[New Notebook]: Overlay NASA satellite data with OBIS occurrences

[MathewBiddle]

Programming Language

python

Module/Software/IOOS data service?

OBIS/ERDDAP

Module/Software/IOOS data service?

pyobis, erddapy

Provide a description.

This is a placeholder for creating a demonstration of what we can currently do with correlating OBIS occurrence data with NASA satellite data (like PACE).

Specifics are TBD.

[MathewBiddle]

I should be getting a brief description of this shortly. I'll post that here when received.

[MathewBiddle]

PACE data can be found at https://www.earthdata.nasa.gov/data/platforms/space-based-platforms/pace/near-real-time-data

I'm wondering if we could use the PACE PAR data as a proxy for phytoplankton, then do a taxonomic query for phytoplankton in the same region as a PACE grid.

Some thoughts:

1. Grab PACE PAR data
2. Generate a taxonomic query for plankton data in OBIS.
3. Grab OBIS data for PACE grid.
4. Make PAR and phytoplankton occurrence data dimensionless to highlight only magnitudes (where is there more PAR? Where is there more phytoplankton observed?)
5. Stick them on two maps side-by-side to see differences/similarities,
6. generate an anomaly map.
7. Describe what we learned from a data exercise. What PACE tricky to work with? What about OBIS? Anything make this impossible? Lacking data about something?

[MathewBiddle]

Access to PACE data https://fish-pace.github.io/hackweek-2025/presentations/notebooks/pace_earthdata_access.html

[MathewBiddle]

tinkered last week here: https://colab.research.google.com/drive/1b18ZnKku3ccLFjmw7ImIip0qo2cvBb8Q

[MathewBiddle]

published as a gist https://gist.github.com/MathewBiddle/7af3ce9b8938d660d5039bbd41df9c5c

Not rendering for some reason.

[ocefpaf]

published as a gist https://gist.github.com/MathewBiddle/7af3ce9b8938d660d5039bbd41df9c5c

Not rendering for some reason.

Looks like the widgets state was note saved in the notebook. I don't remember the menus, etc, but you need to find where you can save/embed the widget state. I'll try to take a screenshot later today.

[ocefpaf]

I'll try to take a screenshot later today.

This is the option you need to click to make sure widgets will get rendered in HTML represantations.

[MathewBiddle]

fixed it https://gist.github.com/MathewBiddle/7af3ce9b8938d660d5039bbd41df9c5c

[ocefpaf]

Matt, as we discussed offline yesterday we are missing a map projection in the notebook to align the two maps. This should do the trick:

import cartopy.crs as ccrs
import matplotlib.pyplot as plt
import matplotlib.colors as colors


norm = colors.LogNorm()

projection = ccrs.PlateCarree()
fig, (ax0, ax1) = plt.subplots(nrows=2, subplot_kw={"projection": projection})

gdf.plot(
    column="n",
    norm=colors.LogNorm(),
    legend=True,
    legend_kwds={"label": "OBIS Occurrences"},
    ax=ax0,
)


cs = chlor_a.plot(norm=colors.LogNorm(), ax=ax1)

We should probably make the images a bit larger for better visualization though.

[MathewBiddle]

Thanks! Much better!

[ocefpaf]

I was wondering...OBIS occurrences are square geometries, maybe we could extract the chlorophyll values for each squares and use that for plotting/correlations. Would that make sense?