Environmental and climate changes can be disruptive, but possible adverse effects can be mitigated and even turned into positive outcomes if there is an adequate information base to understand and predict the changes, thereby allowing adequate planning to occur. This requires not only observations, of which there are increasing volumes especially from space-based platforms, but also the ability to analyze and make sense of the disparate observations by synthesizing them into products of various sorts, recognizing user needs, and making the information accessible and available. This requires the use of sophisticated models and assimilation systems, and their exploitation in prediction on multiple time scales. Interactive data stewardship and management should be carried out to facilitate reprocessing and reanalysis of past data to improve climate records. Research is needed on optimization of the observations, and how to best synthesize the data, build the required models, and bridge the gaps between current capabilities and user needs. Although the need for such a system is increasingly recognized, and a framework for international collaboration to build it is in place through the Global Earth System of Systems (GEOSS), major challenges exist to provide adequate resources and ensure that the observations meet the needs. Recent changes in the NPOESS program and cuts in climate observations provide an example of a particular challenge, as continuity of records is threatened. It seems highly likely that such a system would pay for itself many times over in terms of benefits through improved decision making, as we meet the challenges before us. However, as the benefits may not accrue to those who pay for the system, institutional challenges also exist to ensure that those who benefit most contribute to the maintenance of the system.