QBOSingapore

Extract monthly means of mean zonal wind velocities from singapore radiosonde data

1. Save file from Singapore and unzip in TEMP YYYY-MM in the folder ~/Git/QBOSingapore/wmo_archiv
2. Create TAC files using the script ~/Git/QBOSingapore/wmo_archiv/singapore.sh
3. Run FUB processing script ./QBO.sh -g -w -a YYYY.MM in folder ~/Git/QBOSingapore/QBO
4. Create qbo.highres.dat by running singapore2highres.ipynb in the folder ~/Git/QBOSingapore/data
5. Complete the file qbo_100hPa.dat by adding a line with the wind at 100 hPa for the newly added month YYYY.MM.
6. Add 1 for the new month in line 86 qbo4.neu <- array(NA, c(442, 16)) in file qbo_plot.R in folder ~/Git/QBOSingapore/qbo_Rplot.
7. Make new FUB plot by starting R in folder ~/Git/QBOSingapore/qbo_Rplot and running source("qbo_plot.R").
8. Copy singapore.dat and qbo.dat from ~/Git/QBOSingapore/QBO/qbo_data/ to ~/Git/QBOSingapore/srv
9. Update most recent singaporeYYYY.dat file.
10. Create tar and zip files and cp the FUB plot to server directory by running make in ~/Git/QBOSingapore/srv
11. Create jpg file by running make jpg in ~/Git/QBOSingapore/srv
12. Create netcdf file of data by running qbo_fub2netcdf.ipynb in ~/Git/QBOSingapore/srv
13. Copy the following files to the webserver:
    • ~/Git/QBOSingapore/srv/singapore.dat
    • ~/Git/QBOSingapore/srv/singapore2021.dat
    • ~/Git/QBOSingapore/srv/singapore2022.dat
    • ~/Git/QBOSingapore/srv/singapore2023.dat
    • ~/Git/QBOSingapore/srv/qbo.dat
    • ~/Git/QBOSingapore/srv/qbo_data_tgz.tgz
    • ~/Git/QBOSingapore/srv/qbo_data_zip.zip
    • ~/Git/QBOSingapore/srv/qbo_singapore_fub.nc
    • ~/Git/QBOSingapore/srv/qbo_wind.jpg
    • ~/Git/QBOSingapore/srv/qbo_wind_pdf.pdf

DATA PROCESSING OF RADIOSONDE DATA

I. Overview

(1) EXTRACT SINGAPORE

To generate the monthly QBO time series currently the original Singapore radiosonde soundings are used, as they are distributed via the WMO global telecommunication system. All available soundings for 00 and 12 UTC are used, decoded and error checked.

This is done with sub-scripts:

• generate.sh
• parse.sh

The procedure to interpolate for data at 10 hPa for the early data is currently unknown.

The wind information for the main level 100.0, 70.0, 50.0, 30.0, 20.0, and 10.0 hpa (if available) are directly decoded from the sounding. In addition to these main pressure levels, there are some significant points encoded in the sounding, that give a lot of additional, vertically better resolved wind information. As these additional information are on non standard pressure levels, and change from sounding to sounding, interpolation to a set of standard pressure levels is needed.

(2) INTERPOLATION

The interpolation to the final pressure levels 100, 90, 80, 70, 60, 50, 45, 40, 35, 30, 25, 20, 15, 12, and 10 hpa is done linear in p**(R_d/c_p), i.e. p**0.286.

The interpolated data is than averaged to get the final monthly mean wind data.

This is done with sub-script:

• interpolate.sh

(3) PLOTTING

Currently the plotting is done with R. See folder qbo_Rplot, script qbo_plot.R.

(4) Date for web-page

The data files for the web-page are stored in the folder srv. Currently these are: qbo.dat - low resolution QBO since 1953 singapore.dat - high resolution QBO since 1987 singapore2023.dat - high resolution QBO for the current year (2023) singapore2022.dat - high resolution QBO for the last year (2022) singapore2021.dat - high resolution QBO for the last incomplete year to catch up (2021) qbo_data_tgz.tgz - compressed tar file containing current data files qbo_data_zip.zip - zip-file containing current data files

After updating the QBO to the current month, the files singapore2021.dat and singapore2020.dat have to be replaced by singapore2023.dat and singapore2022.dat.

II. QBO-WORKFLOW

QBO.sh - Start script

With this script the following sub-scripts are executed for one single month.

