Skip Navigation

September-October 2013

ResizeResize Text: Original Large XLarge

Weather Queries

Q: I often look at the Storm Prediction Center's Web page to check current weather conditions and the outlook for hazardous weather for the next several days. My question has to do with storm-relative wind at the anvil level of a thunderstorm. Of what value is this?

Robert Skurski

Chicago, Illinois


A: The Storm Prediction Center is interested in storm-relative winds at the anvil level of thunderstorms, in particular, supercell thunderstorms, which have a rotating updraft and often produce severe weather on the ground: damaging winds, large hail, and sometimes tornadoes. Some meteorologists look at the storm-relative winds (to be described) at altitudes of 9-11 kilometers, a layer high in the troposphere where many thunderstorm anvils spread out. Alternatively, meteorologists key in on the equilibrium level—a level computed from a temperature and moisture sounding obtained by a radiosonde balloon or from a model forecast of temperature and moisture. If a thunderstorm develops on a particular day, the equilibrium level will always cut through the anvil. In this sense, the equilibrium level is a more reliable estimate of anvil altitude than the climatological average of 9-11 kilometers.

A thunderstorm forms when condensation of water vapor inside a growing cloud releases latent heat within the updraft. If this makes the updraft warmer than air outside the cloud at the same level, the updraft becomes buoyant and rises of its own accord, much like a hot air balloon. In fact, the updraft accelerates. The decrease of temperature with altitude outside the cloud determines how long the updraft remains buoyant and how high a towering cumulus cloud can grow. On days when the atmosphere is unstable, the cloud can grow all the way to the tropopause, the top of the troposphere, which is always marked by a stable layer of air, in which the temperature stops decreasing with altitude. Once the updraft encounters this layer, it quickly becomes colder than air outside the cloud and decelerates. The top of the thundercloud then spreads out in the characteristic anvil shape. The altitude at which the temperature in the updraft first becomes equal to the temperature outside the cloud is called the equilibrium level.

To calculate the storm-relative winds at anvil level, one needs the vector wind (speed and direction) at anvil level and the storm motion vector. The two most common choices for the vector wind at anvil level are the average vector wind in a one-kilometer-deep layer or a 50-millibar thick layer, both centered on the equilibrium level. The average can be computed from an observed wind profile or from a model forecast. The next step is to find the storm motion vector. For an existing supercell storm, that is determined by the motion of the storm cell as measured by a Doppler radar. For supercell storms that have not yet formed, one must estimate the storm motion from a measured or predicted vertical wind profile. To estimate the storm motion ahead of time, it is customary to assume a direction of motion 30 degrees to the right of the 1000-500 millibar mean vector wind and a speed of 75% of the mean wind speed in the same layer. This estimate is subject to error because wind profiles may change substantially in the few hours preceding severe storm development, and, in rare instances, a rotating updraft storm may move to the left of the mean wind instead of to the right.

To find the “storm-relative wind near the anvil level,” we take the vector difference between the mean wind at anvil level and the storm motion vector and then find the magnitude of the vector difference, which is a wind speed. This number is meant to discriminate among supercell types. In general, anvil-level storm-relative winds less than 40 knots have been associated with “high precipitation” supercells, 40-60 knot storm-relative winds with “classic” supercells, and greater than 60 knot storm-relative winds with “low-precipitation” supercells. Why should this be so? The working hypothesis is that strong high-level shear, represented by the greater-than-60-knot criterion, will carry much of the frozen condensate in a supercell storm downstream in the anvil, away from the column of precipitation. When high-level shear is relatively weak (the less-then-40-knot criterion), the frozen condensate may fall back into the storm, effectively seeding it and thereby enhancing precipitation efficiency. These rules of thumb work best for isolated supercells. If the anvil from one supercell merges with another directly downstream, one supercell could “seed” its neighbor.

I thank Howard B. Bluestein, at the University of Oklahoma, for help in answering this question.


Q: I have seen many storm-chasing videos on tornadoes and have noticed that some tornadoes have rain, lightning, and hail falling, and others have a nearly clear sky in the background with no precipitation. What makes the difference?

