Nick Hodgskin

I have been trying to migrate v3 of Parcels to use Zarr v3, but have been struggling to do with minimal code changes due to the custom behaviour in `ParticleFile.write()`....

Zarr v2 is currently in "support mode" meaning no new features will be added to it. I don't know whether that means that at some point it will leave support...

> probably there is some way along the lines of [pydata/xarray@8abc993](https://github.com/pydata/xarray/commit/8abc993be1519571a570a023d993a1084a5078ae) to get around this using [pixi.sh/latest/advanced/override](https://pixi.sh/latest/advanced/override/) Exactly this - `distributed` pins the version of dask so you have to...

> It's a shame this doesn't happen automatically for packages installed from source like this Going out on a limb here - I disagree with this. `2025.11.1.dev2+gc7d9c55e5` should not satisfy...

@erikvansebille On the topic of performance, are you also experiencing it taking something like 10s occasionally to run `import parcels`?

Setting to draft until we have some actual benchmarks that we can include in this.

From meeting: We can use `tutorial_nemo_curvilinear.ipynb` as well

> My recommendation is _to forget about anything along the lines of "automation", "continuous", etc._ > > ASV provides a great way of defining benchmarks, running them against a bunch...

Another note: As part of this PR can we remove dependence on the `parcels/tools/timer.py::Timer` class? (which is currently used in a couple examples as simple benchmarking) Ideally once this is...

I think the suggestion by @fluidnumerics-joe is good, though I'm not entirely sure what exact the implementation would look like (partly since this is closely related to how we handle...