40 Years Ago: The Major 13-14 Nov 1981 Windstorm

Four decades ago, a super-intense extratropical cyclone brought one of the biggest and most far-reaching windstorms to the US Pacific Northwest on record.

Figure 1.1. A large black cottonwood toppled during the 13-14 Nov 1981 windstorm and landed right on a sidewalk. Nearby, hidden by the mass of foliage, the trunk of a mature red alder tree snapped under the force of the wind. The street is oriented north-south, and the fall direction of the cottonwood supports winds from the south. Renton, WA, 14 Nov 1981.

November 1981 began relatively quietly in the Salish Sea region. During the first two days of the month, a baroclinic band brought some precipitation to British Columbia’s South Coast and Northwest Washington. The moisture stream lifted northward on the 2nd, and then swept southeast to revisit the South Coast, Washington and Oregon as a quick-moving front on the 3rd. After this, surface high-pressure ridging built-in, and mainly dry conditions prevailed through the 7th. On the 8th, a weak system brought a little precipitation mainly to Vancouver Island, followed by another pause before a fast-moving and weakening front swept ashore on the 10th. Immediately on the heels of this impulse, a vigorous front associated with a low-pressure system that tracked near Haida Gwaii brought widespread precipitation to the region, including a narrow cold-frontal rain band that brought some heavy downpours to southwest Oregon. This front heralded a complete opening of the “storm door” and the development of a southwest flow pattern over the region that brought one of the most significant windstorms on record from the 13th to 14th.

The windstorm began in the northern Pacific on 12 Nov 1981 far offshore, a weak low tracking into the base of a broad upper-level trough centred near 140 to 135ºW near the left-exit region of a strong jet stream (Figures 1.2, 1.3 and 1.4), a classic setup for the development of strong lows that end up moving near and into the Cascadia region. Carried by a strong zonal flow, the weather system rapidly moved east-northeast, with the centre located near 36ºN as it crossed 150ºW late on 12 Nov, then gradually reaching 40ºN by the time it neared 130ºW late on 13 Nov 1981.

Figure 1.2. ERA5 reanalysis for Friday, 13 Nov 1981 at 1500 UTC (0700 PST). Shown are 500 hPa heights in m (black contours), 500 hPa temperature in ºC (white contours) and 300 hPa jet stream wind in km/h (colour fields and black wind arrows where a half-barb equals 10 km/h, a full-barb 20 km/h and a pennant 100 km/h up to a maximum of 2 pennants). Latitude and longitude are shown with a 5º grid, covering 30º to 55ºN and 150 to 120ºW. The developing storm system is amid the deep oranges, browns and purples where the fastest winds are indicated, with the centre near 38.8ºN 135.3ºW. The surface-low track (white line with shadow) with 3-h positions (colour-filled circles) is included: blue circles indicate the developing stage, orange circles peak intensity and brown mature/degrading. Central pressures in hPa are indicated every 12-h–the first three are from Reed and Albright (1986), with the fourth, 14 Nov 15Z, from the reanalysis data. Also see Figures 1.3 and 1.4.

The extratropical cyclone, supported by a strong jet stream with speeds up to 80 m/s (290 km/h), rapidly deepened as it followed this track. Indeed, an analysis by Reed and Albright (1986) suggests that the central pressure fell at least 50 hPa (1.48″ Hg) in the 24-h ending 1800 UTC 13 Nov, among the highest on record. Given that the mean latitude of the low-pressure centre was 39ºN (range 36º to 42ºN) during this fast intensification period, this development ranks a whopping 2.9 Bergerons where 1.0 Bergeron equals 17.44 hPa/24-h for the given average latitude.

Figure 1.3. ERA5 reanalysis for Friday, 13 Nov 1981 at 1500 UTC (0700 PST). Shown are the surface pressure isobars in hPa (black contours), precipitation intensity in mm/h (colour fields), surface wind speed and direction in km/h (black arrows where a half-barb equals 10 km/h and full-barb 20 km/h) and surface temperature in ºC (black numbers on grid). See Figure 1.2 for further explanation of items in the figure.

A value of 2.9 Bergerons is among the highest on record and may in fact be the most extreme–for storms that have brought high winds to the Cascadia/Salish Sea region in the modern surface airways record. It far exceeds the 1.9 Bergerons the recent and much discussed 21-22 Oct 2021 extratropical cyclone, and even the extreme 2.4 Bergerons of intensification exhibited by its phenomenally deep sister storm on 23-24 Oct 2021. Another recent extratropical cyclone that received attention for deepening rates tracked southeast into northern California on 25-26 Nov 2019. This one also ranked 2.4 Bergerons. Even the 2.0 Bergerons of the infamous 1962 Columbus Day Storm, deepening 35 hPa in 24-h along a mean latitude of approximately 39ºN on a similar track to the 1981 event, falls short.

