On April 23, 1999 a downslope windstorm along the western side of the Wasatch Mountains of northern Utah produced wind gusts to 113 mph at valley elevations. Twenty semi tractor trailer trucks were blown over on I-15, a construction crane collapsed, damage to hangars and aircraft was reported at the Brigham City airport, numerous power lines were downed, and there was widespread damage to roofs, trees and small structures throughout Davis, Weber, and eastern Box Elder counties. Damage to residential property was estimated at $3-5 million. This was the strongest downslope wind event along the Wasatch Front in more than a decade (since December 15, 1988) and the 113 mph observed at the Brigham City Airport represents a new record for observed windspeed at low elevation (<5000 feet) in Utah. Look at the Local Storm Report to see specific wind speed and damage reports.
This case study will document the event as it appeared in observations and analyses centered around 1200 UTC April 23, 1999 and will attempt to highlight the most significant aspects of this event. The next section will cover some significant concepts associated with downslope windstorms. An overview of the synoptic situation will then be provided. A section describing the observations and weather is included. Additionally, links are provided to a subset of graphics from five other downslope windstorms of similar magnitude for comparison and reference. These are the March 21, 2000, December 15, 1988, April 4, 1983, February 20, 1971, and November 11, 1978 events. Finally, the High Wind Warning issued for the April 1999 event is presented as an example that might be considered in future situations of this magnitude. It is hoped that this document will serve as a useful tool when forecasters are confronted with similar events in the future.
Windstorms along the western side of the Wasatch Mountains occur over a spectrum of events ranging from nearly pure gap flow to other events that are characterized as mountain wave events. The strongest events, such as the 4/23/99 event are primarily mountain wave events, although some gap flow is present in all events. This situation is very similar to windstorms that occur along the west slopes of the Cascade Mountains in western Washington where both gap and mountain wave events take place in easterly cross-barrier flow with higher sea-level pressure on the east side of the barrier. The literature is rich with theoretical and observational research results pertaining to downslope windstorms. Most of this work is based on regions where downslope windstorms occur on the east side of mountain ranges, in westerly flow. At this time, one of the best references for events that are similar to Wasatch windstorms are the two papers by Colle and Mass (Monthly Weather Review, Jan. 1998, in particular see Part II, p. 53). Much of the discussion below is taken from the results of Colle and Mass Part II, and other research, which were also presented at a COMET-sponsored NWS workshop held in Seattle in late 1996.
The following 4 factors were found to be most significant for Cascade windstorms:
What is a critical level?
When strong flow encounters a topographic barrier, a vertically propagating gravity wave is generated. A key to mountain-wave formation is when the energy from the vertically propagating gravity wave can no longer go up, the energy may be redirected to the surface. This energy propagation is why it is possible to produce surface wind speeds at the base of the mountains far in excess of the wind speeds observed at any level in the free atmosphere. A critical level is the level in the atmosphere that prevents the gravity wave energy from continuing upward.
A critical level can be either a "mean-state" or "self-induced" type. A mean-state critical level means that it exists in the larger-scale observed or forecast data, while a self-induced critical level is generated by the mountain-wave itself, and probably cannot be observed except perhaps with a field program. A mean-state critical level is often defined as a level in the atmosphere where the cross-barrier flow goes to zero. This can be a place where the winds become very light, or where the winds become parallel to the barrier, or it can be the level where a wind direction reversal takes place (going from easterly to westerly flow). A mean-state critical level can also be a stable layer topped by an unstable layer. A fluid oscillation can take place in the stable layer, but not in the unstable layer, so the energy from the oscillation (a gravity wave) cannot propagate upward. A self-induced critical level can be visualized as the turbulence generated by a breaking wave. Consider a breaking ocean wave, where, as the wave builds and then finally breaks, a region of turbulence is created where the flow may either reverse or go to zero. This region acts to prevent energy from propagating vertically through it.
Wave breaking seems to be enhanced in the presence of "reverse shear". Reverse shear is cross-barrier flow that decreases with height. It is thought that even in the absence of a mean-state critical level, that reverse shear can lead to wave breaking, and thus, a self-induced critical level.
Synoptic-scale at 1200 UTC April 23, 1999
A deep trough had developed over
the Great Basin by the morning of 4/22/99. Jet level energy was dropping
into the base of the trough with cyclogenesis forecast to take place near
Las Vegas by 1200 UTC 4/23. Moderately cold air over Montana was
forecast to move south into Wyoming at the same time and easterly upslope
and dynamic ascent were also forecast to produce 700 mb cooling over Wyoming.
The forecast cyclogenesis took place. The result was a strong easterly
(cross-barrier) 700 mb flow as the upper low closed off and intensified
and similarly, an increasing sea-level pressure gradient between SLC and
Riverton, WY (cross-barrier pressure gradient). Below are analyses,
observations and some imagery from around 1200 UTC April 23, 1999.
The sea-level pressure difference between SLC and RIV was over 19 mbs at 1200 UTC. Given that 8 mb is often enough to generate moderate gap winds, this is an extreme value. Pressure tendencies seen in the METAR observations, show pressures rising in northeast Wyoming, while falling over southern Nevada. Again, it is interesting to note that previous ETA forecasts were accurate in predicting this strong pressure gradient, but that the ETA analysis and sea-level reduction method was unable to draw for the data shown by the METAR observations.
