24-06-2014, 12:16 PM
WIND DAMAGE TO AIRPORT: LESSONS LEARNED
[attachment=65253]
ABSTRACT
High winds at a maximum speed of 96 mph (43 m/s) hit the
Columbia Regional Airport in Missouri on June 17,.1985, causing heavy
damage to parked aircraft, hangars, building glass windows, automobiles,
and so on. A post-disaster investigation reveals a wealth of
information, such as the finding that the storm was a microburst, and not
a tornado as it was originally classified, that the aircraft tiedown system
was flawed, that a gravel road was the principal source of damage to cars
parked at the airport terminal, that the gust factor of this type of wind is
much higher than normally assumed for structural design, and so on.
Additional findings are that the atmospheric pressure of the storm
measured was greatly affected by the wind-generated pressure of the
building in which the barometer was housed, and that west is the
predominant direction of high winds at this airport. Lessons learned
from the investigation can be very helpful in reducing future wind
damage at airports and in improving understanding of weather data
pertaining to severe storms and how to use these data in engineering
practice.
INTRODUCTION
On June 17, 1985, a severe storm packed with high winds and heavy rain
hit the Regional Airport at Columbia, Missouri. Heavy damage was caused
to 24 lightweight aircraft parked outdoors, to a hangar, to many parked
cars, and to certain other airport facilities. An investigation was carried out
by the writers to determine the nature of the storm and of the failure
mechanism of the major damages caused by the storm. Details of the
investigation are contained in a report (Liu and Nateghi 1986). Lessons
learned are summarized herein. As will be shown later, much valuable
information has been generated from the investigation. The lessons
learned, if publicized, can help reduce future wind damage to airports
through the development of improved mitigation strategies. Note that wind
damage to airports is not a rare event. For instance, in 1985 at least two
other airports in the United States also suffered wind damages that made
national news
WIND CHARACTERISTICS
The storm hit the airport at midnight and passed right through the
National Weather Service (NWS) Station Building at the airport, which
was equipped with sophisticated weather instruments. Meteorologists at
the station watched the storm closely on a radar screen but did not detect
any features of tornado. When the wind peaked, the building shook and
DAMAGES CAUSED
The only aircraft parked at the airport at the time of the storm were
lightweight aircraft—mostly single-engine planes. 24 planes were damaged
or destroyed. They were made up of 11 Cessnas, 2 Mooneys, and 11 Pipers
(10 single-engine Pipers and 1 twin-engine Piper). A typical damage scene
is shown in Fig. 2. Most of the aircraft damaged were parked outdoors; one
was damaged inside a hangar after the hangar door had failed.
Automobile Damaged
Practically all the automobiles parked on the north wing of the airport
terminal parking lot lost their windows and windshields. Typically, the
broken glass from car windows and windshields facing south and west
landed inside the cars, and the glass from windows facing north or east fell
outward, indicating that the damaging wind was from the southwest. A
close examination of the car bodies (exteriors) revealed that the windward
side of the cars had numerous tiny spots of paint damage on the car body,
indicating that the windward side of each car was blasted by small gravel.
It is envisioned that once the windward glass was shattered by the gravel,
the wind pressure pushed the broken glass into the car. It is quite possible
Storm Classifications and Internal Pressure
The storm was classified by the National Weather Service and reported
in the press as a tornado. Interviews with local meteorologists revealed
that because no tornado was sighted by eye and no tornado features such
as a hook echo were detected on the local radar, they were unsure and not
unanimous about the classification. One meteorologist thought that it
might have been a downburst instead of a tornado. Another thought that it
was not a downburst, but the formative stage of a small tornado.
The senior writer conducted a post-disaster investigation on the day of
the storm. Almost all the damaged objects observed, including trees, light
poles, a ticket booth, automobile glass, and so on, fell or moved toward the
east or northeast, indicating a pattern of straight line wind from the west or
southwest, covering the entire airport damage area. The only exception
was in regions near buildings where the wind was deflected by buildings.
