[R-sig-Geo] Raster comparrison

[Ivailo]

On Mon, May 13, 2013 at 8:13 PM, ping yang <pingyang.whu at gmail.com> wrote:
> Dear Ivailo,
>
> Is this method can be used to quantify the spatial variability of two
> raster grids? Or Is there other measurement will be better to do this?

Dear Ping,

according to the author of the KappaV package (Vincent Bonhomme),
rasterizing the vectorial mosaics would be "neither elegant nor exact"
and that's why he implemented the calculation of the confusion matrix
(i.e. the matrix quantifying the mismatches in different regions of
the mosaic)  directly from the vector maps.

> Previously I posted a topic on how to compare two grids regarding to see
> the difference on spatial variabilities(e.g. two temperature grids). I am
> still wondering how to do it.

I am not aware of any R code to do such a comparison on raster
objects, and Cohen’s Kappa just quantifies the degree of agreement
between the two mosaics of interest. While writing this post I found a
software package at http://www.riks.nl/mck/ called the "Map Comparison
Kit" that might lead you further in your efforts.

HTH,
Ivailo
--
UBUNTU: a person is a person through other persons.