Meteorological Observations Collected During the European Tracer Experiment (ETEX)

Meteorological observations collected during the European Tracer Experiment (ETEX)

Description of the database

by

Anne Grete Straume

Katrin Nodop

Environment Institute, Joint Research Centre, Italy

August 1997

TABLE OF CONTENTS

Introduction ...... 1

1. Ground observations

1.1 Sonic anemometer data at the release site ...... 1

1.2 Routine synoptic weather observation data ...... 2

2. Upper air data

2.1 At the release site

2.1.1 Constant volume balloons ...... 4

2.1.2 SODAR data ...... 6

2.1.3 Radiosonde data ...... 7

2.2 At remote sites 6

2.2.1 SODAR data from France ...... 9

2.2.2 Upper air data from Europe...... 10

3. Other meteorological information ...... 13

4. Structure of the database...... 14

Acknowledgements ...... 14

References...... 15

Appendix...... 16

Figure A1 Map of stations...... 17

Table A1 List of files in the database...... 18

Table A2 ETEX list of stations for tracer sampling and
meteorological observations...... 20

Table A3 List of synoptic observations in the north-west of France 23

1

Introduction

A variety of long range transport models have been developed in different European countries for modelling gas dispersion over Europe in connection with accidents. After the nuclear accident at Chernobyl in 1986, one saw the necessity to improve the numerical models used to predict the dispersion of a substance released in the atmospheric boundary layer became apparent. The Commission of the European Communities (CEC), the International Atomic Energy Agency (IAEA), and the World Meteorological Organization (WMO) therefore established the European Tracer Experiment (ETEX) to produce a database which can be used to validate some of these models. The experiment consisted of two releases of perfluorocarbons in Brittany, France, during October and November 1994. The outspread of the gas was measured at 168 stations in the central and northern parts of Europe (Figure 1). The gas plume was also detected in the vertical by three air craft from the United Kingdom, Germany and Switzerland. This data is not yet available.

During and after the experiment meteorological information on the state of the atmosphere was collected and stored in a database at the Joint Research Centre, Ispra. This information is vital for the processing and the interpretation of the measurements and the modelled outspread from the experiments. The database contains both ground and upper air observations collected from the time of the tracer release and three days ahead. The ground observations contain sonic anemometer soundings taken from the release site. Further, they include synoptic weather observations from the 168 stations where the gas concentrations were measured. Synoptic weather observations from 80 additional sites in France were also collected from the area near the site of release. The upper air data contains radiosonde, SODAR and constant volume balloon soundings from the release site. Radiosonde data from all WMO radiosonde sites in Europe are also stored for the period from October to December 1994. Further SODAR soundings from 10 nuclear power plants in France are stored for a period of 3 days after each release.

This report gives an overview of how and where these meteorological data are stored, and how to access them. The data are structured in an easy format, but they are not checked for possible typing mistakes before they were provided to us. They have therefore to be checked for inconsistencies before use.

1 Ground observations

1.1 Sonic anemometer data at the release site

At the release site a sonic anemometer was mounted to measure the small scale wind and temperature fluctuations during the period of the release. These measurements give a picture of the turbulent motions within the atmosphere. The data collected are averaged into 10 minute intervals, and in combination of three blocks for half an hour. Each file contains data for one day and are stored for the periods October 23 to October 28 and November 14 to November 19. They are named in the following way: SON10230000 contains data from October 23. The structure of the files is explained in SON_readme. An example of the format of the files is given in Table 1a and b.

Table 1a: Parameters in the sonic anemometer files

decday: / decimal day, which indicates the day of the month.
dechour: / decimal hour from 0 to 24. Each file starts with 3 10-minute averages from 23.30 ,UTC the previous day, and ends at 23.30 UTC the current day.
u: / horizontal velocity.
dir: / wind direction.
t: / temperature
u*: / friction velocity
w'T': / kinematic temperature flux in the vertical
u'T': / kinematic temperature flux in the horizontal x direction
u'u': / kinematic flux of U-momentum in the horizontal x direction
v'v': / kinematic flux of V-momentum in the horizontal y direction
w'w': / kinematic flux of W-momentum in the vertical
T'T': / kinematic temperature flux
u'v': / kinematic flux of U-momentum in the horizontal y direction
u'w': / kinematic flux of U-momentum in the vertical

