EW and IW images not aligned

[grantmac89]

I have processed two Sentinel-1 SAR images from the same date, one is an EW and one IW. I want to compare how a particular feature is represented in the two different modes, but they are misaligned significantly - by about 1.5km. The images are in Antarctica at -73S. I have processed both in SNAP as so:

Calibration
Speckle Filter (7x7 Lee)
Convert to dB
Ellipsoid Correction - Average height range doppler - project to UTM Zone 59S.

(The features of interest are in the sea ice, so do not do terrain correction)
If you focus on the island near the centre of the image/screenshot of the two images overlaid with transparency, can see the misalignment by about 1.5km. In both cases the feature of interest is close to the edge of the image.

[piyushrpt]

This is the same issue as

Range Errors between granules in S1-GRD captures

Hi, I’m trying to understand what correction (mathematically) still needs to be applied to GRD images in order to project them to a regular WGS84 such that points in a rendered tiff align between different granules. Note that in this case, I’m monitoring the edges of islands, which are (approximately, and certainly to a greater degree than would explain the error I’m seeing) at sea level so there should be no need for DEM-based RTC correction. Take, for example, these 2 images, which are small…

the difference in average height used to GRDs can be much higher due to the difference in swath widths.