Environmental Engineering Reference
In-Depth Information
elucidate the problems of natural variability, representativeness of measuring site
and model grid averages, etc. A method for validation of models based on esti-
mates of the variability of measured atmospheric parameters was suggested by
Batchvarova and Gryning (2010) and was tested here as well.
Following an applied method of Sreenivasan et al. (1978) for estimation of the
standard deviation of the wind speed and sensible heat flux for a given averaging
time, T, was expressed as:
z
z
σ
θ
=
8
w
'
θ
'
σ
=
12
u
and
.
w
,
T
u
,
T
T
u
T
u
It can be seen that the standard deviation σ increases with height, z, and wind
speed, u, and decreases with averaging time. The standard deviation
increases
σ
w
θ
w
'θ
'
with height and sensible heat flux,
and decreases with averaging time and
wind speed.
2. Sofia Experiment and RAMS6.0 Simulations
In the autumn of 2003, a boundary layer experiment was carried out in Sofia,
comprising turbulence measurements at 20 and 40 m above ground and high
resolution (both in space and time) vertical profiles of temperature, humidity and
wind fields (Batchvarova et al., 2004, 2007) based on radio soundings. Therefore
runs of RAMS6.0 were performed for specific days in September and October
2003 with horizontal resolution up to 1 km for part of Bulgaria. This task required
powerful computing facilities, as the complex topography of the area did not allow
choosing very small innermost grid (
Table 1)
. The CINECA (Bologna, Italy)
computing facilities were used as part of visiting project within HPC Europe ERA
FP6 project. Even with these facilities, a compromise was needed between high
vertical resolution near the ground and high horizontal resolution. Simulations
were performed with three configurations of RAMS6.0 model.
Table 1.
RAMS6.0 configurations for Sofia Experiment 2003
Configu-
ration
code
Simulation period
Number of grid
points in nested
grids
Resolution of nested
grids, time step for
outer grid and time
step ratio
Number of vertical
levels, height of first
level and increase
factor
Case 1
28/09-4/10 2003
start at 0000 GMT
62, 132, 202
25, 5, 1 km
t step 30 s, ratio 1,5,3
42 Levels, start 50 m
increase factor 1.15
Case 2
27/09-30/09 2003
start at 0000 GMT
42, 132, 252
25, 5, 1 km
t step 10 s, ratio 1,5,4
56 Levels, start at 10 m
increase factor 1.15
Case 3
30/09-04/10 2003
start at 0600 GMT
72, 202
25, 5 km
t step 10 s, ratio 1,5
50 Levels, start at 20 m
increase factor 1.15
