Sandy’s left hook
On Halloween I received an e-mail from Tim Queeney, Ocean Navigator’s editor, with the subject line “Sandy’s left hook.” Tim asked, “What happened with Hurricane Sandy so that the storm took such a radical leftward turn? Doesn’t the Coriolis effect tend to make storms curve to the right in the northern hemisphere?” I replied to Tim telling him that the topic would make a great weather e-mail newsletter, and here we are in early December, and it’s time to make good on my promise.
By now, everyone is well aware of the profound impact that Sandy had on the eastern U.S. with the worst casualties and destruction occurring in the New York metropolitan area, but effects covering a huge area including hurricane force wind gusts as far north as Maine, as far south as North Carolina, and as far west as Lake Huron. Excessive rainfall occurred throughout the northeastern U.S., south into the Carolinas and west into the eastern Great Lakes, and heavy snow fell through much of the central Appalachians with over two feet at many locations in North Carolina, Tennessee, Virginia, Maryland and West Virginia. Before any of this happened, Sandy traversed Jamaica and the Bahamas producing hurricane conditions in those areas which led to considerable damage and fatalities.
Typically a hurricane which moves from the Caribbean northward through the Bahamas would turn more to the north-northeast or northeast as it moved into higher latitudes over the western Atlantic, but Sandy was not a typical system and, most important, the weather pattern over the western Atlantic and eastern North America was not typical either. All of these factors led to Sandy tracking north-northeast to the east of the Carolinas after departing from the Bahamas, and then taking a significant turn toward the northwest, and then the west-northwest, with its center making landfall in extreme southern New Jersey. And this is what generated Tim’s questions.
First, let’s address the Coriolis issue. The Coriolis effect was described by Gaspard-Gustav Coriolis in the early 1800s and allowed Newton’s Laws of Motion to be applied in a rotating frame of reference by applying a correction factor. This correction factor causes large-scale motions on the earth to appear to be deflected when observed on the rotating frame of reference. Those who have studied basic meteorology are familiar with the application of the Coriolis effect to help determine wind speed and direction when the pressure field (i.e. the isobars) is known. Barometric pressure differences lead to forces causing air to tend to move from higher pressure toward lower pressure, but the Coriolis effect deflects this motion to the right in the northern hemisphere, and so air moves (or wind blows) nearly parallel to the isobars rather than directly across them as would be the case if the pressure force was the only factor to consider.
What is important to understand, though, is that the Coriolis effect applies only to individual motions of air, and not to an entire system like an area of high or low pressure or, in this case, Hurricane Sandy. The air circulating around Sandy was indeed being deflected to the right as it was pushed toward the center of the system, thus leading to the rotational airflow around the system. However the track of the entire system was not subject directly to the Coriolis effect, but rather was directed by an unusually complex weather pattern in the atmosphere made up of many different air motions, all of which individually were subject to the Coriolis effect.
Looking at this weather pattern will provide the answer to Tim’s first question about why Sandy made such a hard left turn in the western Atlantic leading to the landfall in New Jersey. As noted above, hurricanes moving north through the western Atlantic will often turn more toward the northeast or the east as they move to higher latitudes (or to the right, as it were). During the summer and early autumn, upper level winds are usually rather weak in lower latitudes allowing the motion of tropical cyclones to be directed by weaker lower level wind patterns. As these systems move into higher latitudes, they begin to fall under the influence of stronger upper level winds, which usually flow generally from west to east, and this leads to the eastward turn.
In late October of 2012, though, the upper level pattern was unusually strong and unusually amplified. This is shown by looking at the 500 millibar chart from 0000 GMT on Oct. 29 (Figure 1). This chart shows a large and strong upper level high well to the east of Newfoundland while the flow pattern over the eastern U.S. dips quite far south before then turning north and moving through eastern Canada and around the upper high. Of particular interest is the prominent dip in the flow (a trough) extending from Lake Michigan southward into the southeastern U.S. This feature was associated with an unusually cold system for the time of year, and it is actually about to become a closed low center over Indiana. The closed low that is located in the Atlantic near 70W is a reflection of Sandy. Typically hurricanes do not have a strong signature at the 500 millibar level, and this is an indication that Sandy at this time was already beginning to show signs that it was not a typical hurricane since it was beginning to acquire some characteristics of a non-tropical system, one of which is increasing size.
The strong upper high east of Newfoundland would prevent Sandy from moving north into that region, and as the low center formed in Indiana, it would reinforce the southeast windflow in the mid-Atlantic coastal waters, which would capture Sandy and pull it toward the northwest.
Figure 2, which is the 500 millibar chart 12 hours later, shows the upper low which formed in the Ohio valley region having shifted south to the southern Appalachians and the strong upper high east of Newfoundland having drifted west and became stronger. Sandy by this time had begun to turn to the north-northwest, and given the strong high to the north and the strong low in the southern Appalachians, the upper level flow leaves little question as to where Sandy will end up. In fact, through the ensuing 12 hours, the hurricane and the upper low to its west would do a bit of a dance with one another, and eventually merge. The two features would also interact dynamically resulting in a much stronger upper level low by 0000 GMT Oct. 30 (Monday evening Oct. 29, eastern time - Figure 3). At this time, the structure of Sandy was no longer tropical in nature, but it was a very powerful extra-tropical low, and its center was moving through extreme southern New Jersey.
So it was the unusual upper level pattern, partly a factor of the lateness of the season, which resulted in the hard left turn Sandy made toward the mid-Atlantic coast. The Coriolis effect still causes air motions (not system motions) to be deflected to the right in the northern hemisphere, and this will not change unless the earth somehow begins to rotate in the opposite direction.