An Introduction to Atmospheric and Oceanographic Datasets

11. DATA AVAILABLE AT NCAR AND METHODS OF ACCESS

The types of data available from NCAR cover a very wide range of subjects, including most of the data types discussed in previous sections of this document. The data archive is organized in a decimal system similar to the library's Dewey system (see Appendix E).

The datasets are grouped such that similar datasets are numerically close to one another. Appendix F has a complete list of datasets available from the DSS as of the date of this publication.

Researchers can obtain an up-to-date list of all of the DSS datasets and other information (including pricing) via anonymous FTP to ftp ncardata.ucar.edu , or by using a WWW browser and opening the URL

http://www.scd.ucar.edu/dss

Each dataset has one or more volume serial numbers (VSNs) associated with it. A VSN is a file on the MSS. It usually consists of a single file, but sometimes contains many files, separated by logical file marks. Sometimes there is only one VSN for the entire dataset; more often, there are many VSNs, for example one VSN for each month of a large, multi-year dataset. The VSNs are of the form KnnnnK or Ynnnnn, where each n represents a digit. For example, the character version of the dataset ds754.0 is in the VSN named K0240K and the MSS path would be /DSS/K0240K.

Information on a particular dataset, say ds754.0 is contained within the subdirectory `datasets/ds754.0' If you are using ftp, you can "cd" to that directory; from a WWW browser, you can select the menu of datasets from the DSS home page and simply select the dataset of interest. Within each subdirectory there is a file called MASTER which provides a brief outline of the dataset contents and lists the VSNs in that dataset. In addition, FORTRAN routines and, occasionally, inventories for each VSN are available.

The DSS staff provides assistance and expertise to help researchers locate data appropriate to their needs. Questions may be sent via email to datahelp@ncar.ucar.edu . [see Appendix A]

Climate Analysis Section

The CAS archive includes a variety of datasets (generally) archived as T42 Gaussian grids. (In the near future, CAS will also archive NMC Reanalysis data (Chapter 6) with T63 resolution and will produce datasets in the netCDF format (Chapter 2). These have been reformatted from equally spaced grids at various resolutions to a common grid to facilitate comparisons. (The DSS is the original source of the equally spaced grids.) The datasets include climatologies, monthly means, gridded satellite data and analyzed data. Of particular interest are sets of ECMWF and NMC daily global analyses which are available at various spectral resolutions determined two or four times per day. The datasets have been subject to considerable quality control and have been used as the basis for calculating monthly averages of the basic state variables and deriving value-added diagnostic quantities on a T42 Gaussian grid. The size of the archive with monthly quantities is much smaller and more manageable than the original daily archive. They are archived in the format used by the NCAR Community Climate Model. However, software exists that can be used to interpolate the data to arbitrary grid arrangements. Information about these datasets has been cataloged and can be obtained by using e-mail, gopher or the Web. Specifically:

1. For those with access to a WWW client, the CAS home page is at http://www.cgd.ucar.edu/cas/catalog The home page has a link to the CGD gopher mentioned above.

2. The catalog may be accessed via gopher, either directly or through the World Wide Web (see [i]). The CGD gopher is located at isis.cgd.ucar.edu , from that menu one can directly move to the CAS Catalog.

3. For email access, the user should send mail to casmail@isis.cgd.ucar.edu with a subject of `help'. This will result in a help file being sent to the user.

Climate Modeling Section

Research in the Climate Modeling Section (CMS) encompasses a variety of modeling studies of the physical mechanisms governing the global climate system. In particular, the CMS developed the Community Climate Model (CCM2 FORTRAN code for CCM2 may be obtained via anonymous ftp ( ftp ftp.ucar.edu ; cd ccm ; binary ; get ccm2.tar ; get other files also). ) and has made a number of control integrations available for general use (Table 11.5). These include simulations with different spectral resolutions than the one at which the model was developed (T42); simulations with observed sea-surface temperatures; and, a simulation with the optional Biosphere-Atmosphere Transfer Scheme (often referred to as "BATS"). Details may be found under the NCAR gopher or in the NCAR Technical Note by Williamson (1993). These datasets can be used by researchers with access to NCAR's Mass Store System (MSS).

Datasets produced by atmospheric and oceanographic models can be quite large. The dataset size is a function of horizontal and vertical resolution and how many and how often variables are saved. A typical ten year integration of the CCM2 at T42 horizontal resolution with daily averaged data archived consists of about 55GB. Figure 11.1 illustrates a typical data reduction path one might follow to produce results.

Data Support Section

Climate Analysis Section

Climate Modeling Section: