I’m attempting to use the netCDF format conversion tool to analyze L1C SMOS data (.DBL) in Python, but I’m noticing that the variables in the netCDF file after conversion are different than when I open up the original .DBL file in SNAP Desktop (in particular, brightness temperatures in SNAP are available in both X/Y and H/V polarizations, whereas the netCDF file only has two variables for brightness temperature, a real and imaginary value).
I’m a little new to working with this data- could anyone explain to me why this is? I’m not quite understanding from the documentation of the tool.
can you be more specific? Which L1C type are you using.
The tool should only reformat the data and the names should stay the same.
In section 5.2 of the user manual, the known variable names are listed per product type. Do you find the missing variables here?
Unfortunately, the experts for the SMOS toolbox are currently not available. They might give a better answer next week.
Thank you for your response! And absolutely no worries.
I’m using L1CS MIR_SCSF1C data. The variable names in the user manual for this data match those in the netCDF file I’m reading in Python. However, when I open the original .DBL file in SNAP Desktop, the list of data provided is:
There seems to be data associated with each variable, i.e. they’re not just empty categories. Is this just SNAP Desktop naming the data wrong? Am I opening it/looking at it wrong?
the variable names do not change, this is just an illusion
Due to the format and storage of the original SMOS data in the DBL/HDR data, SNAP needs to perform some processing when opening the data (i.e. mapping onto a raster, angular averaging, etc.).
Because this has to be done anyways, ESA requested that SNAP also calculates some variables “on-the-fly” which are useful to the users. These are the variables that you highlighted in your last post.
All Variables displayed in SNAP which are related to BT’s are calculated from those two real and imag variables using the polarisation of the snapshot.
