EW mode, Projection related

Hi all,

I have been working on EW mode S1 GRD medium resolution data over South pole.
During the pre processing step - post noise corrections, i used Ellipsoid Correction as my scenes are over ocean
My graph:
Apply orbit file
Remove border noise
Remove thermal noise
Calibration
Linear to dB conversion
Speckle filter
Ellipsoid correction:
Here, i selected - wgs/84- stereographic south pole as my projection whole geodetic datum as WGS 84.
But, when I’m calculating the same masked area in stereographic projection in arcmap its 25000, whereas its around 585000 in snap.
Can someone please tell me where am i going wrong?
Becasue, reprpjection (raster) doesnt seem to work. Through Radar-> geometric → EC (it shows its already projected).

I dont understand what’s causing this huge differnce.
Cam someone pelase help me with any steps im missing
Would be grateful!!!

This step is not clear to me. What do these numbers refer to and what kind of mask do you use? Maybe you can add screenshots.

Thanks a ton for responding.
I’m trying to mask out pixels (of a particular threshold - say water).
Not just pixels based, even geometry doesn’t seem to work.
Please find below
the screenshot from SNAP
image
(in box is the geometry I’m testing)

And, the same area when tested in QGIS - shows

I’m quite sure the area from QGIS is right.
But, I’m doing the rest of the processing in SNAP - and cannot switch now to QGIS/ArcGIS.
So, I badly need to get this sorted.

It certainly looks like a projection issue.
Please do the needful, TIA
Let me know if something is not clear here yet

Another example.
The area below in blocks is an area below 20000 sq km


The same area when masked in SNAP
image

This is what the geocoding info looks like @ABraun


But when imported in Arcmap:

Shows GCS_WGS_84 and nothing abt Stereographic Projection

Do you suggest me to opt for QGIS/ArcGIS?

A sidenote: Linear-to-dB should be the last step, otherwise you’ll mess up your image statistics.

So, applying dB conversion post-EC would work fine?
Also, in EC - as I’m working on Poles, I used GG (Geolocation Grid) - Is that the issue? for such huge areal inconsistencies with other sofwares?

@ABraun & @mengdahl - or if there is a way I can export my masked pixels into geotiff ?
I can see that I can export only as text file and not geotiff.

geometry is the sum of all vectors. If you want the areas of single shapes you have to put them into different vector containers.

I tried using different vectors too… What I noticed is - that the problem is only around 180E.
RS-1 (2)
image

Whereas, anything towards all East and all West is working perfecting fine.
Another thing I notices is- all the files are around Atarctica but 1,2,3 appear else where
image