Determine 10-20 AOIs for determining necessary spectra

[pomadchin]

In terms of #70 there is a need input AOIs and AVIRIS scenes.

This task is to extract from the GeoPackage we have AOIs that potentially can be interesting for the further research.

The result AOIs would be used in the future to schedule AVIRIS scenes download and preprocessing #55.

[jpolchlo]

In order to get down to a final list of AOIs, I could use COGified versions of the scene IDs listed below. There are a lot (546) such scenes. I can surely reduce the number if that's useful. Probably get it to substantially under 200. LMK if that's a benefit.

As a side note, a Franklin query for these IDs returned a feature collection with the context line {'returned': 30, 'matched': 3741}, which if I'm reading right means there were almost 7 times as many hits returned as scene ids requested? Wonder if there's any clarification to be found on this point.

[list clipped]

[jpolchlo]

That long list was bothering me. I went back, picked only the largest 15 landfills per climate zone, and then selected only the most recent half dozen images per landfill. This pared the number of scenes way down. There are now a more manageable 106 scenes. If this seems too large still, I can easily trim the number.

Here's the list:

https://gist.github.com/jpolchlo/417bbe91042c0f3120fb75d265b3c9c1

[CloudNiner]

Looks great. We'll let you know when we're ready to ingest if we need to cut the list further.

Franklin query for these IDs returned a feature collection with the context line {'returned': 30, 'matched': 3741}

That seems wrong -- the 3741 is the total number of items in the collection. I'd expect the 'matched' number to equal or be slightly less (if it couldn't find some) than the number of items you pass in the list.

Can you post or DM me the code you used to do the query?

[pomadchin]

The next thing with this list of scenes is to kick the activation step. It is captured in the follow up issue #85.

Thanks!