Recent polar satellite observations show disintegration of ice shelves in West Antarctica and speed-up of several glaciers in southern Greenland. The great ice sheets in Antarctica and Greenland interact with the global climate in a complex manner, and the impact on global sea level of their retreat would be profound. Most of the existing ice-sheet models, including those used by the Intergovernmental Panel on Climate Change (IPCC), cannot explain the rapid changes being observed. The Center for Remote Sensing of Ice Sheets (CReSIS) is developing new technologies to perform 3-D characterization of ice sheets to understand the physics of rapid changes, and develop models to explain observed changes and predict future behavior. In particular, CReSIS has demonstrated that Synthetic Aperture Radar (SAR) can image the beds of ice-sheets. This will enable a new-generation of high resolution ice-sheet models with realistic boundary conditions, but it will require distributed Cyberinfrastructure to gather and process data and assimilate them with large simulations. We propose a sophisticated Cyberinfrastructure instrument that will both enable the crucial ice-sheet science and educate and train a diverse workforce in both Polar science and Cyberinfrastructure. Instrument: This proposal addresses these compelling scientific challenges with a PolarGrid Cyberinfrastructure, aimed at research, education and outreach. PolarGrid consists of intermittently disconnected field and base grids feeding information to “lower 48” data and computing resources. True real-time processing at the field camp is backed up with increasing fidelity but increasing delay at the base and “lower 48” systems. The requested system includes an expedition grid consisting of ruggedized laptops in a field grid linked to a low power multi-core base camp cluster; a prototype and two production expedition grids feed into a 17 Teraflops "lower 48" system at Indiana University and Elizabeth City State (ECSU) split between research, education and training. This will give ECSU a topranked 5 Teraflop MSI High performance computing system, building on its distance education and undergraduate laboratory infrastructure to create tremendous outreach capabilities. We request upgrades to student laboratories, video-conferencing and key scientific software to support the ambitious education and training goals. PolarGrid will be integrated with TeraGrid for both resource utilization and curricula sharing. We will follow modern open data access standards so that raw, processed and simulated data can be archived outside PolarGrid by and for the full science community.