Following the post-disaster investigation, a detailed investigation was
launched to determine why some meteorologists felt that it was (or could
have been) a tornado.
A review of literature showed that a downburst is a mass of cold air
sinking rapidly in a larger body of warm air. It induces an outburst of
damaging winds on or near ground level. The wind in a downburst can be
either straight or curved. According to Fujita (1985), downbursts can be
categorized into microbursts and macrobursts. A microburst has a small
horizontal scale, and has damaging winds lasting typically from 2-5 min.
On the other hand, macrobursts cover a larger area, and the damaging
wind last 5-30 min
Fastest-Mile Wind and Gust Factor
Although the peak wind speed measured in this storm was 96 mph (43
m/s), the fastest-mile-wind recorded was only 50 mph (22 m/s). This yields
a velocity gust factor of 96/50 = 1.92, and a pressure gust factor of 1.92 x
1.92 = 3.8—both are much higher than normally assumed in structural
design. For instance, using Fig. 2.3.10 of Simiu and Scanlan (1986), the
fastest-mile wind corresponding to a 1-s gust of 96 mph (43 m/s) is 78 mph
(35 m/s). This corresponds to a velocity gust factor of 1.23 and a pressure
gust factor of 1.51, which are much lower than observed in this storm.
A study is needed to determine how often high winds have sharp peaks
and gust factors as high as those in this storm. Should this be a common
occurrence for storms in any region, the gust response factor used in
contemporary structural design would need to be increased to insure
building safety. The gust factor used in structural design depends not only
on the ratio of the peak gust to the fastest-mile wind, but also on the
sensitivity of the structures to wind—thus the term "gust response
factor". This storm also raised questions as to the adequacy of using the
fastest-mile wind for structural design, for the damaging part of the storm
(see Fig. 1) lasted only a few seconds. Furthermore, since more and more
Wind Direction
The strongest wind in this storm was either from the west or southwest.
It was not possible to determine the wind direction more precisely because
the wind sensor used readings in only eight directions. Since it has often
been reported that most high winds in the Midwest are southwesterly, a
study was conducted to determine whether the high winds in Columbia
have preferred directions. To accomplish this, the NWS climatic data for
Columbia were used; they are listed in Table 3. The data covered a period
of 36 years (from 1950-1985). Note that all the strongest winds of each year
during this period, except in 1979, were either from the west, the
southwest, or the northwest. Due to the poor resolution of the wind
direction sensor mentioned before, it is clear that west must be the most
preferred wind direction. Therefore, in planning future airports and other
facilities, one should take this preferred wind direction into account. For
example, at the Columbia Airport and at other airports in the Midwest
having the same preferred westerly high-wind direction, airplanes should
be parked heading toward the west to minimize wind damage. Furthermore,
if aircraft are parked near the east walls of a building, they will be
protected by the building. Also, planting bushes on the west side of a car
CONCLUSION
The study shows that the thorough investigation of this storm has
yielded a wealth of valuable information including the following:
1. The main reason for severe damage to aircraft in this storm was rope
failure. The ropes used to tie the aircraft were greatly weakened by
weathering. Improper tiedown practices also may have contributed to rope
failure. FAA Advisory Circular AC-20-35C (1983) should be revised to
address the problems.
2. The main cause for the breakage of glass of cars and buildings was
gravel suspended by the high wind. The gravel came from a gravel road
adjacent to the airport.
3. The storm, officially classified as a tornado, was really a downburst.
This study shows that when no tornado is sighted by eye and no hook echo
is detected on radars, it is often difficult to determine without a thorough
investigation and analysis whether the storm is a tornado or a microburst.
In cases of uncertainty, there is a tendency for meteorologists and disaster
preparedness officials to classify the storm as a tornado and issue a tornado
warning because the general public takes tornadoes more seriously.
