A Review and Analysis of the Addition of TAC and High Resolution BUFR Format Sonde Data and GPS Information to EMC's Observations Processing

The World Meteorological Organization (WMO) has been coordinating the transition of observational data from the traditional alphanumeric codes (TAC) format to the Binary Universal Form for the Representation of meteorological data (BUFR) format. While the distribution and use of TAC formatted data was initially scheduled to end in 2014, there are still many stations across the globe that transmit radiosonde data exclusively in TAC format. As a result, atmospheric modeling centers have had to ensure their codes can handle both formats. Upper air sonde data is one such data type that exists in both data formats. The National Oceanic and Atmospheric Administration (NOAA) Environmental Modeling Center (EMC) is updating pre-processing code to ingest and process Upper air sonde data in the TAC and BUFR formats.

Observation Processing software is now designed to generate a new "uprair" data file (complementary to the existing “adpupa” file) to contain all available radiosonde and dropsonde profile data originating in the BUFR format. The code verifies whether a station profile exists in both formats (TAC and BUFR) and chooses the higher-resolution BUFR profile if it exists. For the generation of so-called prepbufr files, which combine a few observation types by applying a simplified bufr template table, the high density of vertical pressure data is reduced to 255 select levels; “mandatory” pressure levels are retained, and a spaced increment is applied to capture the remaining levels throughout the depth of a profile.

Much of the profile data available in the BUFR format features sonde drift information measured against the Global Positioning System (GPS) constellation. The GPS, the European Union's Galileo, China's BeiDou Navigation Satellite System, and Russia's Global Navigation Satellite System (GLONASS) are all considered part of the Global Navigation Satellite System (GNSS) due to their worldwide coverage, and collaboration between them has brought about more precise location positioning. When GNSS-measured drift data is available in a profile, the Observation Processing code will extract and use that information in place of a wind-derived, constant-ascent rate estimation used in processing TAC profile data.

Assimilation experiments will be conducted to evaluate the quality of the BUFR sonde profile data and the impact of including vertical density and GPS-measured drift to NCEP NWP model performance.