Design and implement end-to-end OGDC prototype

[mfisher87]

Pick some simple-ish dataset that doesn't really provide special challenges, this is just a step above "hello world" through the entire planned infrastructure.

• Develop a recipe that cleanses / fixes up input data
  • QGreenland use case: pipeline branch that outputs analysis-ready data optimized for QGreenland's offline use case.
  • PDG use case: pipeline branch that outputs cloud-optimized 3D tile data for PDG's online visualization use case.
• Deploy recipes to execute on some ADC infrastructure (as a Pangeo Forge Bakery?)
• Write outputs to DataONE
  • QGreenland use case: download analysis-ready outputs and compile in to QGIS project
  • PDG use case: access 3D tile outputs by HTTP requests to display in online viewer.

[mfisher87]

Do we want to still consider this an epic, or include it in OGDC2A?