Figure 1.4. GOES-4 infrared satellite photo taken 1445 UTC 13 Nov 1981, close to the time depicted in Figures 1.2 and 1.3. The developing extratropical cyclone stands out clearly with a classic comma shape off the coast of California. A bent-back occlusion, with clouds wrapping around the low-pressure center creating a shape similar to the tip of a scorpion’s stinger, is clearly evident in this photo. This marks a vigorous storm that is nearing peak intensity. Satellite photo courtesy of the US NOAA National Centers for Environmental Information.

The low-pressure centre of the 13-14 Nov 1981 followed a recurving track up the Pacific coast. This is known as the “classic path” (Figure 1.5). With the low moving northward very close to the coastline, damaging winds can occur across a large region from northern California into southern British Columbia. Thus, these storms tend to have the widest-reaching impacts. Extratropical cyclones with a more zonal–west to east–tracks tend to affect smaller areas with their strongest winds.

Figure 1.5. The low-pressure centre tracks for 13 extratropical cyclones that followed the “classic path” for Cascadia windstorms during the 1940-2017 period. The closer the storms track to the coastline, the greater the likelihood of high winds over land (gusts ≥93 km/h, 58 mph) in the region highlighted in white–though at times strong winds from these storms can develop even further inland. The 130ºW line–highlighted in yellow–is considered to be the critical distance. Strong storms tracking “inside” 130ºW (orange field) have significant potential for high winds. The rule is not absolute, and very intense storms well outside of 130ºW have on occasion brought damaging winds to the region.

The major windstorm of ’81 reached peak intensity just off Northern California, and remained very strong as it headed north-northeast along the Pacific Coast (Figure 1.6). Due to striking the California/Oregon border region during Friday evening and night on 13 Nov, the system has been given the moniker “Friday-the-13th Storm”. A decent title for such a potent system, striking just two weeks after that spookiest of non-holidays. However, for the major population centres from about Salem northward, the maximum winds arrived on 14 Nov. It has been said that Saturday the 14th can be worse than Friday the 13th.

Figure 1.5. At 2200 PST 13 Nov 1981, the storm rakes northwest California and southwest Oregon. The map focuses on surface conditions. Sea-level pressure is in hPa (black contours), near-surface temperature in ºC (gridded numbers), and near-surface gust in km/h (colour fields with scale on the right-hand edge and wind arrows showing direction and speed). Latitude and longitude is on a 2º grid covering 40 to 52ºN and 132 to 120ºW. ERA5 reanalysis data.

The Southwest Oregon coast received particularly intense winds from this storm. The ERA5 reanalysis data paints 120-160 km (75 to 100 mph) gusts throughout this region and extending into extreme northern California. This is supported by actual measurements (Figure 1.6). The peak gust of 148 km/h (92 mph) at Coos Bay, Oregon, is markedly higher than the 130 km/h (81 mph) measured during the 1962 Columbus Day Storm. Wind speeds approaching 150 km/h puts terrific stress on trees, causing widespread stem and root failures. Entire swaths can be flattened. The winds at the 850 height, about 1.25 km above sea-level during this event, are suggestive of near-surface gust potential. They reached 175 km/h (110 mph) right along the coastline near Cape Blanco late on the 13th and this fits well with the available observations.

Figure 1.6. Peak gusts in km/h (left) and mph (right) for the 13-14 Nov 1981 windstorm. High-wind criteria gusts (≥50 knots, 58 mph, 93 km/h) are highlighted with white-filled circles. The track of the low-pressure center is shown with the yellow arrow. In a rather typical manner for this type of storm, the most intense winds affected the Oregon coast.

The extratropical cyclone, now in a mature occluded state, gradually weakened as it tracked northward up the Oregon coast. One mark of this is the cloud field exhibiting a multiply-wound spiral as the occluded front wraps around the low-pressure centre (Figure 1.7). However, very intense storms can still deliver a damaging windstorm even as they are weakening.

Figure 1.7. GOES-4 infrared satellite photo taken 1145 UTC 14 Nov 1981, close to the time depicted in Figure 1.8. The multiply-wound spiral, like a giant cinnamon bun, is the mark of a mature extratropical cyclone that has begun to weaken. Very intense storms at peak intensity can still pack a serious punch even as they gradually weaken. Satellite photo courtesy of the US NOAA National Centers for Environmental Information.