4. The sudden drop of the barometric pressure at the peak of the storm
was caused by wind-induced building internal pressure. Most meteorologists
do not yet understand the phenomenon and hence do not take internal
pressure into account when interpreting or reporting barographic results
[attachment=65253]
ABSTRACT
High winds at a maximum speed of 96 mph (43 m/s) hit the
Columbia Regional Airport in Missouri on June 17,.1985, causing heavy
damage to parked aircraft, hangars, building glass windows, automobiles,
and so on. A post-disaster investigation reveals a wealth of
information, such as the finding that the storm was a microburst, and not
a tornado as it was originally classified, that the aircraft tiedown system
was flawed, that a gravel road was the principal source of damage to cars
parked at the airport terminal, that the gust factor of this type of wind is
much higher than normally assumed for structural design, and so on.
Additional findings are that the atmospheric pressure of the storm
measured was greatly affected by the wind-generated pressure of the
building in which the barometer was housed, and that west is the
predominant direction of high winds at this airport. Lessons learned
from the investigation can be very helpful in reducing future wind
damage at airports and in improving understanding of weather data
pertaining to severe storms and how to use these data in engineering
practice.
INTRODUCTION
On June 17, 1985, a severe storm packed with high winds and heavy rain
hit the Regional Airport at Columbia, Missouri. Heavy damage was caused
to 24 lightweight aircraft parked outdoors, to a hangar, to many parked
cars, and to certain other airport facilities. An investigation was carried out
by the writers to determine the nature of the storm and of the failure
mechanism of the major damages caused by the storm. Details of the
investigation are contained in a report (Liu and Nateghi 1986). Lessons
learned are summarized herein. As will be shown later, much valuable
information has been generated from the investigation. The lessons
learned, if publicized, can help reduce future wind damage to airports
through the development of improved mitigation strategies. Note that wind
damage to airports is not a rare event. For instance, in 1985 at least two
other airports in the United States also suffered wind damages that made
national news
WIND CHARACTERISTICS
The storm hit the airport at midnight and passed right through the
National Weather Service (NWS) Station Building at the airport, which
was equipped with sophisticated weather instruments. Meteorologists at
the station watched the storm closely on a radar screen but did not detect
any features of tornado. When the wind peaked, the building shook and
DAMAGES CAUSED
The only aircraft parked at the airport at the time of the storm were
lightweight aircraft—mostly single-engine planes. 24 planes were damaged
or destroyed. They were made up of 11 Cessnas, 2 Mooneys, and 11 Pipers
(10 single-engine Pipers and 1 twin-engine Piper). A typical damage scene
is shown in Fig. 2. Most of the aircraft damaged were parked outdoors; one
was damaged inside a hangar after the hangar door had failed.
Automobile Damaged
Practically all the automobiles parked on the north wing of the airport
terminal parking lot lost their windows and windshields. Typically, the
broken glass from car windows and windshields facing south and west
landed inside the cars, and the glass from windows facing north or east fell
outward, indicating that the damaging wind was from the southwest. A
close examination of the car bodies (exteriors) revealed that the windward
side of the cars had numerous tiny spots of paint damage on the car body,
indicating that the windward side of each car was blasted by small gravel.
It is envisioned that once the windward glass was shattered by the gravel,
the wind pressure pushed the broken glass into the car. It is quite possible
Storm Classifications and Internal Pressure
The storm was classified by the National Weather Service and reported
in the press as a tornado. Interviews with local meteorologists revealed
that because no tornado was sighted by eye and no tornado features such
as a hook echo were detected on the local radar, they were unsure and not
unanimous about the classification. One meteorologist thought that it
might have been a downburst instead of a tornado. Another thought that it
was not a downburst, but the formative stage of a small tornado.
The senior writer conducted a post-disaster investigation on the day of
the storm. Almost all the damaged objects observed, including trees, light
poles, a ticket booth, automobile glass, and so on, fell or moved toward the
east or northeast, indicating a pattern of straight line wind from the west or
southwest, covering the entire airport damage area. The only exception
was in regions near buildings where the wind was deflected by buildings.