The SLC skewt shows a mean-state critical level near 400 mb as the winds become southerly and the cross-barrier component of the flow goes to zero. There is also a suggestion of reverse shear in the skewt with the flow dropping to only 15 knots near 400 mb. Stable layers near or just above ridge-crest can also be found in the skewt at 700 mb and 600 mb. There is of course some question about exactly where the SLC skewt was measuring the atmosphere in this event, since the balloon was likely some distance to the west of the Wasatch Front during its ascent.
The cross-section generated from the ETA analysis shows some reverse shear around 600 mb, but no mean-state critical level until near the tropopause. It does show a decrease in stability above 500 mb where the isentropes are further apart. This less stable region above a more stable region may act as a mean-state critical level. There is also some slope to the lowest isentropes as they tilt downward on the west side of the Wasatch Mountains, but not much as the temperature contrast across the barrier is not as large as might typically be found in a mid-winter event.
In summary, the synoptic situation was nearly ideal for a strong wave-type downslope windstorm. Strong cross-barrier flow, an unusually strong sea-level pressure gradient, and a variety of indicators pointing to the presence of a mean-state critical level were all present. Additionally, the reverse-shear found in the observations and analysis suggest a self-induced critical level was also possible.
The following links display various
observational data from the 4/23/99 event.
The Brigham City airport AWOS observations show the easterly winds picked up just after 0600 UTC and continued through the morning. It is interesting to note the measurable precipitation indicated during some of the strongest winds. The 113 mph gust was reported about 1/2 mile from the AWOS on a Davis-II recorder. Hill AFB observations are similar with light rain and snow reported through the wind event. Eye-witness reports confirm the precipitation during the downslope windstorm.
The strongest winds in Salt Lake county were reported at the University of Utah, where a 60 mph gust was reported at around 5 am. There were some large trees blown down in the Avenues part of Salt Lake City. Strong winds were not reported from Utah county or the Cache Valley in this event, although gusts to 74 mph were reported on Logan Peak.
Hat and Gunnison Island observations show that the strong easterly winds extended quite far downstream from the mountains in this event. Gusts of over 50 knots were reported from both sites.
Also of interest was that nearly a foot of snow fell in the mountains near Bear Lake and Monte Cristo, and 7 inches of snow in Randolf combined with 53 mph winds to produce blizzard-like conditions.
The following link is to a Deseret News article written just after the April 23, 1999 event, listing other significant "canyon" wind events in the past 40 years. The accuracy of this list is somewhat questionable as a check of the data from at least one event, May 23, 1970, indicates there was no windstorm on this date.
The following links display graphics from five other downslope windstorms that produced gusts in excess of 100 mph along the Wasatch Front. The graphics are generated from NMC gridded analyses, but on a rather course grid.
The 2000 event produced widespread damage with 19 rail cars
toppled in Davis county. The max gust in the valley was 86 mph in Davis County, but
the rail car damage indicates winds may have exceeded 100 mph. Peak gusts in the
mountains included 102 mph from the northeast at Hidden Peak and 83 mph at Francis Peak.
The 1983 event produced gusts to 104 mph near Hill
AFB with many reports of 60-80 mph from Utah county northward along the
Wasatch Front. The 1988 event produced gusts to 104 mph at Centerville
(unofficial estimates of 110 mph since 104 was the anemometer limit) with
numerous reports of 60-70 mph from Spanish Fork to the Idaho border.
The 1971 event toppled 8 railroad cars near Farmington. Although
the maximum reported anemometer wind was 88 mph, blown-over railroad cars
suggests winds likely exceeded 100 mph. There is a media report of
110 mph gust in Bountiful with the 1978 event.
The 2000 event had only a 9 mb sea-level pressure difference between SLC and Lander, but the cross barrier flow was quite strong. The sea-level pressure difference between SLC and Lander, WY reached 20 mb in the 1988 event and 13 mb in the 1983 event. This compares to just over 19 mb in the 1999 event. Only the analyses are available for the 1971 and 1978 events, so the exact pressure difference is unknown, but it looks relatively weak in 1971 and at least 20 mb in the 1978 event.
In the 2000 event, a critical level definitely exists in the 12z SLC sounding, although forecasts from the models did not show one developing. In the 1988 event, a critical level appears to exist near 500 mb, while in the 1983 event the easterly flow is deeper. There is strong cold advection at 700 mb in the 1983 event, while the cold advection is not as strong in the 2000 event and is more along just the northern Wasatch Front in the 1988 event. The 2000 and 1971 event are similar to the 1983, 1988, and 1999 events in that they represent a strong wave-type event with strong cross-barrier flow. The 1978 event is quite different. It appears to be driven only by the low-level pressure gradient as there is little or no cross-barrier flow and no cold advection at 700 mb.
As in the 1999 event, the cross-barrier flow is difficult to determine from raob data. The 700 mb wind on the SLC raob appears to be near 20 knots in the both 1983 and 1988 event, although as seen in the 1999 and 2000 event, this is unlikely to accurately represent the cross-barrier flow.
The Warning Issued for the April 1999 Event
The warning issued for the April 1999 event was very strongly worded. This excellent warning is probably as strongly worded as any High Wind Warning that has ever been issued by the SLC forecast office for a downslope wind event in that it explicitly mentions the possibility of 100 mph winds.