CODEWITHSUNDEEP
rpolygongeospatiallatticekriging
ckluss
ckluss

Reputation: 1271

Spatial interpolation (Kriging), polygon instead of raster output

Is there a (easy) way to convert a SpatialPixelsDataFrame (from krige) to e.g. a SpatialPolygonsDataFrame (vectorgraphic instead of pixelgraphic).

It would be fine to set value ranges and interpolate the raster to a polygon or use another krige method that generates a SpatialPolygonsDataFrame. I'm looking forward to a simple example.

krige example: i.e. in the book https://oscarperpinan.github.io/spacetime-vis/ 8.1.5 Spatial Interpolation, complete source https://github.com/oscarperpinan/spacetime-vis/blob/master/bubble.R

library(gstat)

airGrid <- spsample(NO2sp, type="regular", n=1e5)
gridded(airGrid) <- TRUE
airKrige <- krige(mean ~ 1, NO2sp, airGrid)

spplot(airKrige["var1.pred"],
       col.regions=colorRampPalette(airPal)) + ...

enter image description here

Upvotes: 1

Views: 1010

Answers (1)

Robert Hijmans
Robert Hijmans

Reputation: 47081

Something like this might work:

library(raster)
x <- raster(airKrige["var1.pred"])
y <- cut(x, c(10,20,30,40,50,60,70))
z <- rasterToPolygons(y, dissolve=TRUE)
spplot(z)

Upvotes: 2

Related Questions