The space radiation environment presents serious challenges to spacecraft design and operations: adding costs or compromising capability. Our understanding of radiation belt dynamics has changed dramatically as a result of new observations. Relativistic electron fluxes change rapidly, on time scales less than a day, in response to geomag - netic activity. However, the magnitude, and even the sign, of the change appears uncor - related with common geomagnetic indices. Additionally, observations of peaks in radial phase space density are not readily explained by diffusion processes. These observations lead to a complex picture of acceleration and loss process all acting on top of adiabatic changes in the storm-time magnetic field. Of even greater practical concern for national security applications is the threat posed by artificial radiation belts produced by high altitude nuclear explosions (HANE). The HANE-produced environment, like the natural environment, is subject to global transport, acceleration, and losses. Radiation belt remediation programs aim to exploit our knowledge of natural loss processes to artificially enhance the removal of particles from the radiation belts. The need to open up new orbits and new capabilities has raised questions about the space environ - ment that, up to this time, we have been unable to fully answer. Here we describe the development of a next-generation model for specifying natural and HANE-produced radiation belts using data-assimilation based modeling. We exploit the convergence of inexpensive high-performance parallel computing, new physical understanding, and an unprecedented set of satellite measurements to improve national capability to model, predict, and control the space environment.