STORM TRACK: September 30, 1982 (Volume 5 Issue 6)

Back Index Fwd


By John F. Weaver

Most events, which manifest themselves as sudden and violent. deviations from the more normal behavior of nature, turn out, to be simply local adjustments to broader scale changes. Thus, if two large plates of the earth's crust are slowly drifting past each other over the centuries, and the two should 'stick' at some spot for a time (while the forces on the plates continue), eventually the jam will release with a little jerk, and the drift will go on. To the humans living nearby, that 'jerking loose' represents a potentially disastrous earthquake, throwing them to the ground, leveling their fragile structures, and entering their history books as an incident of great importance. Such is the case with the severe thunderstorm. In reality, what are thunderstorms except the result of transferring a portion of the earth's surface heat into the atmosphere? Or rather, random 'bubbles' of convection which spring up and disappear as part of the process. And a few of these bubbles may create miniscule eddies in their wake, as they rise. Perhaps tornadoes are nothing more than just that. However, to us these tiny circulations are of great importance. Furthermore, as impossible as it sounds, the thunderstorm forecaster must take on the job of alerting the rest of us when one of these random little eddies is about to occur.


It is a difficult, decision as to where the history of thunderstorm forecasting (as opposed to forecasting in general) begins. Aristotle deduced a few facets of the thunderstorm's nature. For example, he realized that the upper atmosphere is cooler than the air near the surface, that cooling causes air to 'turn to water,' and that this process produces clouds. However, Aristotle's deductions concerning storms were, when considered in total, more wrong than right. Uncountable proverbs permeate 'weather lore' regarding the forecasting of rain and thundershowers. However, weather lore is simply a collection of regional climatology, and, as has been "pointed out in previous sections, climatology breaks down when used to predict weather on a day to day basis. As the old proverb goes: 'Climatology is what is expected ...Weather is what happens instead.' At what point then should our history begin? I think I shall (arbitrarily) revert to my argument in the first section; namely, that once the motivating force for a natural event is understood, then true prediction becomes possible. And since this approach wasn't taken until the time of the Renaissance, it is there that I shall begin this part of the story.

Even after we've established the period in which the history begins, the 'first' relevent discovery is difficult to identify. Was the development of meteorological instrumentation (beginning in 1450 A.D.) the true debut of modern thunderstorm forecasting? Perhaps. When van Helmont, in 1635, showed that water vapor was a gas different from air, did we begin to understand enough to predict more accurately? Probably not. What about Boyle's work in the 1650s or Isaac Newton's laws of motion (1687)? Well, all of these discoveries, as well as many others, were probably requisite to the beginnings of scientific meteorology, of which thunderstorm forecasting is a small subset. However, if I were to have to choose one event, a discovery that applied to the most unique characteristic of the thunderstorm, my choice would be Joseph Black's discovery (in the 1670s) that condensing water gives off heat. Latent heat. This knowledge gave us the mechanism of buoyantly driven convection.

As was mentioned in an earlier section, it required the latter half of the 1700s and the early part of the 1800s to begin to evolve a conception of the general nature of the vertical structure of the troposphere. With this knowledge and with careful observation of many thunderstorms, it wasn't long before meteorologists began to understand storms a little better. In 1841, James P. Espy of the Franklin Institute wrote a book entitled 'The Philosophy of Storms,' which was, by any definition, the first work to logically and scientifically describe the mechanism by which the thunderstorm forms and maintains itself. Some of the more important points made with regard to thunderstorms include:

1. The thunderstorm forms when ascending currents carry vapor laden air to higher altitudes. Expansional cooling then causes condensation to form and latent heat to be released. This, in turn, causes the updraft air to be buoyant relative to its environment.

2. The height of the cloud base depends, therefore, on the dewpoint depression.

3. Environmental air continues to 'feed' the storm throughout its lifetime.

4. There is often a lowered cloud base under the most intense updraft (which he felt was due to a relative pressure deficit beneath the updraft).

