HR: 16:00h
AN: NG44A-03 INVITED [Abstracts]
TI: Modeling Systems Involving Interactions Between Scales
AU: * Murray, A B
EM: abmurray@duke.edu
AF: Division of Earth and Ocean Sciences, Nicholas School of the Environment and Earth Sciences/Center for
Nonlinear and Complex Systems, Duke University
, Box 90230, Durham, NC 27708-0230 United States
AB:
When we think of numerical models, `simulation modeling' often comes to mind: The modeler strives to include as many of the
processes operating in the system of interest, and in as much detail, as is practical. The goal is typically to make accurate quantitative predictions. However, numerical models can also play an explanatory role. The goal of explaining a poorly
understood phenomenon is often best pursued with an `exploratory' model (Murray 2002, 2003), in which a modeler minimizes the processes included and the level of detail, to try to determine what mechanisms-and what aspects of those mechanisms-are
essential. These strategies are closely associated with different approaches to modeling processes across temporal and
spatial scales.
Simulation models often involve `explicit numerical reductionism'-the direct representation of interactions at scales as
small as possible. Parameterizing sub-grid-scale processes is often seen as an unfortunate necessity, to be avoided if
possible. On the other hand, when devising an exploratory model, a top-down strategy is often employed; an effort is made to
represent only the effects that much smaller-scale processes have on the scale of interest. This approach allows
investigation of the interactions between the emergent variables and structures that most directly explain many complex
behaviors. As a caricature, we don't investigate water-wave phenomena by simulating molecular collisions.
In addition, basing a model on processes at much smaller scales than those of the phenomena of interest leads to the concern
that model imperfections may propagate up through the scales; that if the small-scale processes are not treated very
accurately, the key interactions that emerge at larger scales may not occur as they do in the natural system. However, this
risk can be bypassed by basing a model directly on larger-scale interactions, and examining which of these interactions might cause a phenomenon. For this reason, it has been suggested recently that a top-down approach leads to models that are better able to make practically useful predictions. The strategies described here represent end members of modeling continua, and
what blend of approaches leads to predictions most useful to science or society likely varies from case to case.
DE: 3210 Modeling
DE: 3220 Nonlinear dynamics
DE: 4546 Nearshore processes
DE: 4558 Sediment transport
SC: Nonlinear Geophysics [NG]
MN: 2005 Joint Assembly