Add Glider metrics to gliders.ioos.us?

[kbailey-noaa]

For consideration: I've just discovered on gliders.ioos.us, this link - https://gliders.ioos.us/metrics - that leads to a sad blank page. It used to be a bar graph of glider days broken down by RA.

Perhaps we could instead add a bar graph that grabs information from the glider DAC and tallies # glider days per year, starting in 2013? (I believe 2013 is the official "start" of the GDAC). And of any glider submitted to the GDAC, not necessarily RA-owned. In fact it would be quite difficult to tally RA-owned glider metrics, so I'd suggest we steer away from that for now.

I'm open to other thoughts on the types of glider metrics we could add here.

Overall this gliders.ioos.us page needs a major refresh, so this was one thing I was considering.

[MathewBiddle]

@kbailey-noaa Thank you for bringing this here. I've been struggling to figure out the right approach for glider metrics (see issues below).

I like the idea of tallying # glider days per year, I just want to be sure we are calculating those correctly. I think John has a script, I have a script, and @ocefpaf built a script too. Which one is "right" and easily deployable?

I'm sure we can figure out different ways to slice and dice based on the metadata available on the GDAC ERDDAP https://gliders.ioos.us/erddap/index.html.

• https://github.com/ioos/ioos_metrics/issues/7
• https://github.com/ioos/ioos_metrics/issues/4

Would you want to add the glider metrics to https://gliders.ioos.us/metrics? Or, add a new subpage on https://ioos.github.io/ioos_metrics/index.html?

If https://gliders.ioos.us/metrics, I can't seem to figure out where those docs are hosted to do any updates.

[ocefpaf]

I like the idea of tallying # glider days per year, I just want to be sure we are calculating those correctly. I think John has a script, I have a script, and @ocefpaf built a script too. Which one is "right" and easily deployable?

I've been running John's old gdutils but, b/c that is not packaged, it has become quite a challenge to ensure reproducible results there. I think we can do that with gliderpy. I'll try to create an example that uses both to see if we can replace one with the other.

PS: note that, in gdutils, John does much more than just glider days from a certain period/area, however, if that is all you need we can use a simpler notebook + gliderpy.

[ocefpaf]

@kbailey-noaa and @MathewBiddle this is the latest version of the notebook that could be an alternative to gdutils: https://nbviewer.org/gist/ocefpaf/90828d3c97db06e593759e32e49963db

Let me know what you think.