How to get data on my own machine about esmvaltool_tutorial HOT 11 CLOSED

Not sure if the set-up page is the right place for that, though. Sounds like a big enough topic for a stand-alone episode on obtaining input data/installing synda

from esmvaltool_tutorial.

True, could be own episode. I tried to setup synda and was amazed by the number of steps required.

from esmvaltool_tutorial.

I think that installing synda might be a bit too much for this tutorial. Just going to an ESGF website and clicking the download button might be easier, e.g. https://esgf-data.dkrz.de/projects/esgf-dkrz/

from esmvaltool_tutorial.

We could then mention synda or esgf-python as more convenient alternatives, along with getting access to a cluster connected to an ESGF node.

from esmvaltool_tutorial.

True, could be own episode. I tried to setup synda and was amazed by the number of steps required.

I would recommend following the instructions here, it's a conda install and you have to fill in your credentials. The only thing that is not clearly mentioned there is how to get the esgf account needed to answer the credentials questions.

from esmvaltool_tutorial.

I think we should add download links for every single dataset that is used in the (basic)tutorial to the setup.md so people will not run into trouble during the tutorial.
Related to this is PR #100 where the download link is now inside the lesson. Also for episode 5 we need some data for which we should specify a download command to make the setup of this tutorial easier. We also need:

Exercise: Adding more preprocessor step

• dataset: HadGEM2-ES, project: CMIP5, exp: historical, mip: Omon, ensemble: r1i1p1, start_year: 1859, end_year: 2005 (var: thetoa & thetoaga)

Challenge : How to write a recipe with multiple preprocessors

• dataset: UKESM1-0-LL, project: CMIP6, exp: historical,
  ensemble: r1i1p1f2 (solutions mention var: tas &pr)

The following datasets are used in examples, so not necessary to complete exercises or challenges

Adding different datasets for different variables:

• {dataset: GPCP-SG, project: obs4mips, level: L3,
  version: v2.2, tier: 1}
• {dataset: HadCRUT4, project: OBS, type: ground,
  version: 1, tier: 2}

Creating variable groups:

• {dataset: CanESM2, ensemble: "r(1:4)i1p1", project: CMIP5}
• {dataset: MPI-ESM-LR, ensemble: "r(1:2)i1p1", project: CMIP5}
• {dataset: UKESM1-0-LL, ensemble: "r(1:4)i1p1f2", grid: gn, project: CMIP6}
• {dataset: CanESM5, ensemble: "r(1:4)i1p2f1", grid: gn, project: CMIP6}

from esmvaltool_tutorial.

See #103 about restructuring the setup.md to make it more concise and move all system specific instructions on how to find data to separate pages.

from esmvaltool_tutorial.

Maybe we can include a single wget script (generated from the ESGF datacart?) that downloads all the necessary files?

from esmvaltool_tutorial.

Maybe this could be interesting to link somewhere? It's a recorded webinar introducing ESGF and CMIP6

from esmvaltool_tutorial.

In PR #104 I added a wget command list of dataset.urls to download all data sets approx 2.9Gb.

from esmvaltool_tutorial.

Also to PR #104 a CI job was added to run the example recipe.

from esmvaltool_tutorial.