Following the post-disaster investigation, a detailed investigation was
launched to determine why some meteorologists felt that it was (or could
have been) a tornado.
A review of literature showed that a downburst is a mass of cold air
sinking rapidly in a larger body of warm air. It induces an outburst of
damaging winds on or near ground level. The wind in a downburst can be
either straight or curved. According to Fujita (1985), downbursts can be
categorized into microbursts and macrobursts. A microburst has a small
horizontal scale, and has damaging winds lasting typically from 2-5 min.
On the other hand, macrobursts cover a larger area, and the damaging
wind last 5-30 min
Fastest-Mile Wind and Gust Factor
Although the peak wind speed measured in this storm was 96 mph (43
m/s), the fastest-mile-wind recorded was only 50 mph (22 m/s). This yields
a velocity gust factor of 96/50 = 1.92, and a pressure gust factor of 1.92 x
1.92 = 3.8—both are much higher than normally assumed in structural
design. For instance, using Fig. 2.3.10 of Simiu and Scanlan (1986), the
fastest-mile wind corresponding to a 1-s gust of 96 mph (43 m/s) is 78 mph
(35 m/s). This corresponds to a velocity gust factor of 1.23 and a pressure
gust factor of 1.51, which are much lower than observed in this storm.
A study is needed to determine how often high winds have sharp peaks
and gust factors as high as those in this storm. Should this be a common
occurrence for storms in any region, the gust response factor used in
contemporary structural design would need to be increased to insure
building safety. The gust factor used in structural design depends not only
on the ratio of the peak gust to the fastest-mile wind, but also on the
sensitivity of the structures to wind—thus the term "gust response
factor". This storm also raised questions as to the adequacy of using the
fastest-mile wind for structural design, for the damaging part of the storm
(see Fig. 1) lasted only a few seconds. Furthermore, since more and more
Wind Direction
The strongest wind in this storm was either from the west or southwest.
It was not possible to determine the wind direction more precisely because
the wind sensor used readings in only eight directions. Since it has often
been reported that most high winds in the Midwest are southwesterly, a
study was conducted to determine whether the high winds in Columbia
have preferred directions. To accomplish this, the NWS climatic data for
Columbia were used; they are listed in Table 3. The data covered a period
of 36 years (from 1950-1985). Note that all the strongest winds of each year
during this period, except in 1979, were either from the west, the
southwest, or the northwest. Due to the poor resolution of the wind
direction sensor mentioned before, it is clear that west must be the most
preferred wind direction. Therefore, in planning future airports and other
facilities, one should take this preferred wind direction into account. For
example, at the Columbia Airport and at other airports in the Midwest
having the same preferred westerly high-wind direction, airplanes should
be parked heading toward the west to minimize wind damage. Furthermore,
if aircraft are parked near the east walls of a building, they will be
protected by the building. Also, planting bushes on the west side of a car
CONCLUSION
The study shows that the thorough investigation of this storm has
yielded a wealth of valuable information including the following:
1. The main reason for severe damage to aircraft in this storm was rope
failure. The ropes used to tie the aircraft were greatly weakened by
weathering. Improper tiedown practices also may have contributed to rope
failure. FAA Advisory Circular AC-20-35C (1983) should be revised to
address the problems.
2. The main cause for the breakage of glass of cars and buildings was
gravel suspended by the high wind. The gravel came from a gravel road
adjacent to the airport.
3. The storm, officially classified as a tornado, was really a downburst.
This study shows that when no tornado is sighted by eye and no hook echo
is detected on radars, it is often difficult to determine without a thorough
investigation and analysis whether the storm is a tornado or a microburst.
In cases of uncertainty, there is a tendency for meteorologists and disaster
preparedness officials to classify the storm as a tornado and issue a tornado
warning because the general public takes tornadoes more seriously.
4. The sudden drop of the barometric pressure at the peak of the storm
was caused by wind-induced building internal pressure. Most meteorologists
do not yet understand the phenomenon and hence do not take internal
pressure into account when interpreting or reporting barographic results