Parts | Days | Selection | Search | Downloads | Help

AKPhil: Arbeitskreis Philosophie der Physik

AKPhil 6: Wissenschaftstheorie I

AKPhil 6.2: Talk

Tuesday, March 4, 2008, 17:15–17:45, KGI-HS 1015

Scale separation as a condition for quantitative modelling — •Rafaela Hillerbrand — Faculty Of Philosophy, University of Oxford, UK

In many applied contexts, it is not qualitative but quantitative forecast that is required. For example climate models shall predict not only the correct shape of the probability distribution of a possible change in mean temperature, but also the numerical values associated to its mean, its variance and so forth. This paper addresses the question on when the description of an empirical phenomenon allows for such a quantitative modelling. The distinction between working and non-working quantitative predictions does not coincide with the boundary between social and natural sciences. Even for the latter numerous examples exist for which quantitative models are still missing. A well known example within physics is hydrodynamic turbulence.

It is argued that the key issue in deriving quantitative forecasts is the separation of the relevant scales. It is shown that this holds both for purely phenomenological models as well as for models derived within a deeper theoretical framework. As an example for a working quantitative model the semi-classical description of the Laser is opposed to hydrodynamic turbulence, which up to now does not allow for quantitative modelling due to a lack of scale separation. Similarly scale separation distinguishes models within QED from QCD models.