![]()
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ICOADS Web information page (Monday, 21-May-2012 17:02:13 UTC): UK Met. Office VOSClim GTS BUFR Data(A) Comparisons of Release 2.5 input data with deck 700 expanded UK Met. Office VOSClim GTS BUFR test dataFigures 1-3 and Examples 1-3 provide duplicate elimination results comparing Release 2.5 (R2.5) input data with a new test file (vosclimbufr700_test_imma.dat) of deck 700 data. In these results we have looked at the ship data versus real-time and delayed-mode receipts from these other decks:
![]() Figure 1. Deck 700 vosclimbufr700_test_imma.dat 12/2003 certain duplicates (DUPS 8-13) with deck 792. For each element, stacked bars of agreement (blue) vs. disagreement (red) (where missing vs. extant data counts as disagreement); and total extant deck 792 (+) and deck 700 (x). Thus any displacement between the symbol and the top of the bar represents missing data. Examples 1. Transient matches between decks 792 vs. 700 (NCEP BUFR vs. VOSClim BUFR).
![]() Figure 2. Deck 700 vosclimbufr700_test_imma.dat 12/2003 certain duplicates (DUPS 8-13) with deck 926. For each element, stacked bars of agreement (blue) vs. disagreement (red) (where missing vs. extant data counts as disagreement); and total extant deck 926 (+) and deck 700 (x). Thus any displacement between the symbol and the top of the bar represents missing data. Examples 2. Transient matches between decks 700 vs. 926 (VOSClim BUFR vs. IMMT).
![]() Figure 3. Deck 700 vosclimbufr700_test_imma.dat 12/2003 certain duplicates (DUPS 8-13) with deck 927. For each element, stacked bars of agreement (blue) vs. disagreement (red) (where missing vs. extant data counts as disagreement); and total extant deck 927 (+) and deck 700 (x). Thus any displacement between the symbol and the top of the bar represents missing data. Examples 3. Transient matches between decks 700 vs. 927 (VOSClim BUFR vs. keyed). Small sample all with ID = CGDS or CG2960.
Table A1. Number of extant BUFR data elements for 1 January 2011 in MetO BUFR (file ShipDay110101) and in NCEP BUFR (file sfcshp.ibm.20110101). Note: A question for the Met Office is why bulletin header fields (ID, CCCC, and BUID) were missing in BUFR file BuoyDay110101, but present in the ship data file (those fields are represented as character data in BUFR). The NCEP BUFR data we utilize for ICOADS are subject to dup-merge, which may account for some reduced counts in those data. In the column labeled "MetO_element_number - BUFR number and description," MetO_element number is "9999" for fields present only in the NCEP BUFR format. Where NCEP has chosen a different BUFR field to represent the same data element appearing in IMMA, this entry appears immediately below the Met Office field. Apparently not all the NCEP BUFR numbers are documented in the WMO Manual on Codes, and thus may be national configurations.
[Documentation and Software][Links to additional]
U.S. National Oceanic and Atmospheric Administration hosts the icoads website privacy disclaimer Document maintained by icoads@noaa.gov Updated: May 21, 2012 17:02:13 UTC http://icoads.noaa.gov/deck700.html |