Robert Skurski

Chicago, Illinois


A: The strongest tornadoes usually form within what are called supercell thunderstorms—thunderstorms that have rotating updrafts. Some supercells produce very heavy precipitation, while others produce very little. The preceding question and answer dealt with the effects of high-level wind shear, in particular, the vector difference between the ambient wind at the anvil level of the storm and the storm's motion with respect to the ground. This parameter is also called the storm-relative wind at the anvil level. In “wet” storms, when the value of this parameter is less than 40 knots, the precipitation will often wrap around the tornado, obscuring it from view. In “dry” storms, when the value is greater than 60 knots, there is sometimes scant precipitation, and the tornado is much easier to see. The amount of moisture in the air entering the updraft below the cloud base, as indicated by the surface dewpoint, has little effect on whether a supercell is wet or dry.

Many waterspouts and weaker tornadoes form without a supercell being present. If there is rotary motion in the wind near the ground and a strong updraft develops in a towering cumulus cloud above this rotary motion, the low-level air is literally sucked toward the base of this cloud, forcing the radius of rotation to shrink and sometimes spinning up a small tornado, even without precipitation reaching the ground. Such tornadoes may not be easy to see, especially early in their formation, because there is no condensation funnel. The first indication of a tornado may be whirling dust and debris raised from the ground.

The location and pointing direction of the camera can determine whether a dark tornado is viewed against a bright sky, perhaps the back side of an anvil in sunlight, or against the dark precipitation shaft of the nearby parent thunderstorm.


Q: The summer of 2013 marks the 10th anniversary of a tornado (July 21, 2003) that destroyed an old railroad bridge in Kinzua Bridge State Park, Pennsylvania. Is it possible that the 3,357-ton Kinzua Viaduct, once called the Eighth Wonder of the Engineering World, could be the largest, most massive man-made structure ever to be toppled by a tornado? Both ends of the structure remained standing afterward, though most of Kinzua's steel lay in ruins on the floor of the gorge. The bridge has been rebuilt and is now a spectacular skywalk. It would be a major selling point for tourism if it could be confirmed that the viaduct was the single most massive manmade structure ever to be toppled by a tornado.

Nathan S. Clark, Jr.

Greenville, Pennsylvania


A: It is impossible say whether a large section of the Kinzua Bridge, destroyed by the 2003 tornado, was the longest, largest, or most massive object ever toppled by a tornado, but at least it is in contention, as will be seen.

The Kinzua Bridge was built of wrought-iron trusses and towers in 1882 as part of a shortcut for transportation by rail of coal, oil, and lumber. The viaduct was 2,053 feet long, spanning the Kinzua Gorge, and stood 301 feet above the gorge near its center. By 1900, rail loads had markedly increased. This forced the original bridge to be dismantled and for a stronger one to be rebuilt with structural steel. Towers for the new viaduct were mounted on the same masonry pedestals used for the original one, and the anchor bolts, a point of failure in the 2003 tornado, were reused. By the middle of the 20th century, most of the regional coal had been mined, and rail traffic decreased. In 1957, the railroad company sold the viaduct for scrap. The new owner, apparently having second thoughts, sold it intact to the State of Pennsylvania in 1963. In 1970, Pennsylvania created a new state park with the Kinzua Bridge as the main tourist attraction. A private concessionaire operated an excursion railroad across the bridge until early 2002, when inspections revealed serious deterioration, and the state closed the bridge for repairs.

Repairs were still under way when an F-1 tornado on the Fujita Scale destroyed more than half of the bridge, sending it to the bottom of the gorge. Twenty-three of its 41 spans and 11 of its 20 supporting towers were demolished. See Figure 1.

Thomas Leech, principal investigator for the forensic assessment of the bridge failure, published findings in the July-August 2005 issue of American Scientist (pp. 348-353). According to him, the twisting winds of the tornado caused failure of the bolts attaching towers of the viaduct to concrete pedestals. The wind pushed the towers off their bases. The towers leaned over and, within seconds, collapsed entirely, one by one. Engineers estimated that the critical wind speed for failure was about 94 mph.

If the Kinzua Bridge is the longest, largest, or most massive structure ever toppled by a tornado, there are extenuating circumstances, namely, the bridge was vulnerable to strong winds from the start, and had been weakened by years of rust and metal fatigue. But there may be other contenders for this dubious honor.