5. The nearby environment is 'stabilized' by the storm.

6. The upper, drier environment mixes with the storm (which he deduced from the shape of congestus clouds).

7. Moist air can be forced to form clouds if it is moved to higher terrain by advection.

8. Hail is formed only if the updraft is strong enough to carry raindrops above the freezing level.

9. Storm motion is governed by the flow aloft.

10. Rain-cooled air produces downdrafts and outflow boundaries. Often, new convection will form via forced lifting at these outflow boundaries.

11. Fog is a cloud on the ground.

12. Following a 'cold wave' (i.e., cold front passage), convection is suppressed due to stabilization of the lower layers of the atmosphere. With regard to other scales of action, Espy suggested that the forced upward motion near the center of extra-tropical cyclones helps convection to form there, and that the additional, buoyantly-driven upward motion of thunderstorms might (conversely) help intensify the cyclones. All in all, not a bad set of deductions from a fellow that had only the most limited amount of data to work from.

On February 9, 1870, the United States Weather Bureau was created by an act of Congress and was put under the control of the U.S. Signal Service. By 1878, 284 field stations were telegraphing weather reports to Washington, D.C. three times a day. The weather observations were collected by military personnel at each station, who soon became some of the most knowledgable observational meteorologists of all time. One of these, a Sergeant James Park Finley became fascinated with severe thunderstorms and tornadoes, and wrote several definitive articles on the subject during the 1880s. Among the many important, characteristics noted by Finley were the following:

1. Tornadic storms seem to form on boundaries separating warm from cold air masses;

2. These boundaries (and, therefore, severe storms) usually occur near the center of extra-tropical cyclones;

3. Severe storms most often occur between four and six PM; and

4. A large-scale circulation in the surface wind field is normally present, and it is likely that the severe storm converges this broad, gentle circulation into a small, violent one when the tornado is formed.

These findings constitute but a few of Finley's observations and form a set of important, contributions to the science of thunderstorm forecasting. Notice also, that Finley talked about 'fronts' (in all but name only) thirty to forty years before the Norwegians 'air mass and frontal analysis' method of forecasting.

Let's pause a moment and look at the thunderstorm forecast process at the end of the nineteenth century. Observations of surface weather conditions were being transmitted to Washington regularly. In Washington, surface weather maps were constructed three times a day. However, there were no facsimile or teletype circuits, so that these finished products could not be transmitted in very convenient format to any field stations. Thus, forecasts for the whole nation were prepared in Washington. By 1901, worded forecasts were being disseminated daily by telephone, telegraph and even by mail to nearly 80,000 users" (illustration).


In out-lying parts of the country, forecast maps could be constructed by plotting the telegraphed positions of lows and highs along with regions of forecast precipitation. This type of forecast, map was appearing in many public places around the nation, before the turn of the century.

The forecasts themselves, were based only on the rudimentary knowledge outlined above. It was known that inclement weather was associated with extratropical cyclones (indicated on maps as lows), and that the deeper the central pressure, the stronger were the accompanying winds. Also, deeper lows generally produced more precipitation. It was known that these systems pulled cold air southward in their rear quadrant and warm air north in front. Thus 'cold waves' could be forecast, as well as periods of unseasonably warm temperatures. Large regions, expected to be generally cloudless in summer, could be expected to develop thundershowers, since it was known that strong heating could provide sufficiently rapid ascending currents of air. Finley showed that stronger storms formed near the center of these systems and along boundaries separating warm from colder air masses, so that these regions could be watched for intense development. Climatologically 'normal' temperatures had been established for most cities, and -in the absence of any significant air mass changes, these could be used to provide guidance on general temperatures. However, in terms of area or intensity- specific thunderstorm forecasts, it is evident that no reliable guidance could be prepared. I would guess that the best one could do at this time was to be aware of approaching weather systems, and then apply as much as was known regarding observational meteorology. In terms of a three to six hour forecast, this method serves fairly well even today, depending on the observer's skill.

The skill of the individual. This, of course, is the problem. Most people that have a vested interest in knowing when severe storms are about to strike don't have the time to learn these skills. Not everyone can learn them. The problem that awaited the twentieth century, then, was how to forecast strong thunderstorm activity so that concerned segments of the public could be alerted in a timely fashion, without requiring considerable training on their part.