Catálogo de publicaciones - libros

The Madden-Julian Oscillation (MJO) has long been an aspect of the global climate that has provided a challenging test for the climate modeling community. Since the 1980s there have been numerous studies of the simulation of the MJO in atmospheric general circulation models (GCMs), ranging from Hayashi and Golder (1986, 1988) and Lau and Lau (1986), through to more recent studies such as Wang and Schlesinger (1999) and Wu et al. (2002). Of course, attempts to reproduce the MJO in climate models have proceeded in parallel with developments in our under- standing of what the MJO is and what drives it. In fact, many advances in under- standing the MJO have come through modeling studies. In particular, failure of climate models to simulate various aspects of the MJO has prompted investigations into the mechanisms that are important to its initiation and maintenance, leading to improvements both in our understanding of, and ability to simulate, the MJO.

In April of 2002, a workshop was held that brought together participants with a wide range of geophysical expertise to focus on the problem of sub-seasonal predictability (). This workshop marked a relatively important milestone in the development of our predictive capability of the atmosphere, ocean, and land systems. The fact that it lured scientists with expertise in modeling, theory, and observations, as well as operational forecasters and funding agency administrators indicated that we had reached the point where sub-seasonal variability presented itself as more than a theoretical concern or vaguely observed set of phenomena. In fact, the need for such a workshop was based on the recognition that a number of sub-seasonal features could likely provide near-term opportunities for improving long-lead forecast skill. One of the keynote speakers, H. van den Dool, brought to the participants’ attention the early foresight that John von Neumann (1955) had of the expected progress to be made in the area of “long-range” forecasting. In terms of present-day terminology, von Neumann recognized (see Appendix I for excerpt) that the first gains to be made in the area of [atmospheric] prediction were likely to be made at the short range where the initial conditions are expected to play an important role (i.e., 1950s–1970s). Following progress in this area, substantial gains would next be likely made at the very long range, meaning climate prediction, where surface boundary conditions (e.g., large-scale sea surface temperature (SST)) are expected to play the most important role (i.e., 1980s–1990s). Then, only after considerable understanding was obtained in each of these two extreme regimes could progress be made at the sub-seasonal timescale (e.g., 2 weeks to 2 months) where both the initial conditions and boundary conditions are expected to be important. The occurrence of this workshop and its follow-on activities () indicate that by virtue of our progress with both “weather” and “climate” prediction problems, we had reached a point where it was feasible to consider the intermediary problem of the sub-seasonal timescale.