The Ocean Sciences Decadal Committee of the United States in laying out future efforts suggested “The very few existing time-series stations paint a compelling picture of important oceanic changes in physics, chemistry and biology. Yet these stations capture the time domain at only a single point. New strategies for observing the appropriate spatial correlation are required.” The question confronting the community is how do we get there? Key enabling technologies are opening the door. These include: (1) long-duration moorings or cabled observatories for subsurface time, (2) high-frequency RADARs providing real-time surface current maps over shelf scales, (3) a growing international constellation of high-resolution ocean color satellites, and (4) an emerging class of long-duration remotely-controlled Autonomous Underwater Vehicles (AUVs). The flight patterns of the AUVs will be directed by real-time satellite imagery, providing high resolution maps maps of fronts/slicks, and onboard theoretic decision-making software allowing the AUVs to analyze their own environmental data. Data from the AUVs will be transported back to shore on an hourly basis allowing for vicarious calibration of satellite imagery and adaptive sampling using ships. These real-time data streams are used to feed data assimilative nowcast/forecast models. This observationally rich environment changes the relationship between models and observation. In the well-sampled ocean, forecast errors are dominated by uncertainties in the model formulations or boundary conditions, and ensemble forecasts with differing parameterizations can be compared to observations so as to improve our understanding of errors associated with various model assumptions.