The Friday-the-13th Storm still retained enough intensity to carry the most extreme winds–hurricane-force wind gusts of ≥120 km/h (75 mph)–all the way to Cape Disappointment in southwest Washington. In the early morning hours, just before 0400 PST in fact, strong southerly winds peaked nearly simultaneously across Oregon’s Willamette Valley. The highest gusts were associated with a vigorous leading occluded front that contained embedded thunderstorms. Decent vertical mixing associated with this boundary appears to have helped bring upper-level wind momentum to the surface, enhancing wind speeds.

Figure 1.7. By 0400 PST 14 Nov 1981, the strongest winds of the windstorm had swept into northwest Oregon. Left panel: 500 hPa height in m (solid black curves) and temperature in ºC (white contours), and 850 hPa wind in km/h (colour fields and wind arrows). Right panel: For the surface, sea-level pressure in hPa (black contours), near-surface temperature in ºC (gridded numbers), near-surface gust in km/h (colour files and wind arrows). ERA5 reanalysis data.

Gusts of 114 km/h (71 mph) occurred both at Salem and Portland, the strongest for the former since the intense 02 Oct 1967 windstorm and the latter since an event on 25-26 Mar 1971, a classic-path extratropical cyclone like the Friday-the-13th Storm. Wind speeds at the 850 hPa height were 130 to 150 km/h (80 to 95 mph) over much of the Willamette Valley at this time–this potential was not fully realized at the surface at the official weather stations. However, unofficial reports indicated speeds up to 130-140 km/h (80 to 85 mph) at some locations. Exposed ridgetops even received higher values.

During the late morning and early afternoon, the strongest winds swept through the Salish Sea area (Figure 1.8). With the low continuing to degrade, wind speeds slowed a bit relative to what occurred in Oregon. Even the height 850 hPa winds backed off a tad, generally 125-140 km/h (75-85 mph) over the region. On the Washington Coast, gusts reached 113 km/h (70 mph) at Hoquiam–intense but far below reports from many near-shore locations in the Beaver State. Considering interior sites, Olympia reported a peak southerly (S) gust of 104 km/h (64 mph), strongest since the 1962 Columbus Day Storm. While wind speeds over the Puget Lowlands were generally lower than the reports from Oregon, a small region around the Seattle Area received comparable magnitude. A 107 km/h (67 mph) gust from the S occurred at the Sea-Tac Airport just after noon, highest since 28 Feb 1955. During classic-path storms, peak wind speeds at Sea-Tac are often markedly behind those at Portland–not so much with the Friday-the-13th storm, especially when reports from the nearby Renton Airport are considered. Here, southerly gust speeds reached a damaging 114 km/h (71 mph), strongest since 1962.

Figure 1.8. By 1000 PST 14 Nov 1981, the strongest winds of the windstorm began moving into the Puget Lowlands of Washington. Left panel: 500 hPa height in m (solid black curves) and temperature in ºC (white contours), and 850 hPa wind in km/h (colour fields and wind arrows). Right panel: For the surface, sea-level pressure in hPa (black contours), near-surface temperature in ºC (gridded numbers), near-surface gust in km/h (colour files and wind arrows). ERA5 reanalysis data.

I lived in Renton at the time, and witnessed widespread destruction of trees. This includes seeing some topple during the heavy gusts–in a few cases all too closely! Damage included massive western hemlocks and Douglas-firs that either suffered stem failure or were uprooted, sometimes leaving divots over a meter deep in the ground. Black cottonwoods, red alders and bigleaf maples also experienced numerous failures. Some of these specimens then crashed into other trees, often severely damaging or even toppling the impacted specimen, a sort of arboreal dominoes. Many mature, healthy trees along a cutblock for a new subdivision–individuals that had not had the time to acclimate to heavy wind load after losing their sheltering neighbours–were either damaged or wind-thrown. At least fifty trees fell among the patchy woodlots along the transmission line corridor between Grant Avenue South and the Cedar River. The broken specimens slowly decomposed and became overgrown with new vegetation over the years, but remained an ever-present reminder of the big gale of November ’81. Outside of tree considerations, numerous shingles were stripped from rooftops, littering the ground, and plexiglass stairwell covers on some condominium buildings were punched-out.

Coastal British Columbia is known as the extratropical cyclone graveyard, and for good reason. Many decaying storms land on the coast and then rapidly fall apart upon interaction with the steep and rugged terrain. The Friday-the-13th storm was no exception (Figure 1.9). As the low tracked across southern Vancouver Island, its central pressure shot up to 981 hPa (28.97″ Hg) by 1300 PST. When the low tracked into the mainland Coast Ranges, the low filled even further, increasing by another 5 hPa by 1600.

