J. David Neelin and Jochem Marotzke
Science, 1994.
Paper (PDF 1.1 Mb)
© Copyright 1994 by The American Association for the Advancement of Science.
Excerpt.
Numerical models of the Earth's atmosphere and oceans are used to simulate the
behavior of the climate system, to assess likely impacts of greenhouse warming,
and recently, to predict some aspects of climate fluctuations on time scales up to
a year (1). Typically, the ocean is represented by a large number of grid boxes
some hundreds of kilometers wide and tens to hundreds of meters deep. The
equations of motion are marched forward in time for each grid box, in steps whose
length must be decreased if the grid length is decreased. As a result, cutting
grid length in half in every direction increases computational cost by an order
of magnitude. Developing models that are fast enough, even on massively parallel
computers, to run long climate simulations with smaller grid size is thus widely
regarded as one of the grand challenges for scientific computing. But there is an
even greater challenge to increasing the fidelity of climate models: improving the
representation of the aggregate effect of processes that occur at scales smaller
than the model grid size. In this issue, Danabasoglu, McWilliams, and Gent (2)
report a striking example of such efforts helping to solve several endemic problems
in a global ocean model.
Citation. Neelin, J. D. and J. Marotzke, 1994: Representing Ocean Eddies in Climate Models. Science, 264, 1099-1100.