NOAA KLM User's Guide

Appendix G.6

Blue line drawn across page to separate text from document title.
Introduction Page, NOAA KLM TOC, Acronyms
Previous Section, Next Section

APPENDIX G.6 Changes made to Sounding Products (ATOVS and AMSU-B)

April 27, 1999
ATOVS operational.

May 4, 1999
Update coefficients.

May 11, 1999
Update coefficients.

May 12, 1999

  1. Added AMSU-A channel 6 to the CDB high terrain retrieval channel combination. This will only affect high terrain areas like Antarctica and the Himalayas.
  2. RETDVR was modified to retrieve coast and ice as land.
  3. SPGDVR software was modified to do limb correction only once.

May 17, 1999
Update coefficients.

May 24, 1999
Update coefficients.

June 2, 1999
Update coefficients.

June 8, 1999
Update coefficients.

June 9, 1999
The ATOVS GTS (SATEM) formatter was modified to check for zero thickness values and filter retrievals containing those values from distribution.

June 24, 1999
Implemented new OPCUPD to:

  1. Correct problem with summing of total ozone values when values are missing (substitute the values for the US Standard Ozone for missing and then include in sum).
  2. Replace global S matrix with "S" bin 5 (everything south of 60S) where the S matrix is only based on that bin's samples. Rest of globe still use the global S matrix.
New Coefficients generated from this implementation were not used in the operations until after the implementation described for June 29, 1999.

June 24, 1999
The GTS formatter(SATEM) was modified to perform quality control and then sub sample. The tasks were being done in reverse, which at times led to a reduced sample size. This will ensure that all retrievals selected for distribution are of acceptable quality.

June 29, 1999
Implemented new SPGDVR to remove AMSU-A channel 6 from the CDB for high terrain retrieval channel combination. This will only affect high terrain areas like Antarctica and the Himalayas.

July 6, 1999
Update Coefficients.

July 13, 1999
Update Coefficients.

July 15, 1999
Implemented new SPGDVR to call new RETOD2 which uses correct measurement for precipitable water.

July 19, 1999
Update Coefficients.

July 22, 1999
A new Unified Radiosonde system was implemented.

July 26, 1999
Update Coefficients.

July 30, 1999
Corrected scaling factors on the Retrieval Operational Data File (RODF) which is the primary output file from SPGDVR and primary input file to all product formatters and the retrieval archive. A total of 20 scaling factors were incorrect.

Parameter Old Scaling New (Correct) Scaling
Retrieved Surface Pressure 1 64
First Guess Water Vapor Mixing Ratio (First word) 64 1024
First Guess radiance Temp.(First word) 1024 64
Forecast Relative Humidity 64 256
Forecast surface pressure 256 10
Forecast Pressure 64 10
Potential Temp. time minus Forecast time 10 100
Stability Departure 10 512
Lower Departure 100 512
Time difference (satellite minus forecast) 512 1
Stability forecast increment 512 1
Total Precipitable water (300-500mb) 100 128
Total Precipitable water (500-700mb) 1 128
Sulfur Dioxide 128 1
Polar Redundancy Flag 128 1
Outgoing long wave radiation 1 10
Layer Cooling rate (240-10mb) 1 1000
Layer cooling rate(500-240mb) 10 1000
Cloud comparison flag 1000 1
Library search closeness 1000 1

August 4, 1999
Corrected I1I2I3I4 term for the SATEM to have I3 (Instrument Combination Flag) set to 1 when clear and 3 when cloudy and I4 (Data Processing Technique) set to 3 when cloudy and 4 when clear.

August 9, 1999
Update Coefficients.

August 10, 1999
On August 9, 1999, two NOAA-15 orbits 064344 and 64445 were lost due to antenna problems at Gilmore Creek.

August 17, 1999
Update Coefficients.

August 23, 1999
Update Coefficients.

August 30, 1999
Update Coefficients.

September 1, 1999
Orbit 676364 failed processing.

