Magnetic reconnection enables a magnetized plasma system to convert magnetic energy into high speed flows, thermal energy and energetic particles. It also connects field lines joining previously isolated regions of plasma, allowing rapid transport of heat and particles. The broad importance of reconnection in nearly all plasma systems, ranging from laboratory fusion experiments to the Earth's magnetosphere, the solar corona and the astrophysical environment, render it one of the premier scientific topics in plasma physics. The importance of this area is also linked to its intrinsic scientific interest: the release of magnetic energy on a global scale is linked to the topological change in the magnetic field resulting from kinetic processes occurring at very small scales. Accordingly, the understanding of magnetic reconnection demands the development of innovative analysis methods which address physics coupling across vastly disparate spatial and temporal scales. The absence of techniques that can address such scale differences along with the requisite physics -- kinetic models at short scale and MHD models at larger scales -- has significantly complicated efforts to explore reconnection at the first principles level, and as a consequence has led to a failure to understand some of the most fundamental observations in both fusion experiments and space and astrophysical systems.

In the case of fusion experiments, magnetic reconnection phenomena play a critical role in virtually every configuration which is being explored to confine high temperature plasma. Indeed, three reconnection phenomena can be problematic for burning and non-burning tokamak plasmas: ``sawteeth'' can degrade confinement and expel alpha particles from the plasma core; tearing modes (neoclassical or otherwise) grow slowly and can lead to a loss of confinement and a termination of the discharge; and disruptions can terminate the discharge and cause damage to the plasma facing components. The physics of even some of the most basic processes involved in these reconnection phenomena is still elusive, e.g., the behavior of the magnetic field profile (q profile) across the central plasma core after the sawtooth crash. The inability to model the reconnection of large scale flux and the evolution of magnetic islands, while at the same time including an accurate kinetic description of the currents and transport, is inhibiting the development of a predictive capability to model the performance of future fusion experiments such as ITER.

A broadly important issue, which spans fusion and astrophysical systems, concerns the mechanism for intense plasma heating and energetic particle production during magnetic reconnection. The conversion of magnetic energy to energetic electrons and ions is widely observed during reconnection in laboratory fusion experiments, dedicated reconnection experiments, the Earth's magnetosphere and the solar corona. Ions can be energized in the ion-scale boundary layers and shocks that develop around the magnetic X-line and separatrices. The elucidation of the mechanism for the production of energetic electrons is a greater challenge, especially in the context of large scale systems where the magnetic X-line occupies such a small region of available physical space. Can electron acceleration occur at sites far removed from the X-line -- e.g., around the slow shocks that bound the outflow region from the X-line?

In spite of the strong scientific evidence that non-MHD processes are essential to model magnetic reconnection in laboratory and astrophysical systems, the MHD model remains the model of choice simply because the tools are not available to implement any other model while at the same time resolving the huge range of spatial scales involved. The scale separation problem is generic to modeling reconnection in all plasma systems from the galactic to the laboratory scale. In some sense the scale separation is a strength since it permits some key simplifications which we will exploit. Examples are:

• The region where kinetic physics is critically important is confined to a narrow reconnection layer typically a few ion larmor radii wide -- the sawtooth problem being a notable exception
• The important kinetic scale turbulence evolves on a fast time-scale compared to the the island growth time.

The description of multiscale systems is also now a very important topic in applied mathematics and novel techniques such as projective integration have been developed to model large scale systems in which kinetic dynamics are of critical importance. Whether such approaches can be applied to large scale systems undergoing reconnection remains to be demonstrated. The techniques function best when the systems of interest exhibit a distinct separation of spatial/temporal scales. This is the case for the reconnection issues that are central to the CMPD mission. Thus, joining forces with experts in this emerging area of applied mathematics to address this important multiscale physics problem appears to be a very promising path with the potential to revolutionize the treatment of critical problems in reconnection both in fusion science and the broader space and astrophysics context.