NCDC to satisfy the guidelines of the International Council of Scientific Unions (ICSU) that all WDC data be available to the international research community; data in national data centers, on the other hand, may be restricted to some users. In addition, the primary datasets in WDCs emphasize global, rather than national, perspectives. Holdings in the WDC-A for Meteorology include foreign publications and data from international experiments and World Meteorological Organization (WMO) programs. An important goal of the WDC-A is to fill gaps in historical baseline datasets.
The WDC-A for Meteorology has an active, and growing, visitor program, which is an ICSU measure of success. Over the last few years, the WDC has hosted 25 visitors from 11 countries, trained 3 scientists from Saudi Arabia and Korea, and pursued cooperative programs with 4 Russian and Chinese scientists.
The access to global datasets by the international community, coordination with the appropriate international organizations for exchanging data and developing standards, and the active foreign visitor program are strengths of the WDC. The center has been weak, however, in publicizing and digitizing their catalogs of data holdings, although there have been significant recent improvements in this regard. Nevertheless, the panel believes that the WDC-A for Meteorology fully meets the ICSU guidelines for world data centers.
Impact of National Weather Service Modernization
The end repository for most weather and climate data generated by NOAA's National Weather Service is the NCDC. As automatic electronic sensors slowly supplant the human observer, the rate of data collection continues to increase in a nearly exponential fashion. The annual input from just one of these new systems, the NEXRAD or Weather Service Radar-1988 design Doppler (WSR-88D) radars, rivals in volume the entire existing historical holdings from the traditional data sources. Apprehensive about discarding detailed data that may later prove valuable, and also mindful of the data retention recommendations of various advisory groups, the center struggles to cope with this flood of new information. The
NCDC will have to exploit the latest advances in storage technology simply to keep pace with the rate of input.
Several factors complicate this process:
-
In addition to merely saving the data, the center must be able to retrieve, disseminate, and, in some cases, summarize or interpret the information for the customer.
-
Screening and quality control must take place for most datasets.
-
As custodian of an important part of the nation's scientific legacy, the NCDC follows guidelines set down by the National Archives and Records Administration (NARA). These include redundant and off-site storage for safety backup, and use of a stable medium. For datasets such as NEXRAD, 8 mm helical tapes offer the only practical alternative, even though the long-term properties of this storage medium are not well known.
-
The problems faced in archiving, retrieving, and disseminating the data from an observing system such as NEXRAD receive minimal consideration and resources from the NOAA operating element fielding the system during its design phase.
Another example of the latter difficulty is the Automated Surface Observing System (ASOS). This system is capable of recording and storing in its on-site memory the values of a variety of elements every minute. It is desirable that these values be archived, and the NCDC is to be commended for its active role in attempting to preserve the scientific value of these data. Unfortunately, the data are overwritten operationally every 12 hours. Hourly data and occasional special information are transmitted over the national distribution systems and can be captured automatically in many locations for long-term archival. The only way to retrieve the high temporal (one-minute) resolution data, however, is to dial each site directly by telephone before the data disappear. The initial ASOS deployment will involve nearly 1,000 sites and may approach 2,000. Thus, a substantial investment must be made at the NCDC to acquire a bank of modems and to pay for several thousand phone calls per day (Snodgrass, 1994). There are no plans in the ASOS program to reimburse the NCDC for the incremental cost of obtaining these data, only a small