ascii-grid export

yes barry I really need this.

I tried to use raster or rgdal but with poor results.

I have this function:

VGM_PARAM_A3 <- gstat(id="bos_bison",
formula=combusto~1,locations=~coord_x+coord_y, data=archezoology_table,
nmax = 10)

VGM_PARAM_A3 <- gstat(VGM_PARAM_A3, model=vgm(1, "Sph", 5, 0),
fill.all=TRUE)

ESV_A3 <- variogram(VGM_PARAM_A3, map=True, with=0.1, cutoff=9)

VARMODEL_A3 = fit.lmc(ESV_A3, VGM_PARAM_A3)

plot(ESV_A3, threshold = 5, col.regions = bpy.colors(), xlab=, ylab=,
main="Map - A3")
png("C:\Users\User\pyarchinit_R_folder\A3 semivariogram_map.png",
width=10000, height=10000, res=400)

I make a png file but how can I convert it in ascii-grid?

Why do you want to make an ascii-grid out of this? The variogram map
isn't in geographical coordinates, its in coordinate differences in x
and y

The return value when map=TRUE doesn't seem to be too well documented,
but looks like it is a list with a 'map' element that is a spatial
pixels data frame. Here's an example using the demo data (I can't run
your code because I don't have your data, please try and make your
problems easily reproducible):

require(sp)
require(gstat)
data(meuse)
coordinates(meuse)=~x+y
v=variogram(log(zinc)~1, meuse,map=TRUE,cutoff=900,width=10)
class(v$map)
[1] "SpatialPixelsDataFrame"
attr(,"package")
[1] "sp"

Now that can be written using rgdal's writeGDAL function.

However, you need the AAIGrid driver to work properly:

writeGDAL(v$map,"vmap.ag",driver="AAIGrid")

Error in .local(.Object, ...) : Dataset copy failed

raster package to the rescue:

require(raster)
writeRaster(raster(v$map),"v.asc","ascii")

class       : RasterLayer

[etc]

When I look at the file, I have an ESRI grid file:

$ head -5 v.asc
NCOLS 181
NROWS 181
XLLCORNER -905
YLLCORNER -905
CELLSIZE 10
[+ data]

Now, that's assuming you wanted to write the data, not a pretty image
picture like you get when you plot the variogram map. And there's
still the mystery of why you want to write a non-geographic
coordinate-based dataset to a geographic data format...

 And you should probably have asked this on R-sig-geo where the
geographRs (including the authors of gstat) hang out.

 Hope this helps anyway.

Barry