Figure 1.9. By 1300 PST 14 Nov 1981, the fading low delivers its final peak gusts to British Columbia’s South Coast. Left panel: 500 hPa height in m (solid black curves) and temperature in ºC (white contours), and 850 hPa wind in km/h (colour fields and wind arrows). Right panel: For the surface, sea-level pressure in hPa (black contours), near-surface temperature in ºC (gridded numbers), near-surface gust in km/h (colour files and wind arrows). ERA5 reanalysis data.

For the major population centres on British Columbia’s South Coast, classic-path windstorms are hit-and-miss in terms of generating high winds. Relative to regions to the south, these extratropical cyclones are often at a markedly weakened state by the time they reach Vancouver Island. This limits wind speed potential, though not always. For example, the 1962 Columbus Day Storm (AKA “Typhoon Freda”) while definitely in a weakening state as it landed on the tip of the Olympic Peninsula still delivered the most severe windstorm to the region on record, with wind gusts of 125 to 145 km/h (75 to 90 mph) causing widespread destruction, including the biggest loss of BC Hydro electrical service in terms of percent of total customers affected (~67%).

High winds from the Friday-the-13th Storm just clipped the extreme southern sections of the South Coast. This includes a 114 km/h (71 mph) gust at Victoria’s Gonzales Heights–a well exposed location known for relatively high wind readings. Abbotsford reported a SSE gust of 102 km/h (63 mph), strongest since 10 Dec 1970. Both the Victoria and Vancouver International Airports received gusts below high-wind criteria, SW 72 km/h (45 mph) and SE 85 km/h (53 mph) respectively. Height 850 hPa winds ranged from 90-120 km/h (55-75 mph), much lower than places south and another reflection of the waning storm.

As the Friday-the-13th storm faded from history, a second intense extratropical cyclone developed on its heels (Figure 1.10). This weather system also brought high winds to parts of the Pacific Coast, though overall the speeds were markedly less than with the initial storm. Had the Friday-the-13th storm not occurred, this event would have been the most significant windstorm of the 1981-82 season at many locations. Further examination of this event is a story for another time.

Figure 1.10. GOES-4 infrared satellite photo taken 1145 UTC 15 Nov 1981. Another intensifying extra tropical cyclone is moving up the Pacific coast on the heels of the major 13-14 Nov 1981 windstorm. Satellite photo courtesy of the US NOAA National Centers for Environmental Information.

Pressure gradients–horizontal differences in air pressure–drive the wind. The stronger the gradient, the greater the potential for stronger winds. Peak gradients offer another means aside from wind speed for assessing the strength of a windstorm. For northwest Oregon and southwest Washington, the Friday-the-13th storm produced some intense values (Figure 1.11).

Figure 1.11. Peak pressure gradients for 14 Nov 1981 (left) and 12 Oct 1962 (right) for selected regions. Shown is the magnitude in hPa/100 km, the orientation of the pressure field in degrees and the day and time in PST of the reading.

To put the pressure gradient numbers in context, 4.0 to 5.0 hPa/100 km can support strong winds on the order of 50 km/h G 75 (30 mph G 45). Magnitudes of 6.0 to 7.0 hPa/100 km occur infrequently, especially inland, and can support winds of 65 km/h G 90 (40 mph G 55). Storms that produce 8.0 to 9.0 hPa/100 km are very rare and can support winds of 75 km/h G 105 (45 mph G 65), even higher. Gradients ≥10.0 hPa/100 km are extremely rare, especially inland, and can support winds of 80 km/h G 115 (50 mph G 70), sometimes much stronger. This information puts the 1981 Friday-the-13th Storm into context–just after peak intensity, the extratropical cyclone still yielded gradients in the range of 8.9 to 11.3 hPa/100 km, certainly strong enough to support the observed wind speeds. It is unfortunate that the weather reporting at key weather stations in northwest California, namely Crescent City, was very spotty, preventing computation of peak gradients nearer to the storm when it was at peak intensity–the numbers may have been higher than places to the north.

Peak gradients for the 1962 Columbus Day storm–the baseline windstorm that all others are often compared to–are included for comparison. Many locations had markedly higher gradients during this catastrophic windstorm. The key exception is the southwest Washington interior–given the close proximity of Tacoma McChord Airforce Base to the airport at Olympia, the top two legs of the this pressure-wind triangle, noise in any of the pressure reports from the two stations can lead to significant error. Thus confidence in the numbers is not as high as with other locations. The difference in pressure gradient magnitude far to the north in British Columbia is quite large and readily explains the difference in wind outcomes between the two storms. Indeed, the value of 11.7 hPa/100 km is the highest on record for this pressure-wind triangle–by far–going back to 1953. The southwest Washington Coast value of 15.5 hPa/100 km is extremely high, but interestingly is not the record. That is held by the compact and very intense 03 Nov 1958 extratropical cyclone: 19.8 hPa/100 km, one of the most extreme pressure gradients on record for the Pacific coast from Ukiah, California to Cape Scott on the north tip of Vancouver Island.

