[R-sig-Geo] Circular density estimate to a raster for mapping in GIS - help

[Jonathan Greenberg]

Raphael:

This sounds a bit like how one does skyview calculations with solar
models -- same type of output.  Can you simply have one raster layer
per angle, and then stack all of them?  The z-profile for each pixel
would then be the probability as a function of the angle for that
given pixel.

--j

On Mon, Mar 3, 2014 at 9:47 AM, Raphaël Lavoie <lavoie.raphael at gmail.com> wrote:
> Hi all,
>
> I created a circular density estimate using package "circular" from bearings
> (directions) from 0 to 360° assuming a von Mises distribution.
>
> #############
> library(circular)
> Bearing<-
> c(186,197,158,206,206,185,186,224,183,173,175,240,201,197,
> 175,173,226,227,233,193,175,217,171,193,190,186,222,185,
> 208,194,230,184,223,183,190,229,228,234,187)
> circular(Bearing, type=c("angles", "directions"), units=c("degrees"),
> template=c("geographics"))
> plot(Bear <- circular(Bearing, type=c("angles"), units=c("degrees"),
> rotation = c("clock"), zero=1.570796), ticks=T, stack=T, bins=360,
> xlim=c(-2,1), ylim=c(-2,1))
> mle.vonmises(Bear, mu=NULL, kappa=NULL, bias=FALSE,
> control.circular=list(units="degrees")) ## estimated (±SE)
> x <- rvonmises(n=39, mu=199.2, kappa=7.351,
> control.circular=list(units="degrees"))
> res25 <- density(x, bw=25, control.circular=list(units="degrees"),
> kernel="vonmises")
> circularp(res25$x)
> lines(res25, points.plot=F, xlim=c(-2,1))
> #############
>
> The output is probabilities that an object is going in a given direction
> (destination) from a specific location (origin). More specifically, there is
> a probability estimate for each direction (each 360 degree wedge).
>
> I would then like to create raster (0.5 x 0.5°) from that circular
> distribution with the center being the point of origin and the different
> directions displaying the underlying probabilities that an object is going
> in a direction (destination). Imagine a pie chart with 360 pie slices having
> each a specific probability. I tried to use the "raster" package, but can't
> figure out how to include the probability density to create the raster. Can
> anyone help?
>
> Thanks!
> Raph
>
>         [[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
>



-- 
Jonathan A. Greenberg, PhD
Assistant Professor
Global Environmental Analysis and Remote Sensing (GEARS) Laboratory
Department of Geography and Geographic Information Science
University of Illinois at Urbana-Champaign
259 Computing Applications Building, MC-150
605 East Springfield Avenue
Champaign, IL  61820-6371
Phone: 217-300-1924
http://www.geog.illinois.edu/~jgrn/
AIM: jgrn307, MSN: jgrn307 at hotmail.com, Gchat: jgrn307, Skype: jgrn3007