biodiversity obis-indicators

[7yl4r]

Hello! We have a project carrying forward from the NOAA IOOS Code Sprint which would be a good fit for this event. The project is focused around creating global biodiversity analysis using openly published data. We have written an R library here to set the foundation for this work.

Is anyone here interested? What could we do to move this effort forward at OHW2022?

[emiliom]

Based on my snooping around on Slack during the IOOS Code Sprint, that looks like a great project for OHW22! I think for now parking it here is a great first step. The group shepherding the projects process will start gathering soon-ish, and they'll provide more guidance. As far as seeing whether there is interest in your project, that will have to wait about a month, until we have selected the participants and started entraining them.

One suggestion I'd make after taking a quick look at your repo and comparing it to what the Code Sprint team shared on Slack is to add graphics. You have some great plots, but a visitor wouldn't know it when browsing the repo.

[7yl4r]

Thanks for the suggestion, @emiliom. I just added two pretty images to the README. I am excited to revisit this project.

[emiliom]

Great plots!

[albenson-usgs]

Following

[sophieclayton]

I'm wondering if it would be possible to add some code to partition the biodiversity indices by different taxonomic levels, e.g. by class or by order?

[MathewBiddle]

Hi @sophieclayton! I think that would be doable, but I'd be concerned about the amount of available data. It's already tough creating global maps with all the data. Trying to partition it down to class/order might result in less than useful information. With that being said, it doesn't mean it's not worth giving it a try (I could be wrong).

[sophieclayton]

Yeah, I was wondering about how the availablity of data could be a problem... but maybe we could build in check to flag up whether or not the data in a given taxonomic subset is sufficient?

[7yl4r]

To test this out would only require changing a few lines of code. Some colleagues of mine have suggested plankton may be a good subset, for example.

[MathewBiddle]

I would add another 💡 expanding this topic. From this ticket: marinebon/obisindicators#39

❗ Want to make a really awesome interactive graphic/website for OBIS ES50 indices?!?! 🗺️

Maybe something like this: https://glaroc.github.io/gbif_globe/ 🤯

🥳 This project could be your chance!! 🥳

ℹ️ The source code for that interactive graphic is available at https://github.com/glaroc/gbif_globe.

To present the ES50 indicator in that website, we think it should be as simple as creating the .jpeg images using the obis-indicators package and referencing them in these lines of code.

✔️ The project will need to:

1. create a fork of https://github.com/glaroc/gbif_globe
2. create the ES50 indices with all the obis data using obis_indicators (@albenson-usgs has been working on this but there might be ways to do it more quickly??),
3. do some data munging (because nothing is simple),
4. determine the appropriate coordinate reference system/projection for visualization tool,
5. determine the appropriate image size for the visualization tool,
6. output the .jpeg images as static files,
7. test interactive website by deploying locally,
8. iterate steps 3-7 until the graphic looks good (TBD),
9. serve website via GitHub pages,
10. UPDATE/CREATE DOCUMENTATION
11. (bonus) add different toggles to the website which are relevant to these data

@albenson-usgs @7yl4r please add anything I might have missed.

[albenson-usgs]

Code for 2 is available here 🙌

[MathewBiddle]

Note that this will be using

[maria-kuruvilla]

I am interested in this project!

[MathewBiddle]

@maria-kuruvilla thank you for your interest in this project! While we would be delighted to work with you on moving this forward, we just think there is not enough interest from OHW participants to give you the experience during the week that you should have (unless you would like to work with @lidefi87??). Since it was three mentors and only one participant, we recommend taking a look at the other project ideas and working with a group of OHW participants.

If, post-OHW, you have an interest in pursuing this activity, feel free to contribute to https://github.com/marinebon/obisindicators.

[lidefi87]

I am also interested in this project and we have two additional participants from the Spanish version (Nathalia Tirado and Cristian Jitala) that would also like to develop a version of this project in Spanish. We will be using data from the Galapagos

[MathewBiddle]

@lidefi87 That's great to hear! Feel free to make a fork of the obisindicators package under OHW and work with it. If you need help one of the OHW coordinators (or myself) can help you get set up.

[lidefi87]

It seems that the other two participants are more comfortable with Python, so we will be working on that language. But we will now use OBIS information instead of the data from Galapagos.

[NathaliaGLPS]

I am interested in this project

[Efeogis]

I am interested but do not have any background in R or Python

[emiliom]

Project repository, in Spanish / Repositorio para este proyecto, en español: https://github.com/oceanhackweek/ohw22-proj-biodiversity-indicators

[MathewBiddle]

¡Qué bueno!