Engagin' Science

Tetsuya "Ted" Fujita, Mr. Tornado

Tetsuya Theodore “Ted” Fujita, photo by Roger Tully, PBS.org

Tornadoes are incredibly powerful weather events. They can appear as just one funnel or multiples. Studying how they work can be dangerous research. Japan is known for its violent phenomena such as earthquakes, tsunamis, and volcanic eruptions, but not tornadoes. Tetsuya Fujita's curiosity put him on the map as an expert in such weather, but most of his work took place in the United States. He developed a ranking system (the Fujita Scale) for the intensity of tornadoes which was put in use in the early 1970s.

Tornado forecasting in the U.S. had begun around the 1800s. In 1884, John Park Finley worked for the U.S. Army Signal Corps and used 1,000 people to spot weather conditions that might precede a tornado. In 1887, the Corps banned the use of the word tornado from any forecasting in order to prevent panic, and that ban lasted until 1950. After the Corps was taken over by the U.S. Weather Bureau in 1890, Finley left. Although much data had been collected when tornadoes started, no one had put together a sound theory for how they were created. Ideas ranged from the Earth's distance from the sun, to sunspot activity, various electrical hypotheses, and the unclear notion of air currents with different temperatures colliding. To quote meteorologist Edward M. Brooks: "a major problem in explaining the formation of a tornado is to find the source of the potential energy and the manner in which it is converted to kinetic energy".

John Park Finley (1917, Wikimedia) and his 1887 map of known locations of tornadoes in the U.S. (The Weather Doctor)

Tetsuya Fujita was born on October 23, 1920 in a small town of Sone-machi (a suburb of Kitakyushu city) in the northern part of the island of Kyushu. His childhood interests varied around science topics such as asronomy (especially the tidal effects of the moon), catching clams on the beach, topographic mapping of the seacoast cliffs, and tracking sunspots with a homemade telescope. He graduated from Kokura Middle School in 1939 and won the Science Award for all of his efforts.

Family picture with Tetsuya at 14 on the right; (right) aged 19 ready to enter college

(images from Fujita's memoirs)

He then entered Meiji College of Technology, even though he was also accepted at the Hiroshima College of High School Teachers, on the basis of his father. Tetsuya majored in mechanical engineering (ME) and was a part-time assistant to Professor Tadaichi Matsumoto, who was in the Geology Department. A research project Matsumoto directed him to do was mapping aerial views of four volcanic calderas in the area, which was accomplished easily due to his earlier interest in topography. His own thesis was under Professor Hajime Nakagawa in the ME department. Tetsuya measured the impact force of steel balls hitting the ground, and his thesis was published in English and German. He learned both languages by dictionary translation of reference books from Prof. Matsumoto.

He graduated 6 months early in 1943 when the university changed its policy so that graduating students could enter military service during World War II. He became a full-time assistant in the Physics Department, and he was promoted to assistant professor in just one month.

Fujita's interest in meteorology came about in odd ways.

In 1944, the Navy contracted him to determine location of enemy planes using 3D triangulation with searchlights. He learned about bending of the beams in certain weather conditions.
Assigned to research on coal mines in 1945, he measured flucturating temperature and barometric pressure at points in mine shafts. These events caused him to become more interested in meteorology.
He even used his knowledge to estimate the altitude of the fireball from the atomic bombs dropped on Hiroshima and Nagasaki.

Fujita studying Nagasaki damage: his triangulation method for the Nagasaki fireball altitude calculation

(Images from Fujita's memoirs)

Fujita got a grant in 1946 to reeducate elementary school teachers, and he chose weather science as the topic because, as he wrote in his memoirs: "it could be studied rather cheaply with pencil and paper". After collecting data from a weather station, he created 200-300 booklets per month for the teachers. The weather maps created from temperature and pressure readings fascinated him, but the direction and speed of winds during certain storms didn't fit basic laws of motion that he had learned.