Table 1b: Example of the format of the sonic anemometer files

30.9792 23.5000 3.66 349.3 12.0 0.1780 -0.0121 0.0254 0.1222 0.0688 0.0497 0.0235 -0.0339 -0.0290
30.9861 23.6667 2.98 355.0 12.0 0.1420 -0.0058 0.0659 0.6467 0.1299 0.0546 0.0241 0.1478 0.0151
30.9931 23.8333 2.03 5.4 12.0 0.0580 -0.0081 0.0328 0.1711 0.0663 0.0367 0.0213 0.0169 -0.0029
1.0000 0.0000 2.23 1.8 12.0 0.0650 -0.0011 0.0086 0.0494 0.0326 0.0132 0.0107 0.0010 -0.0035

1.2 Routine synoptic weather observation data

During the two ETEX releases, routine synoptic weather observations were collected from all 168 stations. Additional observations were collected from 76 stations in the western part of France near the release site.

The meteorological parameters which were collected from the 168 sampling stations are given in Table 2a together with an example of the structure of the data files (Table 2b). The location of the 168 sites, their altitude above sea level and the WMO code is listed in the Appendix. All stations are listed in alphabetic order. Missing values are given by -9, -99, -999 depending on number of digits for each parameter.

The observations are stored in files, one for each time of sample. Each file name indicate at which time the data are collected. For example the file GOD10230000 contains the synoptic data (ground observation data) collected 23 October at 00.00 UTC.

Table 2a: Parameters in the synoptic observation files

"Station_name" / string enclosing the station name.
DATETIME / dattimegroup YYMMDDHH yearmonthdayhour UTC.
9999999 / no synop received.
dd / wind direction in tens of degrees (24 = 240 deg).
ff / mean wind speed in meters per second.
10,11,12 / wind speeds ff_910,ff_911,ff_912 according to WMO regulations in meters per second.
TT, Td / temperature and dew point in tenths of deg. C. E.g. 157 equals 15.7 degree. Celsius below freezing.
PP / pressure reduced to sealevel in hPa.
ww / actual weather [099] according to WMOtable 4677 for ww.
W1,W2 / past weather [09] according to WMOtable 4561 for W1 and W2.
RRR / precipitation during the past 6 or 12 hours WMOtable 3590 for RRR.
N / total cloud cover in oktas.
CL,CM,CH / genera of low, medium and high cloud according to WMOtables 0513, 0515, 0509 respectively.

Table 2b: Example of the files containing the routine synop observations

"Station_name" / DATETIME / dd / ff / 10 / 11 / 12 / TT / Td / PP / ww / W1 / W2 / RRR / N / CL / CM / CH
"A 01 Bregenz" / 94102300 / 4 / 1 / 9 / 9 / 9 / 85 / 84 / 1006 / 9 / 9 / 9 / 99 / 9 / 9 / 9 / 9
"A 02 Feuerkogel" / 94102300 / 21 / 9 / 9 / 13 / 9 / 112 / 3 / 8396 / 9 / 9 / 9 / 99 / 9 / 9 / 9 / 9
"A 03 Innsbruck Air" / 94102300 / 25 / 4 / 9 / 9 / 9 / 84 / 30 / 1005 / 9 / 9 / 9 / 99 / 6 / 5 / 3 / 2

The data from 81 additional stations in France are stored the same way as the synoptic data described above, but there are fewer parameters given: wind direction, wind speed, temperature, relative humidity, pressure and precipitation. The first stations are listed in Table 3, together with the geographical locations and the type of parameters available from the station. The complete list is given in the Appendix. O means that the parameter was available at the station, and N that it was not. The data are also collected every 3 hours. Each file also contains data over a period of 24 hours for all stations. They therefore contain 648 lines each. The format of these files are identical to the file format for the other synoptic observation data described below, except from that the relative humidity is given. This parameter is indicated with RH in the heading of each file. The files are named after the date they were collected, so the file FRW10230000 contains meteorological observations collected 23 October from 00.00 to 21.00 UTC.

Table 3: Stations for synop weather observations in the north-west of France. The type of observations available from each station is indicated with an O or an N. O means that this type of observation is available at the station, and N means that it is not.

