Other reports in this collection
An overriding challenge to modelling and to the IPCC is prediction. This challenge
is particularly acute when predictive capability is sought for a system that
is chaotic, that has significant non-linearities, and that is inherently stiff
(i.e., widely varying time constants). And within prognostic investigations
of such a complex system, the issue of predicting extreme events presents a
particularly vexing yet important problem. However, there appear to be coherent
modes of behaviour that not only support a sense of optimism in attacking the
prediction problem, but also these modes may offer measurable prediction targets
that can be used as benchmarks for evaluating our understanding of the climate
system. In addition, predictions of these modes represent valuable contributions
in themselves.
Evaluating the prognostic skill of a model and understanding the characteristics
of this skill are clearly important objectives. In the case of weather prediction,
one can estimate the range of predictability by evaluating the change of the
system from groups of initial states that are close to each other. The differences
in these time-evolving states give a measure of the predictive utility of the
model. In addition, one has the near-term reality of the evolving weather as
a constant source of performance metrics. For the climate issue, the question
of predictability is wrapped up with understanding the physics behind the low-frequency
variability of climate and distinguishing the signal of climate change (see
Chapter 9, Section 9.2.2.1). In other words, there
are the paired challenges of capturing (predicting) “natural” variability
of climate as well as the emerging anthropogenically forced climate signal.
This dual challenge is distinctively climatic in nature, and whereas the longer-term
character of climate projections is unavoidable and problematic, the intra-seasonal
to inter-decadal modes of climate variability (e.g., ENSO, Pacific Decadal Oscillation
(PDO), and North Atlantic Oscillation (NAO) – see also Chapter
7, Box 7.2) offer opportunities to test prognostic climate skill. Here,
some predictive skill for the climate system appears to exist on longer time-scales.
One example is the ocean-atmosphere phenomenon of ENSO. This skill has been
advanced and more clearly demonstrated since the SAR, and this progress and
demonstration are important (see Chapter 7, Section 7.6;
Chapter 8, Section 8.7 and Chapter 9,
Section 9.3.5). Such demonstrations and the insights gained in developing
and making prognostic statements on climate modes frame an important area for
further work.
This opportunity is well summarised in Chapter 8 (in particular, Section 8.7), “The atmosphere-ocean coupled system shows various modes of variability that range widely from intra-seasonal to inter-decadal time-scales (see Chapters 2 and 7). Since the SAR, considerable progress has been achieved in characterising the decadal to inter-decadal variability of the ocean-atmosphere system. Successful evaluation of models over a wide range of phenomena increases our confidence.”
Central to the climate system are the coupled dynamics of the atmosphere-ocean-terrestrial system, the physical processes associated with the energy and water cycles and the associated biological and chemical processes controlling the biogeochemical cycles, particularly carbon, nitrogen, phosphorus, sulphur, iron, and silicon. The atmosphere plays a unique role in the climate system since on a zeroth order basis it sets the radiative forcing. Specific sub-systems that are important and yet still poorly understood are clouds and sea ice; the thermohaline ocean circulation is a fundamentally important phenomenon that needs to be known better, and underlying these sub-systems and phenomena are the still ill-understood non-linear processes of advection (large-scale) and convection (small-scale) of dynamical and thermodynamical oceanic and atmospheric quantities. These sub-systems, phenomena, and processes are important and merit increased attention to improve prognostic capabilities generally.
|
Other reports in this collection |
|