Abstract
Purchase PDF (189 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (1)
Purchase PDF (487 K)
doi:10.1016/j.cpc.2004.06.001 
Copyright © 2004 Elsevier B.V. All rights reserved.
Beyond benchmarking—how experiments and simulations can work together in plasma physics
References and further reading may be available for this article. To view references and further reading you must purchase this article.
Martin Greenwald
Available online 23 July 2004.
Abstract
There has been dramatic progress in the scope and power of plasma simulations in recent years; and because codes are generally cheaper to write, to run and to diagnose than experiments, they have a well-recognized potential to extend our understanding of complex phenomena like plasma turbulence. However, simulations are imperfect models for physical reality and can be trusted only so far as they demonstrate agreement, without bias, with experimental results. This “validation” process tests the correctness and completeness of the physical model along with the assumptions and simplifications required for solution. At the same time, it must be understood that experimental measurements are almost always incomplete and subject to significant uncertainties and errors. For optimum scientific progress, simulations and experiments must be seen as complementary not competitive approaches. We need experiments dedicated to answering critical questions raised by the simulations, which examine the validity of models and which explicitly test their assumptions. A premium should be placed on ongoing collaborations which are open and candid about the sources of error and the strengths and weaknesses of each approach. Ultimately both experiments and simulation have much to gain by adopting an approach of co-development, where simulations are continuously and carefully compared to experimental data and where experiments are guided by the results of simulations.
