I have a 0.5 meter raster that defines presence or absence (cells are either 1 or 0). I have another 5 meter raster, which I created. I would like to calculate a percent coverage of the underlying 0.5 meter raster and pull that value into the corresponding cell in the 5-meter raster. The rasters don't line up and cannot be adjusted.
One thing I now realize is that the cells are slightly less than 0.5 meters, .46999 to be exact. I ran block statistics, using a 10.638298 X 10.638298 rectangle (5/.469999), with snap raster set to, the slightly offset, 5-meter raster. That's taking a while to run, but I'm guessing that it won't work quite right. I also have a 5-meter polygon fishnet and I could also convert the .46999 meter forest cover raster into a polygon feature. Then I could calculate the area within each 5-meter 'cell'.
Best Answer
An easy way to do this is to use block statistics with a Mean rule. Specify a 10 by 10 neighborhood to get 5m blocks from your 0.5m data. Since your values are 0 and 1, the average values will equal percentages. To make sure it lines up with the 5m raster, change your environment settings so that your 5m raster is the snap raster (underneath processing extent) and the output cell size to 5m.