Tuesday, 15 January 2002: 2:30 PM
The Plan to access Real-Time and NWP Operational Model Data Sets using NOMADS
To address the well documented need for climate, atmosphere and ocean information, the National Climate Data Center (NCDC) along with the National Center's for Environmental Prediction (NCEP) have initiated the NOAA Operational Model Archive and Distribution System (NOMADS). NOMADS is a collaboration between the Geophysical Fluid Dynamics Laboratory (GFDL); the George Mason University (GMU); the National Center for Atmospheric Research (NCAR); NCDC; NCEP; the Pacific Marine Environmental Laboratory (PMEL); Unidata, the Center for Ocean-Land-Atmosphere studies (COLA); and others. The objective of NOMADS is to preserve and provide retrospective access to GCM's and reference quality long-term observational, and high volume three dimensional data, NCEP NWP Operational model run history, re-start and re-analysis information, as well as other NCDC available products and data. This paper describes the National Centers for Environmental Prediction role in this highly collaborative effort and specifies the plan for providing a rerun/retrospective model service to the scientific community, public and NWP centers. Metadata file descriptions using the extensible markup language (XML) is agreed upon by NOMADS participants for representing, storing, retrieving and interoperating in a distributed environment. In addition server side utilities are integrated for sub-seting, re-gridding and repackaging datasets for direct file transfer using direct and web based file transfer protocol (ftp). NOMADS utilizes servers such as the Distributed Oceanographic Data System (DODS), the Live Access Server (LAS) and Distributed Metadata Server (DIMES) which develop and display data sets from the distributed data base, their need for granularity of data base searches can become inefficient for large data set transfers. Direct ftp retrieval of original data sets without format change is provided including server side tools and utilities while on the client side users display results according to their own preference.