Some additional analysis will be added over the next week or so.

Further details on this and many of the storms mentioned here can be found on the Storm King website hosted by the Office of the Washington State Climatologist.

40 Years Ago: The Major 13-14 November 1981 Windstorm

Began: 06 Nov 2021
Modified: 13 Nov 2021

Central Pressure and Wind: The Super-Intense Extratropical Cyclone of 24-25 Oct 2021

An intense extratropical cyclone developed in the northeast Pacific, with a record-setting minimum central pressure of at least 942 hPa (27.82″ Hg) while located near 46.1ºN 131.1ºW on 24 Oct 2021. This low-pressure area eventually tracked across northern Vancouver Island on 25 Oct. However, it weakened considerably to about 980 hPa (28.94″ Hg) just before landfall. This likely prevented a potentially catastrophic wind response. Instead, a moderate windstorm resulted, with only minor disruption. What might have happened had the low tracked across northern Vancouver Island at peak intensity instead of in a degrading state?

At 1800 UTC 24 Oct 2021, the centre of an incredibly intense extratropical cyclone passed nearly right over National Data Buoy Center buoy 46005 (Figure 1.1), also discussed in a previous post. This storm system brought the all-time lowest pressure ever recorded by this offshore station in a record going back to Sep 1976, 942.5 hPa (27.83″ Hg). The previous all-time record of 955.6 hPa (28.22″ Hg) occurred on 10 Nov 1983 during a very active month storm-wise as persistent moist south to southwest flow brought a series of vigorous weather systems into the region, including bringing a significant windstorm to the Puget Lowlands on Thanksgiving Day. Needless to say, the more recent storm decisively smashed this long-standing record.

Map depicting the intense low of 24-25 Oct 2021 in the northeast Pacific.
Figure 1.1. US Weather Prediction Center surface analysis for 1800 UTC 24 Oct 2021. An extremely deep low of 942 hPa is indicated sitting practically on top of Buoy 46005.

The 942.5 hPa minimum at buoy 46005 is likely not the actual low for the 24-25 Oct 2021 extratropical cyclone. There is a real chance that the central pressure ended up somewhat lower than this. I note that the US Weather Prediction Center (WPC) analyzed the low at 942 hPa not only at 1800 UTC 24 Oct, but also for the 2100 UTC, some three hours after the Buoy 46005 low-pressure report. There is the possibility that the low ended up deeper between 1800 and 2100. One could also apply the cyclostrophic wind method using the minimum pressure report at 46005. This would take a precise pinpointing of the low centre relative to Buoy 46005 to be fully useful–no attempt is made here. However, winds were still blowing at 21 km/h G 32 (13 mph G 20) at the time of minimum pressure report–this suggests that the actual low center was not right on top of the buoy. One might estimate an additional 0.5 hPa lower than the 942.5 hPa report, supporting the 942 conclusion by the WPC. It could even be more, maybe ≥1.0 hPa lower (~941.5 hPa), depending on exactly where the centre ended up.

After passing Buoy 46005, the powerful extratropical cyclone followed a receiving path to the northeast and eventually landed on northern Vancouver Island (Figure 1.2). Just by glancing between both Figure 1.1 and 1.2, it is quite easy to discern just how much the extratropical cyclone weakened during the 27 h between passing near Buoy 46005 and reaching the Vancouver Island coast. While the pressure gradient remained fairly strong immediately around the low-pressure centre–it had a compact core–it had diminished markedly further from the low.

Figure 1.2. US Weather Prediction Center surface analysis for 2100 UTC 25 Oct 2021. Low-pressure system nearing northern Vancouver Island, much weaker than a day before.

Despite the weakened state, the storm still packed enough punch to deliver peak gusts of SE 89 km/h (55 mph) at Bellingham, 71 km/h (44 mph) at both Vancouver (CYVR) and Abbotsford (CYXX), and 64 km/h (40 mph) at Victoria (CYYJ) on 25 Oct 2021. An associated vigorous leading occluded front, seen landing on the Oregon and Washington Coast in Figure 1.1, also brought a round of strong wind to some locations on 24 Oct, including gusts to 80 and 81 km/h (49 and 51 mph) to Sea-Tac (KSEA) and Boeing Field (KBFI) respectively. Pacific coastal locations received even higher speeds. Key outcomes from these winds included ferry cancellations, and isolated to scattered electrical service outages due to tree branch and stem failures. Overall, however, the storm did not have a major impact–nothing even close to, for example, the 2018 Solstice Eve Windstorm.

