Clarification on preprocessing steps for Sentinel-1 RTC dataset

[bastiloeblein]

Hello Planetary Computer team,

I am currently working with data cubes that combine Sentinel-1 and Sentinel-2 data, and I am using the Sentinel-1 RTC dataset provided through the Microsoft Planetary Computer STAC catalog.

The dataset is described as Radiometric Terrain Corrected (RTC), and the spatial alignment with Sentinel-2 is excellent. However, for my analysis it is important to clearly understand which preprocessing steps have already been applied to the Sentinel-1 data and which have not.

I have reviewed the dataset page and STAC metadata, but I could not find explicit confirmation for some SAR-specific preprocessing steps. I would therefore appreciate an official clarification on the following points:

1. Thermal noise removal

   • Has thermal noise removal been applied to the Sentinel-1 GRD data before producing the RTC assets?

2. Border noise removal

   • Are GRD border noise artifacts already removed in the RTC product?

3. Speckle filtering

   • Is any form of speckle filtering applied as part of the RTC processing pipeline?
   • If yes, which filter (and parameters, if applicable)?
   • If not, can the data be considered “unfiltered” apart from radiometric terrain correction?

More generally, if there is documentation that explicitly lists the full preprocessing pipeline used to generate the Sentinel-1 RTC dataset in the Planetary Computer, a pointer to that would be very helpful.

Thank you for providing this dataset and for any clarification you can share.

Best regards,
Sebastian