Sentinel-3 OLCI TOA reflectance

Optical Toolbox
#1

hi, I’m trying to convert Sentinel-3 OLCI Level1 TOA radiance bands (S3A_OL_1_EFR) to TOA reflectance and export it as GeoTiff (using python and by using SNAP 5.0 GUI) and I have a couple of questions.

  1. As I understood from Sentinel-3 user guide, the data per pixel in the spectral bands (of S3A_OL_1_EFR) is TOA radiance. What is the scaling factor used for? because it seems that the data is originally in floating point.

  2. I was able to convert the data to TOA reflectance using: optical–>preprocessing–>Radiance to reflectance Processor. Does SNAP uses the formula found here to convert to TOA reflectance?

  3. I projected the data using Raster–> Geometric operation–>reprojection. Is there a way to do it using python script? I need to process many Sentinel-3 images and doing it using the GUI is not feasible.

I’m using SNAP 5.0 on win10 machine.

Thanks

#2

Hi,

probably you can find an effective reply to your questions in this thread:

http://forum.step.esa.int/t/best-practice-to-convert-and-reproject-sentinel-3-radiances-to-reflectance/5744

Read carefully what @marpet explains there, you can extend it to your need I think.

Bye

#3

Thanks, I’ll look into it carefully.

However, from what I read so far, I didn’t see explanation for the purpose of the “scale_factor” inherent in each nc file.
I also couldn’t find explanation in the Sentinel-3 user guide.

#4

The original data is not of type float. It is uint6. That’s why it needs to be scaled.

Yes, this is actually mentioned in the help of the processor.

  1. I projected the data using Raster–> Geometric operation–>reprojection. Is there a way to do it using python script? I need to process many Sentinel-3 images and doing it using the GUI is not feasible.

Not only via python but also from the command line:
Bulk Processing with GPT
For Python you can have a look here:

Also to the example scripts @antonio19812 provides can be helpful for:


In addition, several samples can be found in the forum