Ultimately, the Salish Sea region and beyond was spared a major to catastrophic windstorm because the storm deepened and reached peak intensity far offshore. What might have happened if the low had tracked across northern Vancouver Island at peak intensity, say with a 940 hPa (27.76″ Hg) central pressure? Or, a related question, just how significant is central pressure in determining peak wind potential for a given extratropical cyclone?

This is a topic that appeared in my dissertation (Figure 1.3). For the time period 2008-2013, the landfall central pressure of every single low-pressure system that tracked across northern Vancouver Island was recorded. Central pressure was then related to the peak gust response at CYVR, CYXX and CYYJ, capturing the region where the majority of people live in British Columbia. To do this, and also help eliminate some of the strong variance that occurs with peak gust reports, the average peak gust for the three stations was used. Vancouver Island is a good place for this kind of analysis because many lows directly move across the north half of the island in a given year. The five years of data captured 50 events. Landfall central pressure is used because with most windstorms, peak gust speed at the three weather stations of interest tends to occur within +/- a few hours of the low-pressure centre reaching the coast.

Figure 1.3. Scatterplot for ETC central pressure at landfall and three-station average peak gust. The open and filled diamonds depict all high-wind generating cyclones that tracked across NVI from 1994 to 2012 and the dashed line shows the linear regression best-fit for these selected events (n=10, R2=0.45). The filled circles and filled diamonds indicate all lows, regardless of strength, that tracked across Northern Vancouver Island (NVI) from January 2008 to April 2013 and the solid line indicates the linear regression best-fit (n=50, R2=0.62). Note that two events are shared between the datasets.

What is found is that there is a strong correlation (R2 = 0.62, n = 50) between landfall central pressure and peak gust speed at the three stations. In essence, central pressure explains much of the variability in peak gust associated with landfalling extratropical cyclones–at least for British Columbia’s South Coast. For other regions, the correlation may not be as strong–but certainly worth looking into! The correlation is not perfect because there are many factors that contribute to wind speed. For example: exact pressure gradients and their orientation relative to terrain features (pressure slope azimuth), storm path direction, exact landfall location on northern Vancouver Island, stage of development, the upper-level wind environment and available vertical mixing mechanisms. But given the R2 value, central pressure appears to be perhaps one of the best predictors of outcomes. Better than one might at first expect and incidentally a stronger predictor than the pressure gradient in the vicinity of the study stations.

Some factors that likely contribute to making extratropical cyclone central pressure a good predictor of wind speed include: 1) the deeper the low, the more likely a strong pressure gradient will develop in at least some regions around the low; 2) deeper lows likely have an upper-level wind environment conducive to supporting strong surface winds; and 3) as lows become more intense, the region affected by the strongest pressure gradients tends to get larger, making it more likely that a specific region will be affected by a steeper gradient.

Looking at the chart in Figure 1.3, it appears that lows with a ≤990 hPa (29.23″ Hg) landfall central pressure are the most likely to produce strong (65-90 km/h, 40-55 mph) wind speeds. Moving up the wind scale, those with ≤985 hPa (29.09″ Hg) are most likely to produce severe (≥90 km/h, 55 mph) wind outcomes. Given this, what about ≤960 hPa (28.35″ Hg), or ≤945 hPa (27.91″ Hg)? Perhaps catastrophic?

This brings us back to the initial question: What might have happened if the low had tracked across northern Vancouver Island at peak intensity, say with a 940 hPa (27.76″ Hg) central pressure? The linear model shown in Figure 1.3 can be used to make an estimation. A key caveat here is that the data range goes down to 961 hPa (28.38″ Hg). To get to 940 hPa, extrapolation of the best-fit line (the black line in Figure 1.3) has to be done. This is, of course, fraught with peril as there may be a nonlinear response between peak gust speed and central pressure at some point on the depth scale. Also, there are fewer datapoints at the top end, weakening the strength of any inferences. This means that the estimates made here are not high confidence but perhaps closer to medium confidence.

