S1 radiometric correction

[ABraun]

I did a few tests because I still wonder about the best pre-processing parameters for Sentinel-1.

I therefore downloaded a Scene from 12.09.2016 which covers Cyprus (GRD and SLC). I chose this area due to its topographic complexity and the little vegetation cover.

S1A_IW_GRDH_1SDV_20160912T035113_20160912T035136_013014_0149C4_6E2F
S1A_IW_SLC__1SDV_20160912T035112_20160912T035136_013014_0149C4_CF51

I wanted to see the change in the result if different input data and modules are applied. These were the parameters:

1. Incidence Angle Image
2. GRD > RD Terrain Correction with checkbox “apply radiometric normalization”
3. GRD > Calibrate to S0 > RD Terrain Correction
4. GRD > Calibrate to B0 > Terrain Flattening > RD Terrain Correction
5. SLC > Calibrate to B0 > Deburst > Terrain Flattening > RD Terrain Correction

These are the results:
You can compare them best when you download them and skip through them (by Windows Preview/Slideshow, for example)
radiometricCorrection.zip (1.8 MB)

Incidence Angle Image:

GRD > RD Terrain Correction with checkbox “apply radiometric normalization”:

GRD > Calibrate to S0 > RD Terrain Correction:

GRD > Calibrate to B0 > Terrain Flattening > RD Terrain Correction:

SLC > Calibrate to B0 > Deburst > Terrain Flattening > RD Terrain Correction:

This raises the following questions?

• Does radiometric normalization in the RD Terrain Correction module replace/substitute Calibration? You can’t apply both. The result is however the very same.
• Shouldn’t actually Terrain Flattening somehow replace the topographic/radiometric normalization? The first relies on the illuminated area and the second relies on the incidence angle (see a nice comparison here)
• Why are there slight differences between terrain flattened SLC and GRD data?

[lveci]

Andreas, thanks for the comparison. Indeed the terrain flattening should replace the radiometric normalization. The normalization has been there since NEST. I’m not sure if there is any advantage to leave in the software multiple ways of doing similar things with new methods added in over time. I suppose each user could determine which works best for their particular scenario and data.
However, if something is clearly replaced by a better method then we should remove the obsolete method.

[ABraun]

thank you for your answer, @lveci!
I agree with you that, as long as something works, there’s no need for removing it from the software.

It makes sense that the normalization checkbox is a kind of relict from NEST but I’m still a bit confused about it:

Correct SAR calibration steps

Range Doppler Terrain Correction does not have a checkbox for “calibration”, but there is a selection for “radiometric normalization”, which is not the same thing calibration. Calibration is a purely radiometric operation, while terrain correction is a geometric one that affects radiometry via geometric effects.

How does this go along with my findings that processing 2 (normalization) and 3 (calibration) result in exactly the same image?

[lveci]

I believe it’s true you’ll get a similar results for S1 and RS2 because they use an LUT. There is an open issue to address this. Basically the local incidence angle isn’t being used for S1. It may make more sense to disable the normalization for S1 until the issue is addressed.

[Mr.Liu]

Hello，I have a question that I don’t know how to get the “Incidence Angle Image” by the snap just like your showing above .can you tell me how to get the Incidence Angle Image,and what tools I should be use.
thank you

[ABraun]

the incidence angle is present in the tie-point grids and can be used as such in the band maths.

Local incidence angle (SAR grid)

I don’t have a solution but maybe a hint: You can, in a first step, download the DEM in slant range geometry by right clicking on the band > Add Elevation band. You would then be able to derive it based on elevation and the incidence_angle tie-point grid somehow in the band maths operator but I don’t know the formula either. [image]

[samj]

@ABraun I am still confused, I have followed the following steps
GRD image - thermal noise removal- apply orbit file - calibration - filter- terrain correction

Is it necessary to check normalization option?

[lveci]

If you have already calibrated then no you don’t need the normalization in terrain correction. Normalization is essentially calibration with the local incidence angle.

[samj]

and if in a case we calibrated and also normalized it, then will it affect the output?

[eyeinsky]

Shouldn’t actually Terrain Flattening somehow replace the topographic/radiomatric normalization? The first relies on the illuminated area and the second relies on the incidence angle

The flattening just does radiometric corrections in radar geometry. Afterwards, if you want your result in map geometry, you just have to perform orthorectification (range-Doppler terrain-geocoding) to place those flattened backscatter estimates into the same map geometry as your DEM.