Science Briefs

Why Do Things Go Bump In Your Flight?

"The captain has just turned off the seat-belt sign." You settle down to the meal the flight attendant is serving, when — whump! The plane jolts without warning, as if it has just hit a giant airborne pothole. You and your dinner lurch toward the ceiling! Covered with airline food, you look out the window, but there's not a cloud in sight. What happened?

Figure 1
CAT reports as a function of absolute vorticity, a measure of the total spin of the atmosphere at a certain location. The x-axis is cumulative percentage of turbulence reports. The y-axis is the ratio of frequency of CAT occurrence vs. the background level [15%]; y values greater than 1 indicate higher-than-background concentration of CAT reports.

Bumpiness during a plane ride is called "turbulence", which is due to the up-and-down motions of the air the plane is flying through. Thunderstorms cause a lot of turbulence, but bumpy flights can and do occur in blue-sky conditions — hence the catch-all name of "clear-air turbulence," often referred to as CAT.

What causes CAT? The prime suspect is a type of atmospheric motion known as a "gravity wave." Just as ocean waves make a boat bob in the water, atmospheric gravity waves cause the up-and-down motions that can cause turbulence. Furthermore, gravity waves can snake through clear skies as easily as through cloudy skies.

But what, in turn, causes gravity waves? Since the 1960s, researchers have focused on two types of gravity waves: those caused by winds blowing over high mountains and those caused by sharp changes in wind speed and direction near weather fronts. There waves do indeed contribute to CAT, but there is one small problem: CAT keeps popping up in places away from mountains and fronts. As a result, aviation weather forecasters in 1997 continue to have a difficult time forecasting CAT. How can we improve things? Perhaps there are other ways to create gravity waves in the atmosphere that have been ignored?

In my research, I have investigated the roots of modern clear-air turbulence theory to look for missing pieces of the overall puzzle. I found that nearly all current methods of forecasting CAT assume that the weather situation causing the turbulence is a front, which is an area of low pressure. But this is a leap in logic. Although low-pressure areas are the primary weather-makers, they are not necessarily the only ones. Up to 20% of CAT reports, I discovered, come from strong high-pressure areas! Furthermore, highs are not merely mirror images of lows. There is a special type of instability in the atmosphere that happens only with very strong highs, called "inertial instability". This instability is not well known in the aviation forecasting community, but is becoming a "hot" topic in the field in stratospheric research, which was the subject of my Ph.D. thesis. As highs approach the onset of this instability, they have the potential to cause a lot of gravity waves.

Figure 2
Arakawa's turbulence measure at 43°N as a function of total atmospheric spin (divided by f, twice the Earth's rotation rate at 43°N) for the case in which 90% of the local spin is due to change in wind speed from one place to another. Notice that turbulence should increase for very strong lows and very strong highs.

The first figure shows the results of an observational study of CAT that, unlike many other studies, included cases of both strong highs and strong lows. The right-hand side of the graph shows that strong lows are preferred regions of CAT, which all pilots know. But the left-hand side shows that very strong highs are also preferred regions for CAT! Although very strong highs are rare, regions near them had the highest rate of CAT occurrence. This result was noted by the original authors but then left by the wayside and ignored for the past 20 years, until now.

But one observational study doesn't prove a hypothesis. I then unearthed and cleaned up some simple but long-neglected theory by a Japanese scientist named Arakawa that addresses the subject of turbulence (see Figure 2). This theory predicts the same kind of relationship between lows, highs and CAT as Figure 1 -- turbulence should be most intense for the strongest lows and highs!

This research is still a work in progress, though. Now that observational and theoretical support exists for my hypothesis, numerical modeling work will need to be done to confirm the relationship between very strong highs and turbulence. If that research is successful, work can then begin to translate this approach into a workable turbulence forecasting method.

As a relative novice to aviation forecasting I am pleased that I have been able to use my knowledge of an obscure atmospheric instability to help solve the vexing problems of turbulence forecasting. At the conference where I first presented this research, a senior member of the aviation forecasting community came up to me and said, "Thank you. The worst turbulence I ever experienced in a plane was on a sunny day over the Great Plains, and now I know why it happened."

Reference

Knox, J. A. 1997. Possible mechanisms of clear-air turbulence in strongly anticyclonic flows. M. Weather Rev. 125, 1251-1259.

Knox, J. A. 1997. Do we understand clear-air turbulence in anticyclonic flows? Preprints, American Meteorological Society Seventh Conference on Aviation, Range and Aerospace Meteorology, Long Beach, CA, 202-205.