S2TBX FAQs

What are the difference between products L2A and L2Ap?

L2Ap are the pilot products generated by ESA until March 2018. After March, they are operational products (L2A).

Why some bands are missed in L2A products?

Depending on the resolution selected for the Sen2Cor processing, there are some bands that are not generated:

At 60 m: the band 8 is not used and the band 10 is excluded as it does not represent surface information (the cirrus band).

At 20 m: B08 is not used and the three resampled 60m bands B01, B09 and B10 are omitted.

At 10 m: The 10m product processing covers only the four optical bands at 10m (B2, B3, B4 and B8). The other bands are omitted. If a 20 m processing has already been performed in ahead, these data will be used as input. Else, a 20 m resolution will be performed first, in order to have access to the input data.

For more information see the sen2cor user manual in http://step.esa.int/main/third-party-plugins-2/sen2cor/

What is the difference between the generic Resampling operator and the S2Resampling?

The S2Resampling operator resamples the Sentinel-2 products having into account the particularities of the angle bands.

Because of the discontinuity of the angles when the detector changes, the result of the resampling of the azimuth angle bands is not good from the physics point of view since it blurs the border between detectors when we use a bilinear method. (See a more detailed explanation in the SNAP help).

It is recommended to use the specific resampling only when the angle bands are going to be used later. Else, it is better to use the generic operator since the output is going to be similar and it is much faster.

What tools are available to carry out the atmospheric corrections of Sentinel2 L1C products?

The most used tools that can be used for the atmospheric corrections of Sentinel-2 Level 1C products are:

• Sen2Cor: http://step.esa.int/main/third-party-plugins-2/sen2cor/
  Sen2Cor is the "official" tool that is used by ESA to generate the level 2 product that can be downloaded directly from Scihub.

• MAJA: https://logiciels.cnes.fr/en/content/maja
  MACCS-ATCOR Joint Algorithm, result of a collaboration between CNES and DLR. It is able to work with time series.

• iCOR: https://blog.vito.be/remotesensing/iCOR_available
  "The atmospheric correction software, iCOR (previously known as OPERA), is now available to the broad user community through the ESA Sentinel Application Platform (SNAP) for the atmospheric correction of Sentinel-2 and Landsat-8  data".

• C2RCC: https://www.brockmann-consult.de/portfolio/water-quality-from-space/
  The main purpose of the C2RCC processor is the retrieval of Inherent Optical Properties (IOP) but it also performs an atmospheric correction over water. If you are only interested in water pixels then you can give it a try. It is included in the Sentinel-3 Toolbox.

Problems when trying to open an S2 product

Usually, the errors occurred when trying to open an S2 image are caused because the format of the product has been updated. When this occurs, we update SNAP in order to support the new format, so please, update SNAP (Menu->Tools->Plugins->Update) and try again.

It has been also reported some problems with some zip tools: some users have had problems when unzipping the products with WinZip. If you are using this tool, try to use another one (7zip for example) or to open directly the zip file.

What are the units of the Biophysical processor outputs?

Open

More information about the Biophysical porcessor can be found in the ATBD: http://step.esa.int/docs/extra/ATBD_S2ToolBox_V2.1.pdf

What bands are required by the Biophysical processor?

The biophysical processor is implemented as a neural network that requires the following bands in the input layer: B3, B4, B5, B6, B7, B8A, B11, B12, view_zenith_mean, view_azimuth_mean, sun_zenith and sun_azimuth.

What cloud masks are available in S2 products?

In L1C products there are some cloud masks:

Open

In the level 2A products, the cloud mask are more accurate:

• There is a scene classification identifying the cloud pixels

• There is also a band 'quality_cloud_confidence'

It exists also in SNAP the operator IdePix that can compute the cloud pixels of a L1C product.

How to remove clouds from a S2 product?

In the S2 products there are available masks indicating the cloud pixels. There are different ways to mask out these pixels in the bands:

• right-click on the band in the Product Explorer and write the condition in the Valid-Pixel expression

• Use the land/sea mask operator (Menu-> Raster->Masks->Lad/Sea MAsk) by using "Vector as Mask" and selecting the desired mask:

• Use Band Maths: you can create new bands using as expression for example: