[R-sig-Geo] adehabitatHR functions estUDm2spixdf & kernelUD

[Benjamin.Galuardi]

Hi Clare,

1) If you want a probability of occurrence within each cell, you should 
use the getvolumeUD function on the kud object in your script. Depending 
on the format of the fields you wish to compare, you might be able to 
omit the transform to spatialpixelsdataframe. In any case, the results  
are not a binned representations, which is what a 0/1 flag would 
constitute.

2) you are correct: grid=125 means a 125x125 grid.

Check out the 'raster' package as well for your comparisons. It's very 
good at display, overlay, grid math and export to other formats.

# once installed:
plot(raster(kud))

hope that helps!

Ben

On 4/9/2015 8:54 PM, Clare Death wrote:
> Hello,
>
> I am calculating the utilisation distribution (UD) for a group of mammals (n = 37), in order to define their individual exposure to an environmental contaminant. I need to know the meaning of the value returned in each cell of the converted UD output, as I need to multiply this by the contaminant levels in those cells. It is difficult to be certain what each cell value represents.
>
> I have two questions - thank-you for any assistance:
>
> 1) When using the estUDm2spixdf conversion: What does the function return for each cell? The probability that an animal will be found in the cell, the number of fixes per cell, the number of fixes per unit area, or a 1 or 0 flag to identify if the animal is present or absent from the cell, or something else?
>
> 2) Does the grid parameter of the kernelUD script reflect the cell dimensions of the total grid area (i.e. an arbitrary value for resolution of the pixels)? In the code below I�ve entered grid = 125. Does this mean that a grid is constructed where there are 125 cells in the horizontal dimension and 125 in the vertical (i.e. 125 * 125 = 15625 cells in total).
>
> Some example script:
> spat_dat.spdf <- SpatialPointsDataFrame(spat_dat[,6:7], spat_dat)
> kud <- kernelUD(xy = spat_dat.spdf[,1], h = "href", same4all = TRUE, grid = 125)
> kud.spdf <- estUDm2spixdf(kud)
>
> Kind regards,
> Dr. Clare Death
> Faculty of Veterinary and Agricultural Sciences
> The University of Melbourne, Australia
>
>
> 	[[alternative HTML version deleted]]
>
>
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo

-- 
==================================================
Benjamin Galuardi
PhD Student
NMFS-Seagrant Population Dynamics Fellow
SMAST, UMass Dartmouth
================================================


	[[alternative HTML version deleted]]