GED User's Guide
Dataset Selection

ORGANIZATION OF THE DATABASE

  • Directory structure
  • File name convention
    •

    Organization of the Database

    Directory Structure

    The database uses ISO9660 compatible directories (folders) and files, as shown below:

    Volume: GED_II 

    |- [datasets] 
        | -- [ba] {dataset directories} 
                | --- [formats] {.fmt format description files}
                | --- [lists]   {histograms}
                | --- [source]  {source versions of data}
                | --- [reprints]        {published papers}
                | ---   ba*.*           {dataset files>
        | -- etc.       {ba, baz, bec, etc.}
|- [html]               {User's Guide in HTML format}
|- [software]           {Installable software from NGDC}
        | -- [geovu]    {image browse and export}
        | -- [newform]  {data conversion}
|- go.htm               {main menu screen in HTML format}
|- help.txt             {startup instructions in ascii format}

    Organization of the Database

    FILE NAME CONVENTION

    The data file naming convention has been standardized in the GED to aid in searching the database or copying files using wildcard file specifications. Each data-set is named with a two or three letter prefix which is common to all files in the dataset and is also the name of that dataset directory. By specifying these letters followed by wildcard file specifications, all files belonging to the dataset can be selected.

    For time-series data files, the next two numerical characters after the descriptive name refer to the year and the next two numerical characters refer to the month. An example is shown below:

    TK8706 [.IMG] [.DOC]
    DOC => ascii header file
    IMG => data file
    06 = month (e.g., June)
    87 = year (e.g., 1987)
    TK = data-set name (e.g., "Tateishi and Kojiwara")

    Data-sets that provide only average (typical or average) monthly data will show the month designation only, in which case the file name may extend to six characters. For data-sets that do not have a time dependency relationship, the data-set name may occupy up to 7-characters. The 7-character limit allows one-character user prefixes or suffixes for files that are exported (for example, one could add an "X" to the names of exported files). A rigid convention for all data-sets is not possible because of the restrictions of a 7-character file name.

    All data filename extensions conform to the IDRISI name convention (defined above). These file extensions are intended to be somewhat intuitive. For example, all raster (cell-based) data files have an ".IMG" file extension. All vector (point, line, or polygon) data have a ".VEC" file extension. All attribute values files (used to reclassify either raster or vector data layers) have a ".VAL" file extension. All attribute files are provided in two formats, Microsoft Access (".MDB" extension) and ASCII (".TXT" extension). Color palette files, for specific datasets, are also provided in two formats: Idrisi compatible ".SMP" files and ".PAL" files. Geographic Reference files have a ".GRF" extension.