| |
|
Meteorological data are essential for running dispersion
models. Different data types are often required depending on the model chosen.
Bristol has experience of two modelling packages in which detailed met data
were needed, ADMS urban and Indic Airviro.
Indic Airviro
To run the gauss or grid models in Airviro
it is necesary to have at least two years of data.
The most important parameters to have for modelling are wind, vertical mixing and
temperature. The ideal situation is to have a 25m mast with three temperature
sensors at 22m, 8m ad 2m measuring the temperature differences. As the
temperature difference is about 0.07° per 10m then sensors have to be accurate.
In Bristol two sensors which were calibrated in a pair were fixed to a mast at
2m and 22m. In retrospect it would have been better to locate them at 2m and 8
m as the temperature difference decreases with height.
At 25m an ultrasonic anemometer was located to measure the
wind direction in three dimentions. Ultrasonic
anemometers are becoming cheaper and have the advantage that there are no
moving parts to wear out like a conventional cup anemometer. Airviro also recommends an ordinary horizontal anemometer
located at 10 m and a global radiation sensor to measure sun light.
The ideal location for a met mast is in the centre of the
region to me modelled and an open area. In Bristol’s experience this is very
difficult to achieve. Masts on the tops of tall buildings is possible and for
many years the UK Met Office had a regional office in Bristol which had a mast
on the top of it’s building in the centre of the city. This is a problem,
however, for measuring temperaure differences at 2m
and 8m. A mast was eventually found on top of a hill to the south of the city,
which was ideally located as a regional met mast but was probably above the
boundary layer of the city at certain times.
In the absence of a met mast, obtaining meteorological
data from another source is required. Originally, five years of met data were
purchased from the UK met office and sent to SMHI (the company which owns the
Airviro modelling software) for processing and installation into the Bristol
system. Vertical wind speed is not a standard parameter which is measured in
the UK so calculations were done by SMHI to determine this by using temperature
and cloud cover. We were reliant on SMHI to configure met data for us although
with the Unix based system it is possible to reconfigure a number of the
resource files if there is expert meteorological knowledge available.
Measuring
requirements for Airviro
|
Parameter
|
Requirements
|
Time
resolution
|
|
Horizontal
wind (speed and direction)
|
Threshold
< 0.25 m/s
Accuracy
better than 0.3%
|
15
minute mean value and standard deviation based on sensor output frequency 1Hz
|
|
Vertical
wind
|
Threshold
» 0
Accuracy
better than 3%
|
15
minute standard deviation based on sensor output frequency 1 Hz
|
|
Temperature
|
Accuracy
better than ± 0.1%
|
Mean
value over 15 minutes
|
|
Temperature
difference
|
Accuracy
better than ± 0.3%
|
Mean
value over 15 minutes
|
Airviro
suggested sensors
|
Parameter
|
Sensor
type and configuration
|
|
Horizontal
wind
|
A
propeller anemometer of “Air quality” approved type (according to the US EPA)
|
|
Horizontal
and vertical wind at the top of the mast
|
A three
axis ultrasonic anemometer
|
|
Temperature
|
Platinum
sensor PT-100 in a traditional radiation shield or fan aspirated radiation
shield
|
|
Temperature
difference
|
A
bridge configuration of thermo couples for direct measurements of temperature
difference, mounted in fan aspirated radiation shields.
|
|
Precipitation
|
A
traditional tipping bucket sensor
|
|
Global
radiation
|
Apyranometer
based on a photodiode with a spectral response similar to visual light.
|
ADMS
The UK met office can supply data in ADMS format as can
other suppliers of met data such as Trinity Consultants. If met data of
sufficient standard are not monitored in the locality then approaching the
national meteorological organisations is the best option. It may be that data which
are supplied in this way are modelled or interpolated for the locality and so
will not be as accurate as real monitored data.
Example
of part of a met file in ADMS format.
VARIABLES:
9
STATION DCNN
YEAR
TDAY
THOUR
T0C
U
PHI
P
CL
DATA:
4911.0,1996.0, 1.0,
0.0, 2.6, 0.5, 130.0,
0.0, 8.0,
4911.0,1996.0, 1.0,
1.0, 2.7, 1.0, 130.0,
0.0, 8.0,
4911.0,1996.0, 1.0,
2.0, 2.7, 1.0, 130.0,
0.0, 8.0,
4911.0,1996.0, 1.0,
3.0, 2.7, 1.5, 130.0,
0.4, 8.0,
4911.0,1996.0, 1.0,
4.0, 3.1, 1.0, 130.0,
0.4, 8.0,
4911.0,1996.0, 1.0,
5.0, 3.4, 1.0, 100.0,
0.0, 8.0,
4911.0,1996.0, 1.0,
6.0, 3.9, 1.0, 130.0,
0.0, 8.0,
4911.0,1996.0, 1.0,
7.0, 4.0, 0.0,
0.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
8.0, 3.9, 0.0,
0.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
9.0, 4.0, 1.0,
60.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
10.0, 4.3, 1.0,
30.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
11.0, 4.9, 0.0,
0.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
12.0, 4.8, 1.0,
60.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
13.0, 5.9, 1.0,
50.0, 0.0, 8.0,
4911.0,1996.0, 1.0,
14.0, 4.7, 2.1,
50.0, 0.0, 8.0,
VARIABLES:
9 the
number of columns
STATION DCNN the
station ID number
YEAR the
year
TDAY the
Julian day
THOUR the
hour
T0C degrees
celcius
U wind
speed
PHI wind
direction
P precipitation
CL cloud
cover
When inputing
met data into ADMS it is necessary to enter the height at which the wind is
recorded. This is usually 10m. The data are hourly sequential so the wind
sector angle of 10° is selected. As the data are hourly sequential a check box
is selected which will ensure that the software will calculate the boundary
layer height for each hour. It is unlikely that a boundary layer height would
be measured at a standard monitoring station.
If the location
where the met data is monitored is different from the modelled area then the
“Unrepresentative” checkbox can be selected. For example, a met station located
some distance away from a city may be in the countryside. This then gives the
option to specify a precipitation factor if the rain fall is different and a
surface roughness factor if the modelled area is urban and the met station in in the countryside.
Some Cities are lucky enough to have an ideal site for a
met mast. Although Bristol has a number of met masts they most are not suitable
for modelling purposes as they do not measure all the necessary parameters at
the appropriate height. The perfect mast and location has yet to be found.
References:-
www.airviro.smhi.se
www.cerc.co.uk |