SPE effect on cloud cover
Effect negligible. Further research determined only a small subset would likely have an influence (GLEs), not enough GLE events to do a good analysis. (only GLE event with DSCOVR EPIC data was in 2017, and the days around the event do not have any images).
Analysis was completed to some degree! graphs were calculated but no significant impact.
Also did some work with Ashley clustering some data from SWARM. Elbow plot suggested 4 classes was optimal for k-means clustering. The data involved as parameters was the IMF parameter and the three magnetic field parameters (20 samples of these data for each datapoint on the plot). This builds a picture of the places with a similar magnetic field parameter and relationship with IMF. Longitude is time-adjusted (i.e. degrees away from noon), as the magnetic field is influenced significantly by solar radiation.