Station ;Longit;Latitude;Altit.;DD;FF;T;RH;P;Pr

14;CAEN ;49 10'9 N;000 27'8 W; 64 m;O;O;O;O;O;O

14;PT EN BESSIN ;49 21'0 N;000 46'0 W; 63 m;O;O;O;O;O;O

14;DEAUVILLE ;49 22'0 N;000 10'0 E;144 m;O;O;O;O;O;O

14;L'OUDON ;48 59'0 N;000 00'1 W; 60 m;O;O;O;O;N;O

2 Upper air data

2.1 At the release site

2.1.1 Constant volume balloons

During the two ETEX releases, both the Swiss Meteorological Institute (SMI) and the Laboratoire d'Aerologie at the Paul Sabatier University (UPS), Toulouse, France launched constant volume balloons (CVB) from Monterfil, the site of release. Both institutes launched one balloon each in parallel every 3 or 4 hours during the experiments. The balloons were followed up to 100 km from the site. During the second experiment, only 3 balloons were launched by UPS, and 4 balloons by SMI. Table 5 gives the principal characteristics of the CVB flights. The CVB data are stored in files called CVBxxx ( e.g. CVB10231319, data from the launching at 13h19, October 23).

The radio-sounding system used by SMI was a Vaisala Marwin MW12, equipped with Vaisala RS80L (Loran_C) or RS80N (Omega) system for the wind measurements. The temperature and humidity sensors were not ventilated, nor protected from solar radiation. This means that the temperature measured during sunny conditions is too high. For the wind measurements, an Omega was used on October 23 and 24, 1994, and a Loran_C on November 14, 1994. UPS also used a Vaisala Marwin MW12 radio-sounding system, but the wind measurements were only made with the Omega system. The temperature and humidity sensors were ventilated, and the temperature sensor was protected from direct solar radiation. The data from the Swiss CVB sounding are stored in files with the following contents and format (Table 4a).

Table 4a: Contents of Swiss CVB files

Header
Station: / name of station
Location: / co-ordinates and altitude (masl) of the launching site
LoranC or Omega: / type of navigation system used
Phase fitting length: / filtering used for wind measurement
RSnumber: / Radio sounding number
Started at: / date and time in UTC of launching
Caption
Time: / time after launch in seconds
Speed: / horizontal wind speed
Dir: / direction of horizontal wind
Hgt/MS: / height of sound above mean sea level in meters
Ecomp: / East (X) component of wind in m/s.
Ncomp: / North (Y) component of wind in m/s.
AscRat: / ascent rate of sonde in m/s.
Pressu: / pressure measured by the sonde in hPa.
Temp: / temperature measured by the sonde in oC.
Dewp: / dew point calculation in oC.
MixR: / mixing ratio calculation in g/Kg.
RH: / relative humidity in %. (instead of MixR).
/////: / no data recorded

The French CVB files have a different file structure. Each file contains, as for the Swiss data, measurements from one sounding. The data are stored according to the following FORTRAN format statement: (3(I2), 2(I7), 9(I6)). The files contain a heading written in French before the actual data are listed. The heading gives you the following information (Table 4b).

Table 4b: Contents of the French CVB files

Header
SITE DE LACHER: / Release site
Coordonnees: / Geographical co-ordinates for the release site.
ETEX POI NO: / Number of the release
DATE: / Period were the launchings of the balloons took place.
BCV: / Sounding number
fichier: / file name
Heure de lacher: / Time of launching in UTC
Captions
HeMnSe: / Indicate the time of measurement in hours minutes and seconds.
L: / Position of the balloon in latitudes x 1000.
l: / Position of the balloon in longitudes x 1000.
Z: / Height of the balloon above sea level.
P: / Ambient pressure in hPa times 10.
T: / Ambient air temperature in degrees Celsius x 10.
Hu: / Humidity in percents
DD: / Wind direction
FF: / Wind speed in minutes pr. second x 10.
Vx: / Wind speed in the eastward direction
Vy: / Wind speed in the Northward direction
Vzb: / Ascent velocity of the balloon

Table 5: Principal characteristics of the CVB flights (F - French, S - Swiss) for the first release

