It seems like this should be relatively easy to accomplish, but I've been messing with this issue off and on for a couple days and have not successfully figured it out.
Problem: I have a large (size of lower 48 U.S.) 100m resolution raster dataset that has a raster attribute table (RAT). The data were created by someone else. One of the attributes in the RAT has the data values I care about, while the actual raster cell values are meaningless. I've been trying to figure out how to reassign the value of the attribute I care about to the actual raster cell value but have not been able to figure out how to do this. The reason I want to do this is that I have a fairly involved custom Python tool that someone else developed that will do the calculations I need, but unfortunately it doesn't have the ability to use RAT attributes. It only uses the raster cell assigned value which as I said in this instance is meaningless.
I tried a roundabout way of running the raster to point tool to create a point feature class with the target field being the attribute I care about, thinking I could then create a new raster from the resulting point feature class. However, the raster to point tool failed stating I didn't have enough temporary memory (I'm running an Alienware machine with 16 GB of RAM and a 1 TB data drive).
Do I have any options? I'm using ArcGIS 10, but if I could do what I describe readily in another package I'd be willing to try another way (I'm not married to ArcGIS).