In 1947, despite taking measurements from the base and top of a mountaintop at a weather station and writing a research paper about micro-gusts of wind every 10 minutes (micro-analysis), he determined that the wind sometimes gusted strongly downward. As he wrote in his autobiography, "nobody in Japan in 1948 thought about a downard current...in a thunderstorm". He presented his data in a paper titled "Thunder-Nose", but his work received no recognition.

Graph from his paper "Thunder-Nose". Note the red downdraft created in the bigger storm updraft.

This shows a cross-section of the moving storm at altitudes above ground, and the storm moves from right to left. So, the strongest winds occur in the downdraft ahead of the storm.

On September 26, 1948 he surveyed 9.8 km-long damage from a tornado for his first time, walking the length of the path of destruction and observing how it affected buildings and rice fields. He spent a year collecting data from weather stations in the area (more micro-analysis), too, and published it in 1951 in English in Geophysical Magazine. He'd had to borrow money to buy a typewriter, and he translated it himself and typed it all with one finger.

After a presentation to the Fukuoka District Weather Service on the Thunder-Nose data in 1949, one of the meeting participants said he had found a 1942 paper on a similar topic in the garbage can of a U.S. Air Force installation at the top of the mountain where Fujita had collected his own data! During his doctoral studies at Tokyo University on typhoon damage in Kyushu, he sent his earlier journal articles to the author of that paper, Dr. Horacy Byers, at the University of Chicago. Byers was amazed at how detailed Fujita's data were despite much less sophisticated equipment at hand, and responded, "This problem is attracting a great deal of attention in the United States at the present time, and the U.S. Weather Bureau has a special project to investigate these smaller disturbances". He continued, "I have looked over your paper, Micro-analytical Study of Thunder-Nose, and note that in view of the fact that you were not familiar with the work of the U.S. Thunderstorm Project on this subject your conclusions are highly valuable and really represent an independent discovery of some of the factors derived from our work. In particular you deserve credit for noting the importance of the thunderstorm downdraft and outflowing cold air".

Seburi-yama weather station of the USAF, where Dr. Byers' 1942 paper was found.

Fujita finished his doctoral program in 1953 with a thesis titled "Analytical Study of Typhoons". (Typhoons are hurricanes in the west Pacific.) Byers invited him to Chicago to study severe storm phenomena in the U.S. with him.

Fujita in 1950 at age 30 while attending the Kyushi Institute of Technology (From his memoirs)

For the next few years, Fujita (who had now adopted the name Theodore, or "Ted) worked with Byers in Chicago and Dr. Morris Tepper who was doing research at the U.S. Weather Bureau in Washington, D.C. Following a tornado outbreak in Kansas and Oklahoma in June 1953, he applied his style of analysis on many aspects of the weather records before and during the storms. Sharp dips in the barometric pressure data were termed "mesocyclones". The prefix "meso" refers to Fujita's "mesoscale" analysis (mesoanalysis) of the weather data including

the area within a storm
areas around or ahead of storms

Contrast that to a huge scale (100s to 1,000s of km wide) and a microscale (just a few kilometers). A mesoanalysis takes in data between the size of those areas. Fujita felt they were the most important.

Byers asked Fujita to study the photographic evidence from a June 1957 tornado in Fargo, North Dakota that had damaged 1,300 homes. Over 2 years, he pieced together 150 photos from 53 ground sites to plot the path of the tornado and cloud features at one-minute intervals, and from them he was able to identify the storm wall cloud (from which tornadoes often descend) and tail cloud.

Fargo tornado, June 1957 (Wikipedia)

In 1965, he studied aerial photos of 36 tornadoes in the Midwest and noticed that the tornado tracks paralleled each other in "families". That is, where one tornado damage path ended, another one began a few miles east-northeast of it. He also noticed patterns of damage in corn fields which led him to believe that a tornado had multiple smaller ones inside it rotating around the center.

