SSSF has been using a set of netCDF header conventions for its profiler data files, which allow programs to be written to use and expect those conventions. For DBS, there are separate conventions for moment, spectral, and wind data. For MAPR, there are also conventions for time series and correlation data. There are examples of each of these in the CDL files listed below, including examples of the same type of data processed using alternate algorithms. The CDL description should be identical among similar data types, but the examples are still useful for the differences in notation and global attributes.
I'm not suggesting these are optimum arrangements for profiler data. These are examples of past and current practice in SSSF. Applications and netCDF files which follow these conventions will be more interchangeable with SSSF software. If anyone wants to make any changes, let's at least cooperate on the changes to try to maintain compatibility, but our existing software could be slow in tracking changes. We have C code for an API for these conventions which might be useful, if only as a base for a rewrite. Likewise we'd be interested in any other API implementations.
The POP program currently used for DBS in the field records moments and spectra in the POP binary file format. We convert the moments and spectra in the binary files into two separate sets of netCDF files.
POP3 Moments CDL
POP3 Spectra CDL
Bill Brown wrote some IDL scripts which convert NIMA ASCII output to netcdf. These are example CDL definitions for those netcdf files.
NIMA Monitor Moments CDL
NIMA Monitor Winds CDL
NIMA Monitor RASS CDL
POP also produces consensus winds and RASS which a Zebra ingest program translates to netCDF in the field in real-time. The standard field attributes are blank because of laziness on the part of the ingestor; ordinarily, they should be specified by the program writing the files. The same consensus winds CDL is used for both the high- (prof915h) and low-mode (prof915l) platforms.
POP3 Consensus Winds CDL
POP3 RASS Consensus CDL
Moments and winds calculated from raw spectra using ETL algorithms.
ETL Moments CDL
ETL Winds CDL
Moments calculated with Weber-Wuertz QC algorithms.
WWQC Moments CDL
Winds generated from the WWQC moments.
WWQC Winds CDL
MAPR Correlation Data
MAPR Time Series Data
RAP NIMA input module CDL
Gary Granger
$Id: index.html,v 1.7 2002/11/21 22:04:03 granger Exp $