[R-sig-Geo] About plotting maximum data over a region.-
Ashton Shortridge
ashton at msu.edu
Tue Aug 31 21:13:14 CEST 2010
On 2010-08-31, Freddy López, wrote:
> Dear ALL,
>
> Perhaps my problem does not belong exactly to this list... I apologize for
> this.
>
> My situation is the following:
>
> I have a dataset with 26 sites (columns) whose values are annual
> maximum since 1960 up to 1999 (40 rows) and I want to plotting it. Of
> course I have also the coordinates from each site.
>
> I have seen one example similar to what I am looking for (function
>
> simmaxstab() in SpatialExtremes):
> >filled.contour(x, y, log(data[,,1]), color.palette = terrain.colors)
>
> But at first glance it does not work to my data. The error that it report is:
> > filled.contour(coor2[,1], coor2[,2], nueva2)
>
> Error en filled.contour(coor2[, 1], coor2[, 2], nueva2) :
> increasing 'x' and 'y' values expected
>
> Can anybody help me to plot my particular dataset?
>
> Thanks in advance.
Hi Freddy,
Your problem is that filled.contour assumes a matrix input - essentially a
whole field of z-values. I suspect what you have are scattered locations with
z-values for each one. You have two options:
1. Interpolate your scattered points to a grid, and plot that, maybe with
filled.contour. Interpolation methods are very numerous in R and beyond the
scope of this answer, but maybe idw() in the gstat package would be a place to
start.
2. Plot the point locations directly, and color them according to their z-
value. Maybe something like this would work:
par(pty="s") # square plotting region
plot(coor2[,1], coor2[,2], main="Locations") # Just the locations
plot(coor2[,1], coor2[,2], main="Nueva2 at my Locations",
col=rainbow(n=10, start=0.50, end=0.65)[cut(nueva2,10)], pch=19)
points(coor2[,1], coor2[,2], pch=1, cex=1.1) # draws a black outline
Hope this helps,
Ashton
-----
Ashton Shortridge
Associate Professor ashton at msu.edu
Dept of Geography http://www.msu.edu/~ashton
235 Geography Building ph (517) 432-3561
Michigan State University fx (517) 432-1671
More information about the R-sig-Geo
mailing list