Terrain Flattening and Multilooking

Hoping forum users have some ideas…

I would like to process a stack of SLC IW images for basic classification on mountain glaciers. My processing flow thus far is:

-Remove Thermal Noise
-Calibrate to Beta Naught
-Single Image Multilook (I’d like to minimize the number of looks, just to achieve square pixels)
-Speckle Filter
-Terrain Flatten
-Terrain Correct

I am running into trouble upon Terrain Flattening, because the SRTM 1 sec DEM resolution is lower than that of the source image. I’d like to preserve as much detail as possible before further processing, and at least 16x4 range by azimuth looks are required to increase the pixel size accordingly.

If I skip Terrain Flattening and apply all of the steps to the Sigma Naught signal, the SRTM 1 sec DEM works fine for Terrain Correction, but it sounds like Terrain Flattening is a good idea for steep or complex topography.

Does anybody have any thoughts or recommendations for me? I’d greatly appreciate it.

maybe this helps:

Thank you, that was perfectly helpful.

Other examples of IW SLC processing include splitting, debursting than merging each swath. It appears that with just debursting, I am not required to split and then merge subswaths. Can you think of any issues or lost information I might incur by just debursting?

Thanks again for your help,