NOAA’s data stewardship program is a new paradigm in data management consisting of an integrated suite of functions to preserve and exploit the full scientific value of NOAA’s environmental data. These functions are: the proper archival of and timely access to data and metadata; the careful monitoring of observing system performance for long-term applications; the generation of authoritative long-term records from multiple observing platforms; and the assessment of the state of the atmospheric, oceanic, land, cryospheric and space environments. Successful implementation of NOAA’s data stewardship program will ensure NOAA’s environmental data are of maximum use to the Nation now and in the future.

To promote full exploitation of the scientific value of NOAA’s data by current and future users, four functions, each with several constituent components, must be achieved. The first function is information technology oriented and is called the Comprehensive Large-array data Stewardship System (CLASS). The CLASS function, insuring complete archival and access capabilities, requires that metadata, direct observations, and fundamental records from satellite and in situ platforms be comprehensive, complete and preserved, in perpetuity. Open, efficient access to the metadata, products, and data streams must be insured, and data made available in useful formats. These functions and their components must at all times maintain strong emphasis on the physics of the instrument designed to measure an environmental parameter or parameters.

The next three functions are science-oriented and are therefore referred to as Scientific Data Stewardship (SDS). The first SDS function is to provide real-time monitoring of the observing system performance for long-term applications. Such monitoring requires the establishment of tracking tools necessary for the detection of changes in the observing system as well as in the observation record. One example is the detection of small biases in the instrumental record. These biases can then be minimized or eliminated through efficient coordination with network operators.

The second SDS function is generating authoritative, long-term records. This function will preserve and enhance the value of the irreplaceable historical data by conducting rigorous data analysis and research to validate and improve these authoritative records, and by reprocessing and enabling others to reprocess the entire data stream from the fundamental measurements when scientific advances warrant it. SDS will use physics-based techniques to fuse together data from disparate observing systems like direct measurements from ground-based networks as well as indirect measurements from remote sensing instruments on satellites. Note that these elements differ from those that describe operational data, namely timeliness and a minimal reliance on data from outside sources. The authoritative nature and vitality of these products will be maintained through periodic reviews.

The third SDS function uses the authoritative records to assess the current state of the environment and to put it in historical perspective. Long-term trends on local, regional or global scales can be determined and estimated for the future. In addition the authoritative records can be used to detect changes in environmental conditions between different time periods and different environmental regimes.