30
Kirk Bryan
Figure 3.1. At the dawn of the computer era, Joseph “Joe Smag” Smagorinsky (1924–2005) had the
courage, imagination, and entrepreneurial skills to begin building on Richardson’s original vision of a
quantitative model of a climate system at the U.S. Weather Bureau. Photo taken in 1983.
atmosphere, ocean, land surface, and cryosphere. Richardson recognized that many
gaps existed in the basic understanding of the climate system essential for such an
endeavor. As he put it , “We are forced to occupy a house which is still in the hands
of the builders.” Modelers are still in this uncomfortable position today. However,
climate modelers and field workers studying basic processes are now recognizing the
benefits of close collaboration to fill in the gaps. When Richardson wrote his book,
Weather Prediction by Numerical Process, there were no computers powerful enough
to carry out the calculations he envisioned. Even if they had been available, there was
a lack of knowledge of the branch of mathematics required to use computers to model
the circulation of the ocean and atmosphere. Research during World War II gave the
impetus to the development of computers and pioneering work began on numerical
weather forecasting led by John von Neumann and Jules Charney at the Institute
for Advanced Study (IAS). Joseph Smagorinsky (Figure 3.1), who had been a staff
member at the IAS, returned to the U.S. Weather Bureau with the ambitious goal
of reviving Richardson’s vision of a mathematical model of the entire climate system, using what had been learned in numerical weather forecasting as basic building
blocks.
Ocean circulation models are critically dependent on computer technology.
Computer resources available to ocean modelers are still not adequate to fully resolve
the wide spectrum of time and space scales of the ocean circulation. It is not possible
to say after more than four decades since work began that ocean models are mature.
For reasons that will be made clear there is still much work to do. This brief account
will only be concerned with the three decades between 1960 and 1990. I have associated these three decades with the birth, infancy, and adolescence of ocean general
Kirk Bryan
Figure 3.1. At the dawn of the computer era, Joseph “Joe Smag” Smagorinsky (1924–2005) had the
courage, imagination, and entrepreneurial skills to begin building on Richardson’s original vision of a
quantitative model of a climate system at the U.S. Weather Bureau. Photo taken in 1983.
atmosphere, ocean, land surface, and cryosphere. Richardson recognized that many
gaps existed in the basic understanding of the climate system essential for such an
endeavor. As he put it , “We are forced to occupy a house which is still in the hands
of the builders.” Modelers are still in this uncomfortable position today. However,
climate modelers and field workers studying basic processes are now recognizing the
benefits of close collaboration to fill in the gaps. When Richardson wrote his book,
Weather Prediction by Numerical Process, there were no computers powerful enough
to carry out the calculations he envisioned. Even if they had been available, there was
a lack of knowledge of the branch of mathematics required to use computers to model
the circulation of the ocean and atmosphere. Research during World War II gave the
impetus to the development of computers and pioneering work began on numerical
weather forecasting led by John von Neumann and Jules Charney at the Institute
for Advanced Study (IAS). Joseph Smagorinsky (Figure 3.1), who had been a staff
member at the IAS, returned to the U.S. Weather Bureau with the ambitious goal
of reviving Richardson’s vision of a mathematical model of the entire climate system, using what had been learned in numerical weather forecasting as basic building
blocks.
Ocean circulation models are critically dependent on computer technology.
Computer resources available to ocean modelers are still not adequate to fully resolve
the wide spectrum of time and space scales of the ocean circulation. It is not possible
to say after more than four decades since work began that ocean models are mature.
For reasons that will be made clear there is still much work to do. This brief account
will only be concerned with the three decades between 1960 and 1990. I have associated these three decades with the birth, infancy, and adolescence of ocean general