Favor ARCO datasets where possible

[brian-rose]

I tried to create a lot of content in this book that pulls from publicly available data, mostly via THREDDS servers. This has worked ok but does suffer from occasional outages. One example is climlab/climlab#177

Besides outages of THREDDS servers (including our own UAlbany-hosted server at http://thredds.atmos.albany.edu:8080/thredds/catalog.html), a bigger problem for interactive classroom use is the fact that THREDDS doesn't scale well to multiple simultaneous users.

A better alternative is cloud-native datasets served via public cloud storage buckets, or "Analysis-Ready, Cloud Optimized" (ARCO) data. A nice example of using ARCO-formatted ERA5 data is this Project Pythia Cookbook