[spacewx]: Create DSCOVR notebook

[edmundhenley]

The challenge Develop the Oracle of DSCOVR is proposing taking a ~blind / low-level approach to predicting geomagnetic storms, to cope with DSCOVR degradation.

This challenge will be challenging!

This notebook will do following: (note will draw on a notebook I created last year: https://github.com/HarveySouth/spaceapps-2022/blob/main/notebooks/data_preparation.ipynb)

• [ ] add some code to do some of the low-level work, so people can get started faster
  • [ ] download the low-level spectral data
  • [ ] show how to read it into pandas
  • [ ] get a summary of values
  • [ ] plot an example spectra
• [ ] give some background on Kp and geomagnetic storms
  • [ ] cover how solar wind plasma and Bz affect Kp
  • [ ] crosslink to Lockwood work
  • [ ] show how to read in Kp & WIND data from the OMNI dataset
  • [ ] show what correlation of Kp with OMNI-WIND Earth-propagated Bz and plasma parameters is - that this temper expectations
  • [ ] discuss propagation time from L1 to Earth, and that also need to consider L1 halo orbits, spacecraft ephemerides
• [ ] give some background on DSCOVR anomalies
  • [ ] cross-link to the Loto'aniu paper on conditions where DSCOVR data is poor, and why, and when
  • [ ] suggest that anomaly investigation best be done by comparing with other in-situ data (e.g. WIND), not downstream (Kp)
  • [ ] mention SWFO - that there's life after DSCOVR!

[edmundhenley]

Whoops, forgot to add GH-N tags to my commit messages for these two commits:

• 50e8e29: Add DSCOVR notebook from 2022
• 55a2eba: Add conda environment files