CVB / Sonde number / Launch Time (UTC) / End of flight (UTC) / Flight Level (m) / Comments
1 F / 54 / 13:30 / 15:30 / 500 / launching delayed due to local shower
1 S / 441 / 13:19 / 13:58 / 400 - 100 / idem
2 F / 49 / 17:03 / 18:46 / 900 - 700 / launching delayed due to local shower
2 S / 453 / 16:34 / 18:56 / 1500 - 500 / idem
3 F / 44 / 20:10 / 21:40 / 700 - 500 / cvb wet by water deposition
3 S / 842 / 20:00 / 21:47 / 900 - 500 / idem
4 F / 50 / 00:01 / 02:40 / 900 - 900 / cvb wet by water deposition
4 S / 852 / 23:59 / 01:01 / 550 - 100 / idem
5 F / 45 / 03:54 / 05:40 / 900 - 200 / cvb wet by water deposition
5 S / 230 / 03:59 / 05:15 / 550 - 100 / idem

Both constant volume balloon experiments are described in more detail in two reports published by The Swiss Meteorological Institute (Berger, 1995), and Laboratoire d'Aerologie (Benech et. al., 1995).

2.1.2 SODAR data

During the experiment, the Swiss Meteorological Institute had mounted a SODAR (sound detection and ranging instrument) at the tracer release site in Monterfil (2o 30' W, 48o 30' N, 90 m asl). This SODAR collected data from October 22, 19h00, to October 25, 10h00.

The data are divided in blocks, where each block covers one integration period of 30 minutes. Each block starts with a header, giving the block number, date and time (UTC) and number of validations for each beam. The second line contains the operating frequencies, the average Doppler shift in Hz and the environmental noises for each beam. The first column contains the altitude of the measurements. All other columns contain the SODAR measurements. A 9999 value indicates that no measurement is available. An example of the SODAR data files is given in Table 6a-c.

Table 6a: Header description for SODAR files

BL#: / data block number
MONTH, DAY, YEAR: / date of data
HOUR, MIN, SEC: / time of data, corresponds to the end of averaging
period
VAL. 1 to 3: / number of validations on beam 1 to 3
FREQ 1 and 2: / in Hz central frequency for beam 1 and 2
FRASS: / if = 602, the SODAR is in normal mode
be rejected or accepted only with much care.
DOPP 1 and 2: / average Doppler shift in Hz observed on beam 1 and 2
VAL. 4: / average Doppler shift in Hz observed on beam 3
NOIS1 to 3: / average noise on beam 1 to 3. The noise values are given in arbitrary amplitude units. To arrive at dBA values the corresponding formula is: Noise level in dBA = C + 20 log(dec.) NOIS(i). The nominal value for C during ETEX is 16

Table 6b: Parameter description

ALT: / Altitude of measurements above antenna in meters.
CT**2: / Echo strength, no unit
SPEED: / Horizontal wind speed in cm/s
DIR: / Horizontal wind direction in degrees.
S DIR: / Standard deviation of the horizontal wind speed in cm/s
W: / Vertical wind speed in cm/s
SW: / Standard deviation of vertical wind speed in cm/s
SU: / Standard deviation of U axe in cm/s
SV: / Standard deviation of V axe in cm/s

Table 6c: Format of the Swiss SODAR data at the release site

BL# MONTH DAY YEAR HOUR MIN SEC VAL.1 VAL.2 VAL.3
335 11 16 94 17 0 2 152 1171 1258
FREQ1 FREQ FRASS DOPP1 DOPP2 VAL.4 NOIS1 NOIS2 NOIS3
2250 2250 602 11 - 27 0 30 34 34
ALT CT**2 SPEED DIRS DIR W SW SU SV
600 -9999 -9999 -9999 -9999 -9999 -9999 -9999 -9999
580 -9999 -9999 -9999 -9999 -9999 -9999 -9999 -9999
560 429 529 209 4 -26 -9999 41 40
540 412 527 209 5 -30 -9999 52 48
520 396 525 209 5 -32 5 60 48
500 404 523 211 4 -32 5 64 44
480 421 519 212 2 -30 5 73 20

SODAR measurements are affected by noise. Rain, hail or snow and any kind of natural or manmade noise will affect the signal to noise ratio and therefore decrease the range. Limits and accuracy as stated by the manufacturer: Maximum wind speeds are 30 m/s horizontal and +/- 4 m/s vertical. Accuracy for horizontal wind speed is better than 0.2 m/s or 3% for wind speed over 6 m/s. Accuracy for horizontal wind direction is 3 % or better for winds faster than 2 m/s. Vertical speed accuracy is 5 cm/s or better.

