[Deathwatch] Edward N. Lorenz, scientist, 90

[Deathwatch Central <cdw@slick.org>]

Edward N. Lorenz, 90; scientist developed influential chaos theory

The MIT meteorologist's theory -- that very small changes in a system
can have very large and unexpected consequences -- is applied to
scientific and nonscientific topics.

By Thomas H. Maugh II

April 18, 2008

Edward N. Lorenz, the MIT meteorologist whose efforts to use computers
to increase the precision of weather forecasts inadvertently led to the
discovery of chaos theory and demonstrated that precise long-range
forecasts are impossible, died of cancer Wednesday at his home in
Cambridge, Mass. He was 90.

Lorenz was perhaps best known for the title of a 1972 paper,
"Predictability: Does the Flap of a Butterfly's Wings in Brazil Set Off
a Tornado in Texas?" The memorable title pithily summarized the essence
of chaos theory -- that very small changes in a system can have very
large and unexpected consequences.

Although the chaos theory was initially applied to weather forecasting,
it subsequently found its way into a wide variety of scientific and
nonscientific applications, including the geometry of snowflakes and
the predictability of which movies will become blockbusters.

His work "profoundly influenced a wide range of basic sciences and
brought about one of the most dramatic changes in mankind's view of
nature since Sir Isaac Newton," wrote the committee that awarded him
the 1991 Kyoto Prize for basic sciences in the field of earth and
planetary sciences.

By showing that there are limits to the predictability of many systems,
Lorenz "put the last nail in the coffin of the Cartesian universe and
fomented what some have called the third scientific revolution of the
20th century, following on the heels of relativity and quantum
physics," said atmospheric scientist Kerry Emanuel of the Massachusetts
Institute of Technology.

Lorenz was also "a perfect gentleman, and through his intelligence,
integrity and humility set a very high standard for his and succeeding
generations," he added.

One dramatic conclusion of his work is that it is impossible to predict
weather more than three weeks ahead of time with any degree of
certainty.

The roots of chaos theory date to at least the late 19th century, when
French physicist Henri Poincare discovered to his chagrin that it was
not possible to calculate the stability of a celestial system
containing more than two bodies -- at least using techniques available
at the time.

That was a shock because Newton's laws of gravity and motion promise
order and predictability, and Poincare concluded that there must be
other equations that would eliminate the problem. In the absence of
computers, however, there was little anyone could do to test that
thesis.

In 1961, a young assistant professor of meteorology at MIT was using a
primitive Royal McBee LPG-30 computer to study simple models of the
atmosphere based on a series of 12 differential equations.

After one run, he decided he wanted to study the end of the calculation
in greater depth.

Rather than running the entire calculation again, he chose a point in
the middle of the calculation and entered the previously calculated
values for that point.

He then went off for a coffee break to avoid the incessant noise from
the machine.

When he returned, to his surprise he found that the calculated weather
patterns were grossly different from those produced in the first
computation. After determining that the Royal McBee hadn't simply blown
a vacuum tube, he began looking more closely at the calculation itself.


Ultimately, he concluded that the original calculations had been
carried out to six significant figures.

To save space, however, the printout he used had rounded each value off
to only three places.

That subtle difference, less than 1% of the original values, was
sufficient to drive the system in a completely different direction.

Lorenz worked out the math involved and reported his findings in the
Journal of Atmospheric Sciences in a 1963 paper called "Deterministic
Nonperiodic Flow."

The first paper fell on deaf ears, however, and he received little
attention until his 1972 "butterfly" talk at a meeting of the American
Assn. for the Advancement of Science.

Lorenz later said that he had planned to use a gull as an illustration
but that an MIT colleague suggested a butterfly would have more impact.
He chose Brazil for its alliterative value.

According to the Web of Science online database, Lorenz's original
paper has now received at least 4,000 unique citations by subsequent
authors, making it one of the most prolifically cited papers of all
time.

Edward Norton Lorenz was born May 23, 1917, in West Hartford, Conn. He
earned his bachelor's degree in mathematics from Dartmouth College in
1938 and a master's degree in mathematics from Harvard in 1940.

During the war, he served as a weather forecaster for the U.S. Army Air
Corps, earning a master's degree in meteorology from MIT in 1943.

After the war, he continued his studies and received a doctorate in
1948.

He spent his entire career at MIT.

In addition to the Kyoto Prize, he also received the Royal Swedish
Academy of Sciences' 1983 Crafoord Prize, established to honor fields
not eligible for the Nobel Prize.

Lorenz was an active hiker and cross-country skier and made it a point
to visit mountain trails near every scientific meeting he attended. He
was hiking two weeks before his death, according to his family.

His wife, Jane Logan Lorenz, died in 2001.

Lorenz is survived by daughters Nancy of Roslindale, Mass., and Cheryl
Lorenz of Eugene, Ore.; and son Edward of Grasse, France.

Many thanks to Deathwatch Central for posting this obituary