topic Splitting raster cells. in Python Questions

Splitting raster cells.

TLeed — Mon, 04 Mar 2013 02:40:48 GMT

I have a raster with a resolution of .47-meters, it contains one field that defines the presence or absence of forest canopy (1 or 0).

I need to use a 5-meter fishnet, (or 5-meter raster) to average the coverage within each each cell. The 5-meter raster obviously doesn't line up perfectly with the .47-meter raster. How do I get the exact percent coverage of the underlying forest canopy coverage into the 5-meter raster or fishnet.

Doing a block statistic works, but when I run it with snap raster it changes the values to back to 1 or zero. I thought there might be a good solution using python.

I could also convert them to polygon features, but I'm surprised there isn't a good way to do it with 2 rasters.

Thanks in advance for your help!:confused:

Re: Splitting raster cells.

MathewCoyle — Mon, 04 Mar 2013 15:58:30 GMT

Could you use focal statistics with 5m map units and then just con to your 5m raster?

Re: Splitting raster cells.

curtvprice — Sat, 11 Dec 2021 22:23:26 GMT

I have a raster with a resolution of .47-meters, it contains one field that defines the presence or absence of forest canopy (1 or 0).

I need to use a 5-meter fishnet, (or 5-meter raster) to average the coverage within each each cell. The 5-meter raster obviously doesn't line up perfectly with the .47-meter raster. How do I get the exact percent coverage of the underlying forest canopy coverage into the 5-meter raster or fishnet.

The tool you are looking for is Aggregate. You can't get an "exact" aggregation - as .47 does not equally divide into 5 -- but it will be very close if you resample on the fly at a smaller cell size (.2) that equally divides into 5 meters.

from arcpy.sa import *
arcpy.env.cellSize = 0.2
# .2m * 25 = 5m (you could also use other values, say 0.1 and 50)
fiveMeterMean = Aggregate("p47rast",25,"MEAN")
# to get 0-100 percents:
fiveMeterPct = Aggregate("p47rast",5,"MEAN") * 100