The SODAR data are stored in files, named in the following way: The file SOD10221900 contains data for October 22 at 19.00 until midnight. The file SOD10230000 contains data from October 23 at 00.30 UTC to October 24 at 00.00 UTC and so on.

2.1.3 Radiosonde data

The radiosonde data from the release site were provided by MeteoFrance. The radio-sondes were launched every 3. or 4. hour starting on Oct. 23. 12h00 UTC and ending at Oct. 24. 03h00 UTC for the first experiment. There were in total 5 sounding. For the second experiment, the first radiosonde was launched at 12h00 Nov. 14. 12h00, and the last one at Nov. 15 02h00. There were in total 7 sounding. The parameters measured are given in Table 7a and the format of the file is shown in Table 7b.

Table 7a: Parameters of the radiosonde data

1 / Minutes (after the release of the sonde)
2 / Seconds
3 / Altitude (m)
4 / Pressure (hPa)
5 / Temperature (oC)
6 / Relative humidity (%)
7 / Dew point temperature (oC)
8 / Wet bulb temperature (oC)
9 / Mixing ratio (g/Kg)
10 / Water vapour pressure (hPa)
11 / Potential wet bulb temperature (oC)
12 / Potential temperature (oC)
13 / Wind direction (degrees)
14 / Wind speed (m/s)
15 / Launching velocity (m/s)

Table 7b: Format of the files for the French radiosonde launched at the release site

SERVICE : CNRM/GMEI/4M
CAMPAGNE : ETEX
SITE : MONTERFIL
COORDONNEES : 2 0 W / 48 3 N
ALTITUDE SOL : 90
Numero RS : 10
VENT LORAN C
LACHER : 14/11/94 18h37 UTC
RESEAU : 14/11/94 19 UTC
0 0 90 1007.7 13.7 97 13.2 13.4 9.52 15.2 13.1 13.1 237 7.9 0.0
0 2 104 1006.1 14.5 91 13.1 13.6 9.42 15.1 14.0 13.4 244 11.9 7.0
0 4 111 1005.3 14.6 91 13.2 13.7 9.49 15.2 14.2 13.5 245 12.5 3.5
0 6 121 1004.1 14.5 91 13.1 13.6 9.44 15.1 14.2 13.5 246 13.0 5.0

Each file contains data from one sounding. The files always contain a heading which is a combination of 27 characters. The first 5 characters are an identification string. Then the network date and hour is given, and at last the date and hour of the release. The data are displayed in strings of 120 characters, divided in the following format: (15 x A8). Missing data are indicated by -999. The files are containing data from one launching each, and are named in the following way: RAS10231203 contains data from October 23 at 12.03 UTC.

2.2 At remote sites

2.2.1 SODAR data from France

In France 11 nuclear power stations are equipped with SODARs, which are always operational collecting meteorological data. Data from 10 of these SODARs are available for the periods of the ETEX experiment. The SODAR data from France are stored for the following days: 23 - 26 October at 12h00 and 14 - 17 November 12h00. November 12h00. The names and locations of these sites are given in Table 8a.

Table 8a: List of nuclear power stations equipped with SODARs in France

Code / Station name / Latitude / Longitude / Altitude
BV / Belleville / 47o 30' N / 02o 52' E / 144 m
CA / Cattenom / 49o 25' N / 06o 14' E / 189 m
CS / Cruas / 44o 38' N / 04o 48' E / 82 m
CZ / Chooz / 50o 05' N / 04o 47' E / 82 m
GF / Golfech / 44o 06' N / 00o 50' E / 65 m
LH / La Hague / 49o 41' N / 01o 54' W / 178 m
NO / Nogent / 48o 31' N / 03o 32' E / 65 m
SA / St Alban / 45o 25' N / 04o 45' E / 141 m
SL / St Laurent / 47o 44' N / 01o 36' E / 80 m
VA / Valduc / 47o 34' N / 04o 52' E / 492 m

The data set contains surface observations in addition to the vertical data from the SODARs. The data is sorted in 2 data files for each experiment. The file named SOD10231026_surf contains observations at the ground measured between October 23 and October 26. The file SOD10231026_vert contains wind data from the same period at different altitudes above the ground (generally at 50, 100, 150, ..., up to 400 meters). Sometimes the SODARs did not work correctly, and missing data from these periods are indicated by -9, -99 or -999. The parameters and the structure of the files are shown below and in Table 8b.