Geoscience Reference
In-Depth Information
as forced large-scale descent or ascent over the AMregion by ENSO. The large-
scale vertical motions provide a strong control on atmospheric stability and
initiation of convection. Even though the ENSO remote forcing has relatively
slow time scales, its impact may be sufficient to tip the delicate balance of the
aforementioned local feedback processes causing either the amplification of a
given climate state or transition fromone state to the other. The hypothesis needs
to be verified with additional data, analyses, and further experimentations with
other climate models with detailed land surface processes.
In summary, studies up to now have shown that a large portion of the
predictable part of interannual variability of the monsoon rainfall is forced by
the slowly varying boundary conditions at the Earth's surface. However, no
climate model has been able to replicate even the simplest empirical relation-
ship between the SST anomalies and monsoon rainfall anomalies. It is unclear
at this stage whether the inability of current models to simulate and predict
monsoon rainfall is due to model deficiencies or due to intrinsic lack of
predictability of the monsoon. It is likely that both play important roles.
9.4 Future challenges
In this chapter, we have discussed the importance of modeling in providing
better understanding of causes of regional climate anomalies, and in predict-
ing future climate evolution, using the AM climate as a specific example.
Given that models will be increasingly used for climate predictions on all
time scales, it is important to keep in mind that large uncertainties exist and
that not all aspects of model predictions have the same degree of reliability.
The challenge is how to reduce these model uncertainties, and to make
climate forecasting more reliable and useful. The following are suggested
steps that should be taken to move in that direction. Because each of the steps
involves complex procedures and organized efforts, successfully implement-
ing these steps will take years of sustained efforts by the science community.
Use finer spatial and temporal resolution. As stated previously, one of the major
uncertainties in climate models stems from the lack of spatial and temporal resolu-
tion, and as a result regional to subregional scale features are not well represented.
Yet it is these subregional features and short-term events that cause the most
socioeconomic damage. With the advances in computational power, it is now
possible to run climate models with high resolution globally for extended periods.
For example the European Centre for Medium-Range Weather Forecasts is running
its operational model for medium and long-range weather predictions of the order
of 50 km resolution, and the Frontier Climate Change Research Program of Japan is
running the Earth Simulator at 10-20 km resolution. One of the most obvious
improvements in going to higher resolution is the better simulation of orographic
Search WWH ::
Custom Search