[My friend Wolf Read is an expert on wind storms. Last week he wrote the rather technical piece below. He graciously offered to let me post it. – George]
The Minnesota Extra-tropical Cyclone (ETC) of October 2010 deepened further than the Columbus Day Storm (CDS) — 95.4 kPa verses 95.8-96.0 kPa depending on source. However, a striking difference in surface wind speed response resulted. Using sea-level pressure for weather stations STC, PKD and DLH to examine the geostrophic potential winds (Mg), it appears that surface winds at many stations within the triangle tended not to exceed about 0.30 Mg (30% of the geostrophic potential). Many were lower, in the range of 0.20-0.25 Mg. During the Columbus Day Storm, surface winds approached more closely to the geostrophic potential in some regions of western Oregon and Washington. Using sea-level pressure for OTH, AST and PDX to compute Mg, it is clear that many Willamette Valley stations approached 0.35-0.40 Mg, and some (such as PDX), even approached 0.5 Mg. It appears that the CDS packed way more punch.
Some caveats about calculating Mg. The size of the area between the three stations can affect the outcomes. This is because many intense storms can be fairly compact. Under such circumstance, a larger area has a tendency to reduce the geostrophic potential, because a large triangle may not “sense” small-scale pressure differences to the same degree as a smaller triangle. In other words, if a smaller triangle of stations is used, then the Mg may be markedly higher for a given storm, provided that the event is smaller (say on the border between meso- and synoptic). The areas covered by the two triangles in the above analysis cover 10,057 km^2 for MN and 9,286 km^2 for OR. They are very close and area differences would not be a big issue when making the comparison between the ETC and the CDS if the two storms were on similar scales.
Nevertheless, relative to the Oct 2010 Minnesota ETC, the CDS appears to have been more compact. And, when a smaller area is applied to the CDS pressure data, the geostrophic potential appears to go up. Since the surface wind data stays the same, this lowers the proportion of geostrophic potential realized by the observations. Using AST, HQM and OLM (area 3,493 km), a geostropic potential of 105 km/h 0.25 Mg and 168 km/h 0.40 Mg is returned. Most stations in the area only achieved 20-25% of Mg. Note, however, that some stations just outside the triangle, like PDX (88 mph fastest mile = 142 km/h) still approached 0.40 Mg even at this finer level. I have not looked at a finer triangle in Northwest OR as of yet, but will probably do so soon.
Another way to look at the situation is just to ignore the surface observations altogether and focus on the Mg. It is a reflection of a given storm’s pressure gradient. In this regard, the CDS clearly outshone the Minnesota ETC. Note the differences in peak Mg: For the 2010 ETC, 59 km/h for 0.25 Mg and 95 km/h for 0.40 Mg, and for the CDS, 77 km/h for 0.25 Mg and 123 km/h for 0.40 Mg. This tells a significant part of the story. The differences can be seen in the pressure trends (included on the attached graphs). Though they did get very low, barometric readings for the MN ETC just plodded along, partly due to the vast scale of the system. For the CDS, the pressure trace looks like a fairly sharp V. The rapid pressure changes during the CDS not only reflect a compact storm with tight pressure gradients, but also one other key aspect to surface wind response: Storm motion.
I suspect that the key reason for the somewhat stronger surface response, relative to geostrophic potential, is that the Columbus Day Storm moved northward at a very fast pace, in the vicinity of 90 km/h if not more. This northward component of storm momentum likely added a kicker to southerly surface winds–the favored direction in the right rear quadrant of the storm. The October 2010 Minnesota ETC moved very slowly while in its deepest state. Surface winds probably received just a minor kicker from the storm’s overall motion.