Taking the model at face value, here is what can be concluded: A 940 hPa landfall on northern Vancouver Island is a 4 standard deviation event, extremely rare. The predicted average peak gust is 26.4 m/s (95 km/h, 59 mph or above high-wind criteria) at the study stations. If this occurred, it would mark the 3rd strongest windstorm since the 1962 Columbus Day Storm for the region of interest. It would certainly fall into the major windstorm category with significant damage outcomes, perhaps approaching borderline catastrophic in some locations. Given the variance around the best-fit line, an over-achieving event could produce a 31 m/s (111 km/h, 69 mph) average peak gust, approaching but still short of the outcome of the 1962 Columbus Day Storm (34.5 m/s, 124 km/h, 77 mph), and certainly the strongest event since the Terrible Tempest of the 12th in ’62. A windstorm of this magnitude would likely have a catastrophic damage outcome, and might be considered a historical event depending on just how much impact it had. On the opposite side of the spectrum, call it an underperforming scenario, the average peak gust would be close to 21 m/s (76 km/h, 47 mph)–still a notable windstorm capable of causing widespread tree failures and electrical service disruption, but not quite in the major windstorm category. Incidentally, the actual 3-station average peak gust ended up at 19.1 m/s (69 km/h, 43 mph), not too far off the predicted for the degrading 980 hPa low that did arrive.

In conclusion, while many factors contribute to peak gust outcomes from extratropical cyclones that affect the Salish Sea region, central pressure is certainly a good indicator of the potential peak gust speeds. The more intense–deeper–the low, the more likely a severe wind outcome will occur. Lows with ≤985 hPa central pressure at landfall are the most likely to produce strong to severe winds. If the intense 24-25 Oct 2021 windstorm had landed on northern Vancouver Island with a 940 hPa central pressure, a major windstorm appears the most likely result for the southern inner coast of British Columbia. At the extreme end of possibilities, a catastrophic windstorm could be the result.

Super-Intense Storm in the Forecast

An extremely strong extratropical cyclone is forecast to develop southwest of Vancouver Island on Sunday/Monday, 24-25 Oct 2021.

For further analysis of this event after the fact, see this more recent post.

On Sunday/Monday, an extraordinarily deep low is forecast to develop far off the coast of Washington/Oregon and then follow a wobbly path northeast onto northern Vancouver Island. The GFS forecast shown in Figure 1.1 indicates an intense 943 hPa (27.85″ Hg) central pressure at 1000 PST on 24 Oct 2021. This is an extremely rare value for the depicted region (42 to 52ºN and 132 to 120ºW).

Figure 1.1. GFS forecast for 1000 PST 24 Oct 2021. Included in this display is sea-level pressure isobars (1 hPa increments), total 3-h precipitation (colour shades), temperature (gridded numbers in ºC), 50 kPa heights (faint lines) and near-surface wind (classic arrows with barbs).

Due to growing commentary on the extreme nature of this forecast event–it has not happened yet–I have received questions about other such “deep” storm systems. In fact, a big question on people’s minds is whether-or-not this is some kind of new all-time record. If the expected low does intensify into the sub-945 hPa category, then it may in fact be a new all-time record low for the region being shown in Figure 1.1.

While I have not done a comprehensive analysis of extratropical cyclone central pressures for the period of record–say 1948 to present–given the rarity of sub-950 hPa storms, and the tendency of these storms to result in notable extreme weather provided that they track close to the coastline, a reasonable estimation can be made from the available record on windstorms that have affected the region. In this case, the storm that stands out for all-time lowest central pressure is the 13-14 Nov 1981 extratropical cyclone that brought a major windstorm from Northern California into the South Coast of British Columbia. This is sometimes known as the “Friday-the-13th” storm–adds a nice, spooky sense to the event. Shown in Figure 1.2 is the ERA5 reanalysis estimate for conditions when the storm was near peak intensity.

Figure 1.2. ERA5 reanalysis for 2300 PST 13 Nov 1981. Included in this display is sea-level pressure isobars (1 hPa increments), total 3-h precipitation (colour shades), temperature (gridded numbers in ºC) and near-surface wind (classic arrows with barbs).

The reanalysis indicates a 962 hPa (28.41″ Hg) central pressure. This is too high–and not surprising with a reanalysis product as the gridded nature of the data is not likely to capture the greatest extremes. Based on a detailed analysis of the storm done by Reed and Albright published in the Monthly Weather Review in 1986, the minimum central pressure of the 1981 storm reached 947 hPa (27.96″ Hg) at 1900 PST on 14 Nov 1981. The crew aboard the Sea-Land Liberator reported a pressure of 956 hPa (28.23″) at this time, along with 28 m/s (101 km/h) near-surface winds, which informs that the central pressure was even lower than being reported by the barometer.

In essence, 947 hPa is the number to beat to achieve what may be a new all-time record low within the region of concern. Since the Friday-the-13th Storm, a number of extratropical cyclones have challenged the title. This includes the powerful 12 Dec 1995 windstorm (Figure 1.3), one that was a near equal to the Nov 1981 event in terms of wind speed and damage outcomes all the way from San Francisco, California to Victoria, British Columbia. Best estimates put the minimum central pressure at 953 hPa (28.14″ Hg) just off the Washington Coast. Indeed Buoy 46041, Cape Elizabeth, reported a minimum pressure of 958.8 hPa (28.31″ Hg) as the low neared landfall on the Olympic Peninsula. In the figure, the ERA5 reanalysis indicates a 958 hPa (28.30″ Hg) central pressure, closer to the mark than with the 1981 storm.

