Erik van Sebille

Very good idea! A few places where we can do this 1) A notebook. Advantage is that it's a nice way to mix code, text and figures. It's also in...

Another thing to do in this updating of the performance advice is the following - [ ] Update `documentation_MPI.ipynb` notebook with performance curves for v2.1.3

Hi @angus-g, I see that this Issue has not been picked up in a while. Is it still a problem you're encountering? Have you found a solution/workaround? Something we can...

Very strange, especially because Parcels v2.3.1 doesn't even exist... See https://github.com/OceanParcels/parcels/releases and https://anaconda.org/conda-forge/parcels. So no idea where this `_version_setup.py` gets its version number from

Thanks for this question, @jporobicg. We stopped supporting the gitter channel last year; didn't realize there is still activity on it... Anyways, one reason could be that the velocities that...

Good to see that magnitude in your top plot does agree, that supports my hypothesis that the mismatch between 'U' and 'V' in the GRB1 input and the parcels-output is...

Thanks for reporting, @apatlpo. I don't think there is a fundamental reason for this not being implemented for xarray. When we developed this feature of time-varying depth grids, we only...

I just pushed a small fix to the PR; it turned out that the depth fields were never connected to the grid because the `Field.dataFiles is None` for Fields created...

Great to hear that this fix seems to work, @apatlpo! I've just pushed a set of fixes (one related to PEP8, the other to a bug in reading the `depth`...

> I also have a somewhat indirectly related question: could you please confirm using xarray does leverage its lazy capability (i.e. does not load full datasets in memory) ? Hmm,...