This is the first survey of pulsating auroras which is differentiated by type. Pulsating auroras are found to be almost always an early-morning phenomenon and are almost entirely lacking persistent structuring before midnight. Long-lived patches which are known to move with convection primarily appear after midnight. These patches are a less common form of pulsating aurora and are found to originate from the inner magnetosphere, in agreement with past observations of their source region.

Pulsating auroras have recently begun to be reconsidered in the context of the conditions and mechanisms driving them. This research connects one type of pulsating aurora to specific plasma waves and uses in situ plasma observations to infer the source region of this aurora. Shortcomings of the available methods of associating in situ observations with specific auroral features are outlined and a key issue is identified which challenges our understanding of pulsating aurora formation.

The comparison of geomagnetic active and quite events of double substorm onsets responsive to IMF variations shows that the occurrence sequence of all required substorm signatures looks the same and not different for small and large Kp. Double substorm onsets responsive to IMF variations can be characterized with two-stage magnetic dipolarizations in the magnetotail, two auroral breakups of which the first occurring at lower latitudes than the second, and two consecutive Pi2-Ps6 band pulsations.

The solar wind reshapes Earth's magnetic field to create our magnetosphere and powers many dynamic processes in our plasma-filled environment, some of which produce the aurora. Networks of ground-based all-sky cameras are valuable tools that offer a large field-of-view with which to study the aurora. Using sequences of auroral images, this study defines criteria for differentiating an important type of aurora whose subcategories are often conflated.

The spatial and temporal evolution of the aurora provides information about plasma dynamics throughout the magnetosphere. The THEMIS all-sky imager network has been operating for over 10 years and has accumulated millions of auroral images. To speed the extraction of information from this dataset, it is desirable to implement automated algorithms to track and classify the aurora. This paper demonstrates an automatic method of extracting the motion of the aurora from sequences of images.