In the early days of the fusion program the fundamental goals of both the experimental and theory programs were to understand and characterize the transport in fusion experiments. There was little discussion of trying to control the fine scale turbulence driving transport -- transport and turbulence were believed to be instrinsic to magnetically confined plasma systems. The experimental observation that peaking of the density profile in the Alcator experiment could change the confinement properties of the machine and later that the plasma could spontaneously form a transport barrier in the plasma edge (the H-mode transition) profoundly altered this thinking. A major focus of the program is now to control pressure profiles, magnetic shear and plasma rotation profiles to induce the formation of internal transport barriers and thereby strongly reduce the radial transport of energy and momentum. The formation and control of such barriers are now considered critical to the achievement of good performance in future burning plasma experiments such as ITER. For this reason it is essential to develop a full understanding of the physics that controls the onset and development and dynamics of transport barriers. While there has been some progress in understanding the onset conditions for the H-mode transition, we are far from a predictive capability in this area.

The complexity of the problem stems from its multi-disciplinary richness - the physics issues lie at the intersection of the MHD description and the kinetic description of fine scale turbulence and zonal flow generation. It is only recently that electromagnetic gyrokinetic codes have reached the maturity required to describe these processes. Progress on the problem is also made complex by the multi-time-scale nature of the dynamics of barrier formation. Barriers form over times scales of 100's of milliseconds while turbulence time scales are typically 10's of microseconds or shorter if electron scale turbulence plays a dominant role (which is likely in well-developed barriers). Thus, simulations of barrier formation are computationally time consuming and become daunting when the number of control parameters is factored into the mix.

The Center for Multiscale Plasma Dynamics aims to pioneer the next important advance in understanding and modelling turbulence -- the analysis of multiscale turbulent phenomena, including the formation of transport barriers. This would not be possible without the clear evidence that gyrokinetics is ready to be applied more widely. Thus, part of our effort is to further benchmark computational tools, especially in the finite beta regime of present day fusion experiments. Three topical areas have been selected. Research in these areas will strengthen the physics basis of the dynamics of turbulence and our ability to model it and will address head-on the most important issues facing the fusion program in this area:

• Can smoking gun evidence be found for the electron scale turbulence and the predicted large-scale radial streams that are expected to enhance its role in transporting electron thermal energy? What important interactions exist among turbulent fluctuations that are driven at disparate scales? For example, how is ion-scale turbulence affected by the presence of parasitic electron-scale fluctuations, and vice versa?
• Does the gyrokinetic model provide an accurate description of finite-beta electromagnetic disturbances?
• Do basic kinetic Alfvén wave experiments confirm existing theoretical ideas?
• What triggers the formation and determines the structure of transport barriers? In particular, how are large scale sheared flows driven and damped by small scale turbulence, and under what conditions do large scale flows drive or damp instabilities? What limits the pressure gradients that develop in mature transport barriers?