On May 11, 1953, an F5 tornado, 23 miles long, passed through downtown Waco, Texas. One building, the five- or six-story (depending upon the information source) R. T. Dennis furniture store, of brick construction, crumbled to the ground. Bricks filled the adjoining street up to five feet deep. Thirty people, mostly employees, were trapped inside and perished. This was undeniably a massive structure, but its weight is unknown. Of interest, the 22-story office building across the street, now known as the ALICO Building, of reinforced steel construction, withstood the winds intact.

On May 24, 2011, a long-track, EF-5 tornado struck Cactus Rig 117, a 1.9-million-pound (950-ton) drilling rig in Canadian County, Oklahoma. Figure 2 shows the result. Constructed mostly of steel, this massive structure was toppled and rolled. A 20,000-pound oil tanker was moved one mile from the same location, and was apparently airborne for most of the distance. An Oklahoma mesonet station recorded a wind of 151 mph from the edge of this tornado. Using a mobile Doppler radar, Howard Bluestein and his students at the University of Oklahoma measured winds of 100-120 meters per second (224-268 mph) not far off the ground near the downed rig.

Thanks to Joseph Golden, Don Burgess, and Jim LaDue for help in answering this question.


Q: I was looking at data logged by my weather station over the past few years and found something curious: The swings in barometric pressure from day to day are less in July than in January. This doesn't make sense to me. Why wouldn't the pressure swings from day to day be greater during the summer, when there is more energy in the atmosphere?

Steve Dryja

Muskego, Wisconsin


A: Dry static energy in the atmosphere is defined as cpT + gz. In this expression, cp is the specific heat capacity of dry air at constant pressure, or the amount of energy in Joules (J) required to heat one kilogram (kg) of dry air by one degree Kelvin (K) while the air pressure is held constant. The value of cp is 1005 J kg−1 K−1. T is the temperature, expressed in degrees Kelvin. [The temperature in K is 273 degrees higher than the temperature in degrees Celsius (°C)]. g is the acceleration due to the Earth's gravity: 9.8 meters per second per second (m s−2). z is the height above sea level in meters.

Given this definition, the dry static energy of a volume of air depends on its temperature and its height above sea level. In the mid-latitude troposphere below 300 mb, it is almost always true that dry static energy is greater in July than in January because, if one picks a specific temperature in an atmospheric sounding, it will almost invariably be found at a higher altitude in midsummer than in midwinter. If you are referring to dry static energy in your question, you are correct. However, dry static energy has little to do with variations in surface pressure.

To a good approximation, the surface pressure (the force per unit area exerted on the ground by the air) is due to the weight of the overlying air. Imagine the total mass of air in a vertical column. If mass is removed from the column, the surface pressure drops; if mass is added, the surface pressure rises. But what physical process can add or remove air from a vertical column? Basically, wind.

Wind moves the air around. The temperature contrast between pole and equator gives rise to wind—in the upper troposphere, to the jet stream, a ribbon of high-speed air. In summer, when the pole is bathed continually in sunlight, the temperature contrast across mid-latitudes is much less than in winter, when the pole is in constant darkness. Thus the jet stream is mostly absent from mid-latitudes in summer, but it can become quite strong in winter. Examination of any weather map in winter reveals a jet stream circling the hemisphere, sometimes as a single, very strong current and sometimes split into northern and southern branches, but always displaying embedded waves. The largest waves tend to move slowly, while the smaller waves travel quickly, sometimes amplifying and slowing down and sometimes straightening out and speeding up. These waves in the jet stream are associated with divergence and convergence; that is, they tend to remove mass from air columns as they approach and add mass to air columns as they recede. In response, surface pressure drops in advance of a wave and rises after the wave passes by. The parade of surface low- and high-pressure systems in the mid-latitudes in winter is a consequence of the passage of waves (also called troughs and ridges) in the jet stream.

In summer, when the temperature contrast between north and south diminishes, the jet stream weakens and usually moves north into central Canada. The waves in the jet stream are no longer as pronounced or energetic as in winter, and so the day-to-day variations in surface pressure become less noticeable.

Weatherwise Contributing Editor THOMAS SCHLATTER is a retired meteorologist and volunteer at NOAA's Earth System Research Laboratory in Boulder, Colorado. Submit queries to the author at, or by mail in care of Weatherwise, Taylor & Francis, 325 Chestnut St., Suite 800, Philadelphia, PA 19106.

In this Issue

On this Topic

© 2010 Taylor & Francis Group · 325 Chestnut Street, Suite 800, Philadelphia, PA · 19106 ·