Fujita's model of multiple mini-tornado gusts inside a single one (uchicago.edu)

These smaller cyclones would swirl around the center in seconds adding 100 mph to the main tornado's winds and cause specific damage patterns, which Fujita recognized. They were confirmed by aerial photos like the one below. He said this pattern (which Fujita named cycloidal marks) might explain why one house could avoid damage while others nearby would be hit.

(Left) Notice the two curls on the right in this Magnet, Nebraska tornado path from 1975 (PBS).

(Right) Fujita's photo from 1970 indicating the cycloidal marks (Monthly Weather Review).

A multi-vortex tornado which generates cycloidal marks. (weather.com)

Moreover, Fujita's keen eye for detail also noted different levels of destruction of houses. At that time, hurricanes followed a 12-point ranking of wind speeds and sea conditions called the Beaufort Scale, but the maxium for that was 73 miles per hour (64 knots).

Beaufort Scale (Wikipedia)

Since tornado winds are much stronger, Fujita created his own scale in 1971, which he called the Fujita Scale or F-scale. Here it is compared to the Beaufort Scale and the Mach scale for speed of sound. The original F-scale was later modified in 2007.

Comparison of Fujta Scale (tornadoes), Beaufort Scale (hurricanes), and Mach scale (speed of sound) (Wikipedia)

Original Fujita Scale and the 2007 Enhanced Fujita Scale

Keep in mind that Fujita's analysis was done at a time when there were no weather satellites and no radar stations to monitor weather. His work was done all by gathering empirical evidence from ground and aerial photos, weather records, and descriptions of damage. Fujita visited 300 sites himself and took ground and aerial photos, plus conducted interviews with survivors and emergency teams.

(Top left & right) Fujita inspecting tornado damage (KPBS video)

(Bottom) Fujita taking photos from a plane (KPBS)

He also built a tornado simulator using dry ice for laboratory analysis at the University of Chicago.

Fujita's tornado simulator (YouTube)

Fujita with his tornado machine up close (WOUB Public Media)

With his accumulated knowledge growing, he was able to identify patterns called downbursts, which were sudden strong (62 km/hr, 39 mph) downdraft winds from thunderstorms, such as the one he suspected caused the crash of Eastern Airlines Flight 66 at John F. Kennedy Airport in 1975. He later refined them in 1981 as macrobursts and microbursts.

Macroburst: winds up to 188 km/h (117 mph) spreading in a path >4 km (2.5 miles) wide and lasting from 5 to 30 minutes
Microburst: winds ~270 km/hr (170 mph) <4 km (2.5 miles) in diameter and lasting <5 minutes

Result of a microburst, leaving a focused pattern of tree damage (Fujita, 1978)

Although Fujita had tried to distuinguish these types of air movement with specific terms, they still fell under a collective title of wind shear. Many meteorologists did not believe in his concept of downbursts, but eventually he collected enough data from around the country to show them that they were real phenomena. His work led to better pre-flight checks on commercial aircraft.

From 1976 to 1978, he received funding for project NIMROD (Northern Illinois Research on Downburst), and then joined a team called JAWS (Joint Airport Wind Shear) in Colorado. It wasn't until June 12, 1982 while working on JAWS that he actually saw his first tornado! It is said that after he discovered downbursts, he almost never flew without being invited to the cockpit to meet with the flight crew. After this, special Doppler radars were installed at karge commercial airports to improve safety.

Fujita's study of storms also produced the term bow echo, which describes the bow-like shape of a storm. Parts of it may produce very high horizontal winds, and other parts might generate downbursts. These data helped weather forecasters to warn the public.

A real bow echo in Kansas City, 2008, and a diagram showing how such things form.

Note how the diagram shows the storm front moving left to right, but the ends are curling around. (Wikipedia)

Despite retiring in 1990, Fujita continued investigating things like hurricanes and El Niño. He became a naturalized American citizen in 1968, and he got the nickname Mr. Tornado from a National Geographic article in 1972. He was the recipient of many awards in his lifetime ever since his middle school science award. Here is a partial list: