Generating percentage cover by value type, within a polygon

[scra88le]

If I have a raster that contain cells with integer values that represent categorical variables (say 1 to 10), and I'd like to know the percentage cover for each value type, from a subset of the raster as determined by an sf polygon, how should I go about doing this? I'd like to return a vector containing percentages (one for each value type), for a given bounding polygon. I thought it would be something along the lines of:

exact_extract(raster, poly_sf, function(value) count(value)/no_of_cells_in_poly_sf)

But I'm not sure how to extend this to return a vector for the % cover for each value type. Thank you

[dbaston]

Does this return what you're after?

exact_extract(raster, poly_sf, function(value, frac) {
  tapply(frac, value, sum) / sum(frac)
})

[scra88le]

Thanks! I will try it. I ended up using:

      function(values, frac) 
        data.frame(v = values) %>% 
          group_by(v) %>% 
          tally()
      )

[scra88le]

When I run this, as the function works through the set of polygons I get this warning:

(EPSG:NA)Polygons transformed to raster CRS (EPSG:NA)Polygons transformed to raster CRS (EPSG:NA)Polygons transformed to raster CRS

Is there anything I should be concerned about here?

The raster CRS is:

+proj=tmerc +lat_0=49 +lon_0=-2 +k=0.9996012717 +x_0=400000 +y_0=-100000 +ellps=airy +towgs84=446.448,-125.157,542.06,0.1502,0.247,0.8421,-20.4894 +units=m +no_defs

The poly CRS is:

+proj=tmerc +lat_0=49 +lon_0=-2 +k=0.9996012717 +x_0=400000 +y_0=-100000 +ellps=airy +towgs84=446.448,-125.157,542.06,0.15,0.247,0.842,-20.489 +units=m +no_defs

They are very slightly different. Is that the reason?

[dbaston]

Yes, the warning is because the CRS are different. It doesn't look significant.