[R-sig-Geo] gdistance: costDistance with barriers

[Robert J. Hijmans]

Karl,

You are using a default RasterLayer which has lon/lat coordinates. In
that case the earth is considered spherical (or similar), not a plane.
The maximum possible distance in longitude is 180 degrees, and the
distance between -120 and 120 is not 240 degrees, but 60+60= 120
degrees, .

To get the results I think you expected, use a planar CRS. For example
r <- raster(nrows=18, ncols=36, crs='+proj=utm +zone=10')

Robert

On Thu, Mar 12, 2015 at 11:35 AM, karljarvis <karljarvis at gmail.com> wrote:
> Hi all,
> I am seeing confusing behavior from the costDistance function in gdistance.
> In general, when cost is constant, cost distance increases linearly with
> actual distance. However, in this example, it is not doing that for the
> longest few distances. Am I missing something?
> Karl
>
> r <- raster(nrows=18, ncols=36)
> r[] <- 1
> t <- transition(r,function(x) 1/mean(x),4)
> p <- cbind(seq(-120,120,by=40),rep(0,7))
> costDistance(t,p)
>
>> costDistance(t,p)
>    1  2  3  4  5  6
> 2  4
> 3  8  4
> 4 12  8  4
> 5 16 12  8  4
> 6 16 16 12  8  4
> 7 12 16 16 12  8  4
>
>
>
> --
> View this message in context: http://r-sig-geo.2731867.n2.nabble.com/gdistance-costDistance-with-barriers-tp7587891.html
> Sent from the R-sig-geo mailing list archive at Nabble.com.
>
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo at r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo