Beginning in 2006-9, the National Polar-orbiting Operational Environmental Satellite System (NPOESS) Integrated Program Office will launch a new generation of converged multi-spectral and hyper-spectral imagers and sounders into three orbit planes. In 2012, NOAA/NESDIS will launch an equivalent next-generation Geosynchronous Operational Environmental Satellite (GOES-R) imager/sounder capability. This paper examines the upcoming decade and identifies the key requirements to be satisfied and associated approaches to achieve “pervasive” integrated satellite observations and observing systems, along with data and information from other space- and land-based platforms. The numerical weather prediction and climate modeling community continues to drive polar and geosynchronous environmental satellite requirements. At the same time, the physical ocean and land characterization and modeling disciplines are rapidly maturing. The goal is maximizing the utilization of environmental data—both the calibrated radiances and the derived environmental data products. The vision is that of building an optimized, user-friendly 4D data assimilation resource.

This paper reviews the most relevant lessons learned from NASA’s Earth Observing System (EOS) and NOAA’s Polar-orbiting Operational Environmental Satellite (POES) and GOES systems. We consider approaches for best satisfying the user communities through: (1) deploying an architecture that is not static, but continually evolving, blurring the lines between development and sustaining engineering; (2) deploying and maintaining consistent spectroradiometric calibration scales and standards across sensors and platforms; (3) exploiting the diversity of multiple sensing techniques and radiative transfer models while maximizing their synergies; (4) leveraging today’s investments, for example in the NPOESS ground and command and control systems; (5) obtaining maximum benefit from today’s prototype systems; (6) data storage with transparent methods of formatting, categorizing, sub-setting, and distributing the data (both near-real time and reprocessing sets); (7) a research-to-operations transition and deployment concept of operations; and, most of all (8) users’ feedback.