscape Ecology 
11:289-297).
- production of other relevant movement ecology statistics useful in spetial 
scale perception by animals (Nams 2006: Animal Behaviour. 72: 1197-1203).
- optionally, in conjunction with the random movement simulators already 
present in adehabitat (in particular simm.levy and simm.crw), test for 
deviations from a CRW model (Nams and Bourgeois 2004: Can J. Zool, 
82:1738-1747) and detect if oriented movements occur (Nams 2006: Animal 
Behaviour. 72: 1197-1203)

Required skills
The candidate should be both familiar with R programming and wildlife 
radiotracking data analysis. A minimal familiarity with the adehabitat code 
could be desirable.

Mentor
Damiano G. Preatoni, PhD.
Research Associate
Unit? di Analisi e Gestione delle Risorse Ambientali
Dipartimento Ambiente-Salute-Sicurezza
Universit? degli Studi dell'Insubria
(see signature for details)

And, finally, the programming exercise [2]:
the candidate should demonstrate to be able to transform raw irregular VHF 
radiotracking data into an adehabitat ltraj object, which could be the base 
data structure to work with in the proposed package.


[0] http://nsac.ca/envsci/staff/vnams/Fractal.htm
[1] http://www.nabble.com/fractals-to21875728.html
[2] http://xkcd.com/74/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: This is a digitally signed message part.
URL:
<https://stat.ethz.ch/pipermail/r-devel/attachments/20090224/33013710/attachment.bin>