September 2, 1999
The precipitable water content had been incorrectly distributed in the SATEM as distinct layers when it should have been summation. This was corrected. Additionally, cases were found where the terrain was sea but the sea surface temperature was zero, these situations were eliminated in the SATEM. The problem is the terrain flag comes from the SSM/I processing and the Sea Surface Temperature is from another system with (obviously) a different terrain field.

September 7, 1999
Update coefficients.

September 13, 1999
Update coefficients.

September 20, 1999
Update coefficients.

September 22, 1999
Corrected the "No HIRS" designation to not be flagged when the HIRS really is present. Also changed the initialization of some variables to be properly done.

October 4, 1999
Update coefficients.

October 4 to 6, 1999
Problems with orbital processing resulted in the replacement of the first guess library, CDB and OPCDB at 1545L on October 6.

October 7, 1999
Archive changed to report '0' SST as missing instead of '0', see September 2, 1999 entry for explanation of '0' SST.

October 12, 1999
Update coefficients.

October 13, 1999
Copied clear and cloud Matchup data bases from parallel tooperations, changed CDB after orbit 736768 (data time 1650 to 1835Z).

October 19, 1999
Update coefficients.

October 25, 1999
Update coefficients.

November 2, 1999
Change CDB radiosonde match time windows for the latitude bins to reduce the number of matchups in selected bins. (New time windows by bin 3,6,6,6,4,6,6,4,3,2,3,3,3,3,3,2,2,3,3,3, 3,3,2).

November 4, 1999
Update coefficients.

November 10, 1999
Update coefficients.

November 15, 1999 10:35L
Modified thresholds in CDB.

Novmeber 17, 1999
Update coefficients.

November 17, 1999
Allocate archive with logical record length of 1000 and block size of 9000.

November 18, 1999
Modified thresholds in CDB.

November 23, 1999
Update coefficients.

November 24, 1999 11:54L
Modified thresholds in CDB.

December 1, 1999
Update coefficients.

December 2, 1999
Update thresholds in CDB.

December 6, 1999
Update coefficients.

December 14, 1999
Update coefficients.

December 21, 1999
Update coefficients.

December 21, 1999
Spike in HIRS filter wheel motor current in orbit 832122 (1841 -2027Z). Did not appear to impact data.

December 28, 1999
Update coefficients.

January 5, 2000
Update coefficients.

January 12, 2000
Update coefficients.

January 12, 2000 10:47L
Change window channel test threshold.

January 19, 2000
Update coefficients.

January 27, 2000
Update coefficients.

February 1, 2000

Implemented significant upgrade:
  1. Replaced the radiosonde oriented first guess computation for stratospheric levels with an AMSU-A regression scheme. The two schemes are merged between 50 and 10mb inclusive.
  2. The grouping (binning) of the matchups were changed to use ascending/descending node instead of day/night for grouping in addition to latitude zone and terrain. Additionally, the problem of some bins being reported a full (complete set of matchups) when they were actually incomplete and could hold more matches (and hence improve the selection for the first guess)were corrected.
  3. Consistent screening of the matchups between the eigenvector update, the operator component update and the first guess library build is now being performed.
  4. The eigenvector update was also modified to use ice data for land eigenvectors only (not sea). This program also extends the interpolated temperature profile from the RAOB by merging the interpolated profile generated from AMSU-A channel brightness temperatures.
  5. The orbital processing was modified with these changes:

February 8, 2000
Update coefficients.

February 24, 2000
Update coefficients.

February 25, 2000
Copied clear and cloudy matchup data bases from the production parallel system to the operational system.

March 7, 2000
Update coefficients.

March 14, 2000
Update coefficients.

March 16, 2000
Changed CDB thresholds at 11:20 am.

March 20, 2000
Update coefficients.

March 28, 2000
Update coefficients.

April 5, 2000
Update coefficients.

April 11, 2000
Update coefficients.

April 18, 2000
Update coefficients.

April 25, 2000
Update coefficients.

Previous Section Top of Page Next Section