Figure 1.3. ERA5 reanalysis for 2200 PST 12 Dec 1995. Included in this display is sea-level pressure isobars (1 hPa increments), total 3-h precipitation (colour shades), temperature (gridded numbers in ºC) and near-surface wind (classic arrows with barbs).

Another extratropical cyclone of note is one that tracked just east of due north between 132º to 130ºW on on 10 Nov 1983 and passed very near Buoy 46005. This National Data Buoy Center station, located at 46.1ºN 131.1ºW, reported a minimum central pressure of 955.6 hPa (28.22″ Hg), the all-time lowest on record at this site (update: smashed by the 24-25 Oct 2021 extratropical cyclone being discussed here–see the following post). More recently, the stronger of a pair of lows responsible for the 02-03 Dec 2007 “Great Coastal Gale” in northwest Oregon and southwest Washington had a minimum central pressure of at least 959 hPa (28.32″ Hg), this at 1000 PST on 03 Dec 2007 based analysis done by the US Weather Prediction Center. However, this occurred just outside of the region, around 135ºW. The low gradually weakened as it approached Vancouver Island. A month later, on 04-05 Jan 2008, another strong low approached the Pacific Coast. This one achieved a central pressure of 956 hPa (28.23″ Hg) at 1900 PST 05 Jan 2008 near 49ºN 132ºW–just off the Vancouver Island coast. And, of course, there is the infamous 1962 Columbus Day Storm, the most damaging windstorm to affect the region in the modern surface airways record, one that still holds most of the peak gust speed records in western Oregon, Washington and parts of the British Columbia South Coast. This tempest had a minimum central pressure of at least 958 hPa (28.29″ Hg)–later analyzed at 960 hPa (28.35″ Hg) by Lynot and Cramer in their 1966 paper that appeared in the Monthly Weather Review.

Of course, I would be remiss not to mention that a precursor super-deep low moved through the offshore waters on 20-21 Oct 2021 (Figure 1.4). The minimum central pressure reached 951 hPa (28.08″ Hg) during the early morning of the 21st. Like the 02-03 Dec 2007 event, this low reached its peak intensity outside of the 132ºW limit used here, and it tracked nearly due north along 135ºW. Maybe it can be considered an honorary member (like the 2007 storm), in part due to its close timing to the forecast storm of interest here. With the 20-21 Oct 2021 extratropical cyclone, the strongest winds were limited to Northern Vancouver Island northward to Haida Gwaii. This includes a SE 165 km/h (102 mph) gust at Solander Island, usually a leader when it comes to extreme wind speeds. Other reports include SE 113 km/h (70 mph) at Sandspit, ESE 85 km/h (52 mph) at Port Hardy and SE 77 km/h (48 mph) at the Estevan Point lighthouse. All times PST.

Figure 1.4. US Weather Prediction Center surface analysis for 0700 PST 21 Oct 2021. Shows an intense 951 hPa (28.08″ Hg) low near 47ºN 135ºW.

Given the intense central pressure in the forecast for Sunday/Monday 24-25 Oct 2021, and the windy outcomes of the past storms mentioned above, one might expect that some strong to severe winds will affect Oregon, Washington and British Columbia. Indeed, elevated winds are in the forecast, but these are not expected to approach the big storms of history at most locations. The key reason for this is that the extratropical cyclone is forecast to reach peak intensity well off of the coast, then steadily weaken as it approaches Vancouver Island (Figure 1.5). By 1900 PST 25 Oct 2021 in the GFS forecast, the central pressure has increased to 978 hPa (28.88″ Hg) just before landfall on the north tip of the Island.

Figure 1.5. GFS forecast for 1900 PST 25 Oct 2021. Included in this display is sea-level pressure isobars (1 hPa increments), total 3-h precipitation (colour shades), temperature (gridded numbers in ºC), 50 kPa heights (faint lines) and near-surface wind (classic arrows with barbs).

A central pressure of 978 hPa is a far cry from 943 hPa, but is still at the potent ≤980 hPa for area windstorms. Thus, winds speeds along the immediate Pacific coast are still likely to reach the strong to severe categories, say gusts of 80 to 120 km/h (50-75 mph). In the interior, such as the Willamette Valley, Puget Lowlands and along the Georgia Strait, speeds are likely to be much less, more in the 50 to 80 km/h (30-50 mph) range.

For further information on some of the storms mentioned in this narrative, see The Storm King.