The Following figures were presented by Larry Lyons at the Workshop

The points he made are summarized in the captions

 Figure 1
Figure 1 (based on Anderson et al., JGR, p. 15,757, 2000): IMF measurements from three spacecraft indicating that steady, strongly southward, IMF impacted the magnetosphere during the man phase of the October 19, 1998 storm from ~0440 to ~1510 UT. During this period of enhanced steady convection, no substorms were observed until 1325 UT, indicating that substorms generally do not occur when the IMF driver is known to be strongly southward and steady. After ~9 1/2 hours of steady IMF driving, a substorm did occur. However this occurred shortly after a change of Bx with respect to x was observed in the solar wind, implying significant y (or z) variation of By (or Bz), and when a northward turning was observed by Geotail when it briefly entered the magnetosheath (See Figure 2).

 Figure 2
Figure 2: Bz, measurements from Geotail as compared to the IMF Bz measured by Wind on October 19, 1998. Geotail was mostly within the magnetosphere but occasionally entered the magnetosheath; the measurements are shown as a dark black line during intervals when Geotail was identified as being within the magnetosheath. These measurement show that the 1325 UT substorm may have been the result of a northward turning (reduction in negative Bz,) of the IMF that occurred in the vicinity of the Earth, despite this northward turning not being seen in the IMF measurements shown in Figure 1 by three spacecraft in the solar wind. 

 Figure 3
Figure 3 (Lyons, L. R et al., New Perspectives on the Earth's Magnetotail, AGU, p. 241, 1998): CANOPUS meridian scanning photometer observations illustrating the difference between the poleward expanding regions of intense aurora associated with substorms (which initiate in the equatorward part of the aurora oval) and the much more frequent , but small-scale, auroral poleward boundary intensifications which initiate near the poleward boundary of the auroral oval. 

 Figure 4
Figure 4 (based on Lyons et al.,. JGR, p. 349, 2001): Ground X and Pi 2 and geosynchronous energetic electron observations showing that geosynchronous injection (and presumably dipolarizaton0 initiated ~2 1/2 minutes prior to onset 

 Figure 5
Figure 5 (Lyons, JASTP, p. 1087, 2000): Representative IMF measurements from multiple monitors during time intervals surrounding substorm onsets identified with CANOPUS photometer and ground magnetic field data. All times have been corrected to an estimated magnetopause contact UT assuming magnetic structures are aligned at 45° in the x,y plane. Notice that large differences occur between simultaneous measurements of the IMF from satellites that are often used as IMF monitors. Also notice that substorms can appear to occur during a steady IMF (e.g., 2/19/96) or in association with a southward turning of the IMF (e.g., 4/15/96) based on one monitor, but appear to be triggered by a northward turning of the IMF based on other monitors located closed to the Earth-sun line. 

 Substorm-Onset Wrokshop Website