Match-UP

[NickMortimer]

Title
Matchup - Matching up data from moving platforms to remote sensing

Summary
A common problem faced by many researchers using mobile platforms is to enhance point observations using remote sensing products.

Personnel

Specific tasks
Identify some use cases (it would be great to have at least one use case in each ocean!)
Identify sources of remote sensing data with coverage for the given matchup
Produce some recipes for matching up using a range of data sources for example ARGO floats to SST

Data sets and infrastructure support
http://imos-data.s3-website-ap-southeast-2.amazonaws.com/?prefix=IMOS/SRS/SST/ghrsst/L3S-1d/dn/
http://imos-data.s3-website-ap-southeast-2.amazonaws.com/?prefix=IMOS/Argo/dac/csiro/7900905/
http://imos-data.s3-website-ap-southeast-2.amazonaws.com/?prefix=IMOS/ANFOG/seaglider/Bicheno20110406/

The problem
It is often desirable to add value to insitu measurements by augmenting them with satellite products such as SST and Ocean Color.

Application example
Here is where you follow up with a location-specific example of where the data science methodology applies and list example datasets (size, format, how to access) that could be used for this exploration.

Existing methods

Proposed methods/tools
We propose developing recipes specific to desired data stores that fetch time and point coherent data from satellite archives.

[emiliom]

I suggest for this year you look up and connect with the OHW21 project "Sampling high-resolution model output as if by an in situ platform (ship, glider, mooring, etc.)". Kyla Drushka has continued it as a Python package; see oceanhackweek/ohw21-proj-model-subsampling#13. I tried to recruit Kyla for OHW22, but I think she'll be out at sea. I don't know if she's extended it to remote sensing, but she does work with remote sensing data in her research.

[emiliom]

BTW, I think Kyla even has some dedicated funding to work on her package (https://github.com/kdrushka/oceanliner). But it's currently focused on subsampling data from "high-resolution model output". Maybe this project could extend the package to satellite data types??

We love projects that build on previous projects 😍

[aurelaie]

This is a great project idea! And weaving it through a pre-existing project would be interesting too

[cathmmitchell]

Yesterday I had the exact same project thought (more specifically BGC-Argo and satellite chlorophyll). Great minds Nick

[sophieclayton]

@flane00 has been working on this for his MS thesis, using the OceanOptics getOC package

[cathmmitchell]

I'm familiar with that package (and have my own scripts that do something similar)... it's the Argo data part that's new to me

[NgoshaJr]

I am real much interested with this project, it sounds cool

[LauraGomezNavarro]

This package might be useful using satellite data instead of climatologies : https://pypi.org/project/oceansdb/ : "Package to subsample, or interpolate, climatologies like WOA to any coordinates."

[alisonrgray]

I had this same idea and would love to help out on this. Two thoughts:

• Would be great to make it cloud-enabled so that one does not need to actually download any satellite datasets
• Options for nearest neighbor and linear interpolation (in space) would be fantastic

[cathmmitchell]

Breakout Room Discussion Notes:

Paul & Nick CSIRO easyhub code would be a good starting point

Looking for tools to match model output with remote sensing, extract track from model to compare with remote sensing (rather than with field observation)

What remote sensing datasets are people working with?

Point data: Argo API, UNOLS flow-through data, match surface fluorometer satellite

What satellite datasets are ready for cloud usage?

Initial motivation for this project: IMOS data on AODN, connect point data that are AODN with satellite data on AODN

Possibility for expanding that initial motivation into something more general: any point data from any source, with cloud enabled satellite datasets??

[B-Stepin]

The conversation ended before I could pass on the Satellite data we use:
Aqua
NOAA-17,18,19
OrbView-2
Sentinel-1A,1B
Suomi NPP

[alisonrgray]

On the sign-up sheet, this project has only 1 participant listed (and me, as a mentor). Should we shelve it for this year and perhaps revisit next time? Or are there people on here who are motivated to work on this project?

[sophieclayton]

probably shelve it for now, unless someone is actively working on it this week.

[leewujung]

@alisonrgray: do you wanna ask on #ohw22_project to see if anyone is interested in this project?

@sophieclayton @NoraLoose: may be useful to ask at the end of the GVE project report out time today?

[leewujung]

@sophieclayton haha seems our messaged crossed pass each other in the air!