CLIMATE DATA RECORD GENERATING ELEMENTS
Among the key elements of successful CDR generation programs identified in the committee’s first report, seven were related specifically to CDR generation (see Box 1). These elements address the need for high accuracy and stability of FCDRs, met in part by thoroughly characterizing and calibrating sensors before launch and during their lifetime. Likewise, TCDRs should have well-defined levels of uncertainty and an ongoing program for validation. Another critical element of CDR generation is the development of well-defined criteria for selecting which TCDRs to produce, accompanied by processes and policies whereby scientists, decision makers, and other stakeholders can propose new TCDRs and provide feedback considered in the selection of TCDRs.
It is reassuring that many of these generation elements are included in the implementation plan. The descriptions of the FCDR and TCDR teams and their responsibilities are consistent with CDR generation elements 5, 6, 7, and 11. The committee strongly endorses a mix of centralized and distributed activities for CDR production identified in the draft plan (NOAA, 2005, p. 43).
Likewise, a plan for defining progressive levels of reduced uncertainty in the CDRs (Element 10) was described by the plan’s authors when they briefed the committee, although it could have been explained more fully in the written document. A plan for defining progressive levels of reduced uncertainty should be described explicitly in the written plan as well.
Conversely, the plan lacks sufficient detail on how TCDRs are proposed and how feedback is provided for consideration in the selection of CDRs (element 9). As noted in the committee’s first report (NRC, 2004, p. 56), this process should include input from scientists, decision makers, and other stakeholders. It may take place through the advisory structure discussed above or by external proposals submitted by individuals. One mechanism for soliciting wide community input would be to issue a formal Request for Information (RFI) soliciting proposed TCDRs. Final selection should be made by NOAA in consultation with its advisory structure. The SDS Working Group and NOAA program managers would then prioritize the TCDRs based on well-defined criteria (element 8). In developing these advisory mechanisms, NOAA should build upon the successes of NASA, which has longer experience in CDR generation, particularly in areas beyond NOAA’s core observing strengths. The revised implementation plan should state how FCDRs will be prioritized for their conversion to TCDRs and distribution to the science community. Note that the prioritization of CDRs has to consider existing FCDRs and TCDRs produced by the Distributed Active Archive Centers (DAACs) and used widely by the research community, including the Advanced Very High Resolution Radiometer (AVHRR) Pathfinder series and their Moderate Resolution Imaging Spectroradiometer (MODIS) continuations, as well as the Scanning Multichannel Microwave Radiometer (SMMR), Special Sensor Microwave Imager (SSM/I), and Advanced Microwave Scanning Radiometer for NASA’s Earth Observing System (AMSR-E) sea ice products.
Although it is addressed in sustaining CDR element 12, reprocessing is a vital part of generating and regenerating CDRs. New technology will allow much more rapid reprocessing than has occurred previously. It is important to maintain community interaction before, during, and after reprocessing by providing a rationale for and expected results of the reprocessing. Therefore, the plans to reprocess subsets prior to a full reprocessing are laudable. The reprocessing of FCDRs should be done in coordination with TCDR teams that use those FCDRs as their primary input.
While CLASS will be supporting a variety of data, the SDS will have the additional responsibility of merging data from multiple satellite platforms into NOAA CDRs. The implementation plan needs more detail on how this process will be accomplished, including the involvement of the SDS Working Group and Science Teams.