# Raster analysis conundrum: resolution change and cell statistics in same step?

608
0
04-09-2014 09:31 PM
Highlighted
New Contributor II
Hello,

I need some help thinking my way through a raster analysis.  Any suggestions would be greatly appreciated.

I have a series of rasters, and I need to perform two operations:
1. Calculate the median for each pixel across all the rasters
2. Move to a coarser resolution (with pixels roughly 2x as big) using the median

I realize I could use cell statistics for step 1, and aggregate for step 2, but what I'd like to do is perform both operations in the same step. For example, if I have 10 rasters with pixels that are 1x1km, in one step I'd like an output of 1 raster with pixels that are 2x2 km, where the value of each pixel represents the median value across 40 pixels (4 pixels per raster to change the resolution by 10 rasters).  I guess to put it another way, what I'd like to do is find the median of a cube of pixels.

One idea I had for how to do this is to use the cell statistics tool, but under environments enter the coarser cell size.  This would accomplish both steps at once, however I'm unsure what statistic the tool would use to change the cell size.  I'm assuming it would probably use the mean, which wouldn't work in this case.

Alternatively, I have a separate raster with the correct pixel size.  Is there a way I could use this raster as a 'zone' to calculate statistics across multiple rasters? This would be like using the zonal statistics tool but across several rasters as opposed to one. 