-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathREADME
112 lines (78 loc) · 4.97 KB
/
README
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
Doug Schuster
Gregg Walters
July 22, 2010
Updated 14 October 2013
Kevin Manross
USING THE BUFR DECODING SOFTWARE
FOR SURFACE DATA IN THE NCAR RDA DATASET DS461.0
(NCEP ADP GLOBAL SURFACE OBSERVATIONS, DAILY OCT1999-CONTINUING)
Here is a list of the bufr_configdecode sub-directories and their descriptions.
bufrobs Use to hold BUFR data files for the dumpbufr.x decoder.
There is a README which describes how to handle some
machine dependencies.
configs Has a few sample configuration files, to be read by
the BUFR decoder, plus a descriptive template.
doc Contains a file showing some basic parameter mnemonics
and a link to recent NCEP BUFR tables.
exe Where the install places the decoder executables.
-grabbufr- No longer needed
install Has a shell script for compiling and installing BUFR
decoders.
lib NCEP BUFR decoder support library - mostly FORTRAN
source code, plus a few C source codes. Includes
its own setup script 'makebufrlib.sh' which is called
from within install/install.sh so you shouldn't need
to run it, but may if you like.
src Two FORTRAN BUFR decoder source code files. A
flexible, user configurable program plus a simple
dumping program.
textobs Used to receive text files generated by the decoders.
All directories (except "textobs") contain their own README files.
** The following text lists the steps necessary to build the software and
decode the BUFR data. Steps include preparing the decoder software, preparing
the input files, preparing the configuration files,
and finally running the decoder software. *****
1. PREPARING THE DECODER SOFTWARE:
-To compile the BUFR libraries and extraction code, go to the "install"
directory and execute the install.sh script to complete the compilations.
Note: instal.sh is set to use 'gfortran' as the compiler and linker. This
is tested and works for linux (including Mac/Darwin variants), but may need
adjusted for other operating systems.
The executables will be placed in the
"exe" directory. You could copy them to an arbitrary directory,
notably your bin, where they could be executed from anywhere.
bufr_configdecode/exe/bufrsurface.x: used to make a selective dump from either
gdas.adpsfc.tHHz.YYYYMMDD.bufr files
gdas.sfcshp.tHHz.YYYYMMDD.bufr files
bufr_configdecode/exe/dumpbufr.x: used to dump all contents of a BUFR file.
2. PREPARING THE INPUT FILES:
-Place your downloaded files, which have the naming structure "gdassfcobs.yyyymmdd.tar.gz",
in the bufrobs directory.
-Execute the "exe/preprocess.csh" script.
-All individual BUFR files are extracted and placed in the "BUFROBS" directory.
-File names must appear in the following format to proceed further:
gdas.adpsfc.tHHz.YYYYMMDD.bufr (land surface)
gdas.sfcshp.tHHz.YYYYMMDD.bufr (marine surface)
3. PREPARING THE CONFIGURATION FILE:
-Go to the "configs" directory for default configuration examples, and instructions
on how to build your configuration file.
-Copy the completed configuration file into the "bufrobs" directory. The name of the
production configuration file needs to be "bufrsurface_config".
-Ensure that the configuration file includes the names of all BUFR files that you would
like to be operated on.
4. RUNNING THE DECODER SOFTWARE:
-To decode the BUFR data according to the instructions in your configuration file,
go into the "bufrobs" directory and execute "../exe/bufrsurface.x"
-To generically dump output from an individual file, go into the "bufrobs" directory and execute
"../exe/dumpbufr.x". From here you will be prompted to enter the name of the file
from which you would like all output to be dumped.
*** Additional information ***
Downloading the RDADB ds461.0 BUFR data files is supported from the "Data Access"
tab near the top of the ds461.0 home page
http://rda.ucar.edu/datasets/ds461.0/
RDA ds461.0 file names have the form: gdassfcobs.yyyymmdd.tar.gz. Individual
BUFR files must be extracted from the tar.gz files for use with the decoders.
This can be done after files are copied to the "bufrobs" directory, by editing
and running the "exe/preprocess.csh" shell script.
Linux users must preprocess all BUFR files using the "preprocess.csh" script, or
the "grabbufr" tool on individual BUFR files.