Didn’t we just hash this out on this post?
To back up a step, it’s important to note that the netCDF file format is not really a file format from the perspective of ParaView or other data consumers. NetCDF really just holds a collection of multi-dimensional arrays with attributes like a name. That alone is not enough for a ParaView reader to just know how to read the data.
So simply saying “I have a netCDF file” is not sufficient information for ParaView to read it. That’s like saying “I have an ASCII file.” That could mean lots of different things. You need to specify a convention of how the data are arranged. For example, an ASCII file could be a table with delimiters (i.e. csv) or a hierarchy with tags describing polygonal data (vtp). Thankfully, different ASCII file producers with different conventions tend to understand are each a format in its own right and identify the file as such. It kind of drives me nuts that so many netCDF producers don’t recognize this fact and just write out files with a .nc extension and expect consumers to magically understand what they mean.
Because there is this ambiguity on the format of data in netCDF files, there are many different netCDF readers to handle the different known conventions. So simply sending a screenshot is usually not enough information to determine what is going on. We usually need to have an example file or at a minimum the output of ncdump -h.
That said, on the previous post I thought we determined that the appropriate reader was the CDI reader (available through a plugin). That reader is maintained by DKRZ and you would have to go through them for help with this reader. @Niklas already reached out to you and offered help. What happened with that?