TOPSAR SLC Data Deramping

Hello,

I’m trying to understand the processing flow of Sentinel-1 SLC data, particularly regarding the deramping of TOPSAR data:

  1. According to ESA’s ‘Sentinel-1 Level 1 Detailed Algorithm Definition’ handbook, the processing is divided into:
  • Pre-processing: Focusing raw data into an ‘internal SLC product’
  • Post-processing: Converting the internal SLC product into final L1 products
  1. My questions are:
  • Are the SLC products available through the Sentinel Hub already the post-processed SLC images, not the internal SLC products?
  • If they are post-processed, why is additional deramping necessary for users? The documentation indicates that the Doppler spectrum is already deramped during azimuth post-processing. However, in previous discussions in the forum, it was advised that TOPSAR SLC data should be deramped in SNAP prior to any resampling/interferometric operation.

Any clarification on these processing steps and when/why deramping is required is greatly appreciated!

The Sentinel-1 products you downloaded from Copernicus data space or other data hub are L1 product and ready to use.

The additional deramping is for Sentinel-1 TOPS SLC product only because the data in provided in bursts. In case that resample is needed (e.g. in interferometric operation), the data should be deramped first. But SNAP operator such as Back-Geocoding takes care the necessary deramping operation.

Thanks Jun for your answer.

However, in the ESA document (see below), specifically in 7.2 Post-processing Azimuth Processing (page 7-5), it states:

‘’ The post-processing azimuth processing consists of a number of steps (steps 1 to 13)…
2. For TOPSAR only, de-ramp the Doppler spectrum. This de-ramping is
achieved by applying the chirp signal defined in Section 5.2.1.‘’

There is no mention of a subsequent reramping operation, unlike section 6 where at multiple instances the spectrum is deramped, resampled or any other operation is perfromed, and then reramped (e.g. 6.4.1.2 - 6.4.1.4). This is the reason that led me to think that the L1 IW SLC products were already deramped.

If I may follow up with a related question, is there a function in SNAP that performs the inverse of ‘S-1 TOPS DerampDemod’ i.e. reramp/remod? This is performed in Back-Geocoding, but that only works for a pair of products, not individual ones.

The S-1 TOPS DerampDemod operator under Radar > Sentinel-1 TOPS performs deramp and demodulation for a given S-1 TOPS SLC product.

Hello Jun, do we have a generic version of the operator that could be used for TOPS and spotlight-modes too?

No, the current operator is for TOPS only. We should generalize the operator in future to include spotlight mode.

1 Like