Just How Strong Was That Windstorm?


Wolf Read

There are many ways to measure the strength of a windstorm, none of which offer what might be considered a definitive answer. Below is an examination of the pros and cons of various methods for determining storm strength.

Peak Gust: Peak instantaneous gusts can be compared for various locations, which is what I tend to do in the examination of windstorms on this website. I've picked this method mainly out of convienience, for peak gusts are often the simplest figure to find. Peak gust has the limitation of comparing a blast of wind that lasts for just a brief moment in the duration of a storm that may last for hours. It seems logical that lasting winds might do greater damage by slowly weakening trees and structures, even if the long-duration winds aren't as strong as a higher instantaneous gust from a different storm. Gradual weakening of structures is concievable due to, say, nails being worked loose in a flexing structure, or rootlets breaking little-by-little as winds constantly pull at a tree. And if there is rain, the softening of ground over time can contribute to further tree loss. Note, however, that it takes three to six seconds for stable aerodynamic forces to set up around a structure such as a house; thus an instantaneous gust is likely not to have exerted its full force [1]. But a peak gust could be reflective winds in the 3-6 second category that are nearly as high, as this is a fairly short period of time when compared to the standard measures of sustained winds (see below). There is another problem with peak instantaneous gust: readings can vary considerably between the many varieties of anemometers due to differences in their inertial response. In other words, some anemometers are more sensitive than others. This variance can be 30% or more [2]. Thus, it seems, truly meaningful peak gust comparisons should come from the same kind of anemometer sensors. Comparing NWS ASOS stations, for example, is failry valid, for the system is standardized. Comparing unofficial readings is a different matter, for they could be from any number of wind measuring systems.

Sustained winds: Sustained gusts, or systems like fastest one- or two-minute windspeed, and fastest mile (which is the highest velocity maintained by one mile of air passing the anemometer: if the wind blows a steady 60 mph, it'd take a minute for a mile of air to pass), provide a measure of longer-duration winds, but then there is the question of just how much duration is needed to provide significant meaning? Minutes, or hours? It has alredy been pointed out in the discussion about peak gust that it only takes 3 to 6 seconds for a wind to exert its maximum force on an object. Secondly, peak gusts are fairly well correlated with average or sustained winds: in a typical cold-core storm, the peak gusts are typically 1.3 times the average wind speed, with warmer storms, such as hurricanes, approaching 1.5 times [3]. Thus the average wind speed may be no better than the peak gust--one could infer it from the peak gust figure, and a little knowledge about the anemometer's capabilities and the structure of the storm. Another problem with peak sustained wind occurrs in the record for storms that occurred pre-ASOS (About 1993-1996 depending on location). Sustained one-minute winds were typically calculated at intervals of once an hour, or a little more frequently when special observations were being taken. Also, these readings sometimes involved estimation, with rounding to the nearest five knots, and technique probably varied to some degree between observers. Thus, for this time period, peak gust--not peak wind--was often the most accurate "peak" figure obtained.

Duration of gale-force winds: This gets at the problem of the length of time the storm is afflicting a region and the potential for longer periods of high winds to cause more damage from gradual weakening of trees and structures. The cutoffs, like "gale-force" would be arbitrary. Number of hours with 40 mph gusts or greater might be one category, or number of hours with sustained winds of 30 mph or greater. It probably doesn't matter too much--what does matter is trying to capture the length of the damaging winds. How necessary it is to do this, I'll note, is debateable: most big storms produce their highest winds within a similar time frame, say 2 to 4 hours, with unusual ones lasting 8 to 12. It seems that the strongest wind producers, which are often moving at a quick pace (which aids surface wind speeds), don't often last as long as moderate windstorms.

Size of region being examined: As is pointed out on some of these pages, like the one for the November 1983 storm train, windstorm strength is often a local thing. One city may get record wind gusts while another a few tens of miles away might just experience a typical winter's gale. This is due to a number of variables, including terrain factors, size and type of storm, angle of the pressure gradients across various terrain sections, and what path the low center follows. Sometimes a storm merely strikes part of the coast, while other times a Pacific cyclone will sweep an area measured in tens of thousands of square miles. In terms of determining strength, one might argue that the bigger the area affected, the bigger the storm. A storm that consistently produces peak gusts of 60-70 mph across portions of several states might be considered a more significant event than a squall line that affects just a few counties with wind speeds in excess of 90 mph. Clearly, this kind of thing gets difficult to assess.

Time of year the storm strikes: Dead and weakened trees accumulate over the summer months as disease, insect attack, and old age take their toll, such that the first significant windstorm can cause more blowdowns than might be expected. Thus, subsequent storms in the same winter season may produce far fewer fallen trees and branches, even if they have higher windspeeds--for the first storm cleaned out all the material waiting to go! Also, many deciduous trees do not drop their leaves until November in the Pacific Northwest, and some leaves hang on into December. Leafless trees have less surface area for the wind to act on. Early autumn storms are likely to have a stronger impact on such trees than winter gales for this reason. Thus, in terms of a storm's potential threat to human life and property, it is concievable that timing can play a big role--maybe even more so than actual wind strengths. Which, ultimately, means that counting wind-thrown trees (and houses damaged as a result) may not be a good measure for determining just how strong a storm actually was. See my discussion of the January 1986 storm train for more details.

Dollar Damage: How many dollars damage a storm causes could be used as a measure of storm strength, for in a sense, it combines a number of the ideas above: a storm with very high winds affecting a broad region for a longish period of time is likely to produce more dollars in damage than a short gale on part of a coastline. And this might be a more meaningful measure to many people, for it gets at something that most folks can relate to: money. However, it must be noted that some storms can and do cause huge damages to relatively small locations. There have been several tornados in the midwest that have done more dollars in damage than the infamous Columbus Day Storm of 1962, which did $210 million to parts of OR and WA. The February 1979 windstorm, with mostly affected the Olympic Peninsula of WA, did nearly as much damage as the 1962 storm because of the catasrophic failure of just one structure: the Hood Canal Floating Bridge. As intense as the 1979 storm was, most would agree than the 1962 event was still the far larger one. Also, many dollar damage figures are estimates. Not all damage gets reported. Often reassessments in later years change the figure (usually upwards). So, we see that even measures of property losst has its limiations.


[1] Simpson, Robert H., Riehl, Herbert. The Hurricane and its Impact. 1981. p. 205-207.

[2] Simpson, Robert H., Riehl, Herbert. The Hurricane and its Impact. 1981. p. 204.

[3] Simpson, Robert H., Riehl, Herbert. The Hurricane and its Impact. 1981. p. 205.

Last Modified: March 5, 2003
Page Created: September 6, 2001

You can reach Wolf via e-mail by clicking here.

| Back |