• generate.sh
• parse.sh
• interpolate.sh
• file_add (file_add.f90)

All available data for this month are processed. Execute when the month is finished.

generate.sh -> gen_indata (gen_indata.f90) Extract and gather the coded message for station Singapore from the original data files for 00 and 12 UTC. The original data files are available separatly for each section A/C/D in the folder ~/wmo_archiv/<yyyy>/M<mm> Files (e.g.):

 afw2ha.<dd>00 afw2hc.<dd>00 afw2hd.<dd>00
 afw2ha.<dd>12 afw2hc.<dd>12 afw2hd.<dd>12

Results are stored in folder:
 qbo_data/<yyyy>/M<mm>/
Files:
 sing<yyyy><mm><dd>.00
 sing<yyyy><mm><dd>.12

parse.sh -> parse_wind (parse_wind.f90) Decode the sing<yyyy><mm><dd>.<hh> files. Generates one output file per month:

Folder:

qbo_data/<yyyy>/M<mm>

File:

 wind<yyyy><mm>
 -> contains for all available soundings entries as:
 D<yyyy><mm><dd>.<hh>
 press(z)[hpa] direction(z)[degree] velocity(z)[knots]

The first entries are on 6 standard pressure levels.
All data lines are on uneven pressure levels varying from
sounding to sounding.

interpolate.sh -> interpolate_qbo (interpolate_qbo.f90) Interpolate the data of file wind to 21 main pressure levels. Generates two output files per month:

Folder:
 qbo_data/<yyyy>/M<mm>
File:
 tmp.dat
 -> contains one single line with the monthly zonal wind data on
    the main pressure levels
 wind<yyyy><mm>.txt
 -> contains for all available soundings entries as:

E.g.
First block of data contain original pressure level of sounding
 and the zonal and meridional wind components.

Second block of data contain zonal and meridional wind components
 interpolated to the 21 main pressure levels.

Last section give the monthly mean data for the 21 main pressure
 levels.

 <dd>.<mm>.<yyyy>  <hh> Uhr

 99.9  -0.8  -3.0   94.6  -9.3  -6.5   93.9  -9.7  -6.8   86.4  -3.3  -9.2
 80.3  -8.9 -10.6   75.2 -17.9  -1.6   70.6 -13.9   9.7   70.0 -13.1   9.1
 66.2 -11.8   8.3   64.1 -12.2   2.1   62.2 -10.8   0.9   60.4 -11.1   4.0
 57.8 -11.8   0.0   56.0 -12.2   2.1   52.9 -13.4  -3.6   50.0 -15.9  -1.4
 47.5 -14.7  -2.6   45.2 -16.4   4.4   38.8 -21.6   0.0   34.2 -24.7 -11.5
 32.4 -35.5   0.0   30.0 -30.9   5.4   27.5 -34.3   3.0   26.1 -26.6  -2.3
 23.9  -6.3   5.3   23.6  -5.0   7.2   22.9  -9.1   9.1   20.8  -0.8   4.6
 20.0  -2.6   4.5   19.5  -3.0   3.5   19.2  -3.5   0.6   18.7  -3.0  -6.5
 18.3   2.1  -8.0   17.4   9.8  -5.7   16.8  14.0  -6.5   15.3   9.7  -0.9
 14.3   4.6  -5.5   13.8   5.1  -4.3   13.5   3.3  -3.9   11.8   3.1  -5.3
 11.5   7.1  -6.0   11.0   4.4  -7.6   10.6   6.2  -6.2   10.0   3.6  -6.2
  9.7   1.9  -5.3    9.1  12.7  -2.2    8.1  13.9  -3.7    7.2   8.4   2.3
  6.6   7.2   0.6    6.3  13.9   3.7

 99.9  -0.8  -3.0   90.0  -6.4  -8.0   80.0  -9.4 -10.1   70.0 -13.1   9.1
 60.0 -11.2   3.4   50.0 -15.9  -1.4   45.0 -16.6   4.3   40.0 -20.6   0.9
 35.0 -24.2  -9.4   30.0 -30.9   5.4   25.0 -16.6   1.4   22.5  -7.6   8.3
 20.0  -2.6   4.5   15.0   8.2  -2.2   12.0   3.1  -5.2   10.0   3.6  -6.2
  8.0  13.3  -3.1    7.0   8.0   1.7    6.0   0.0   0.0    5.0   0.0   0.0
  4.0   0.0   0.0

 ...

 Monatsmittelwind fuer den  1 2006

 Niveau Anzahl     U     V   DD FF
  [hPa]        [m/s] [m/s]     [kn]
   99.9     54 -18.4   3.5  101 36
   90.0     54 -10.3  -1.0   84 20
   80.0     54  -4.5  -1.2   75  9
   70.0     54  -1.4  -0.7   62  3
   60.0     53  -3.9  -0.0   90  8
   50.0     53 -15.2   0.4   91 30
   45.0     52 -21.6   0.2   91 42
   40.0     52 -27.1   0.1   90 53
   35.0     52 -30.3   0.0   90 59
   30.0     52 -28.8   1.2   92 56
   25.0     52  -3.6   0.1   91  7
   22.5     51   4.2  -0.3  274  8
   20.0     51   9.5  -1.0  276 18
   15.0     47  16.2   0.6  268 31
   12.0     47  17.4   0.6  268 34
   10.0     46  17.1  -0.7  272 33
    8.0     26  13.0  -1.3  276 25
    7.0     18   9.2  -1.3  278 18
    6.0      6   7.9   3.3  247 17
    5.0      2   3.5   0.7  259  7
    4.0      0   0.0   0.0    0  0

file_add (file_add.f90) -> read in tmp.dat and add the data for this month to the two qbo data files: qbo.dat singapore.dat