INTRODUCTION Eddy-resolving Lidar Measurements and Numerical Simulations...
Transcript of INTRODUCTION Eddy-resolving Lidar Measurements and Numerical Simulations...
Eddy-resolving Lidar Measurements and NumericalSimulations of the Convective Internal Boundary Layer
Shane D. Mayor, Gregory J. Tripoli, and Edwin W. ElorantaDepartment of Atmospheric and Oceanic Sciences, University of Wisconsin
Madison, Wisconsin, 53706, USA
10 JANUARY 1998
distance east of lidar
dist
ance
sou
th o
f lid
ar
0 2 4 6 8 10 12
-10
-8
-6
-4
-2
0
13 JANUARY 1998
0 1 2 3 4 5 6 7 8 9 10-5
0
5
10
15
20 x 10-4
10 January 1998, 14:16-14:57 UT
dive
rgen
ce (s
-1)
Distance east of VIL (offshore distance) km
0 1 2 3 4 5 6 7 8 9 10-2
-1.5
-1
-0.5
0
0.5
1x 10-3
10 January 1998, 14:16-14:57 UT
vorti
city
(s-1
)
Distance east of VIL (offshore distance) km
distance east of lidar
dist
ance
sou
th o
f lid
ar
0 2 4 6 8 10 12
-10
-8
-6
-4
-2
0
The VIL uses a Nd:YAG laser to transmit 400-mJ pulses at 1.064-micron wavelength at 100 Hz.. The VIL resides in a semi-trailer van, employs 0.5-m optics, a beam-steering unit, log-amplifier and real-time-displays. Data are stored on write-once optical disks.
During cold-air outbreaks, steam-fog forms over the lake and is an excellent source of scattering for the lidar.
In this poster we present correlation functions and winds derived from horizontal (PPI) and vertical (RHI) scans of the VIL during Lake-ICE. The observations are used to check a large-eddy simulation of the internal boundary layer.
4.5
5
5.5
6
6.5
7
7.5
8
8.5
9
9.5
10
Downwind distance (km)
Altitu
de (k
m)
0 1 2 3 4 5 6 7 8
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
DOWNSTREAM WIND SPEEDS FROM RHI SCANS ON 13 JANUARY 1998
SPATIALLY RESOLVED 5-m WINDS FROM CROSS-CORRELATION OF AEROSOL BACKSCATTER STRUCTURE
WIND SPEED
WIND SPEED
DIVERGENCE
DIVERGENCE
VORTICITY
VORTICITY
Minimum air temperature at the surface on the morning of 10 January 1998 was -16 C in Sheboygan. Winds were from the WSW at 5-10 m/s. The mixed-layer was about 1-km deep.
Minimum air temperature at the surface on the morning of 13 January 1998 was -21 C in Sheboygan. Winds were from the NW at 5-10 m/s. The mixed-layer was about 400-m deep.
FREE ATMOSPHERE(USUALLY NOT TURBULENT)
STRATOCUMULUS CLOUDS
THERMAL INTERN ALBOUNDARY LAYER
(HEAT AND SHEAR DRIVEN)
WEST WISCONSIN(COLD SURFACE) LAKE MICHIGAN
(RELATIVELY WARM SURFACE ~0° TO 5°C)
EAST
(-15° TO -30°C)
VERY COLD OFFSHORE FLOW
STEAM-FOG
MIXED LAYER
Steam-fog near the VIL site from the NCAR Electra on 13 January 1998. Notice the cellular patterns of fog. Photo courtesy Dave Rogers (CSU).
This photograph was taken from the VIL site looking east over Lake Michigan. Notice the steam-fog and clouds offshore.
Speed and direction as a function of offshore distance. These north-south averages of the wind speed and direction show acceleration and veering as the flow adjusts to changing surface friction, pressure-gradient, and vertical mixing over the water.
0 1 2 3 4 5 6 7 84
5
6
7
8
9
10
30 m
180 m
330 m
480 m
Wind
Spee
d (m/
s)
Downstream distance (km)
Consecutive RHI-scans at a constant azimuth angle (pointed downwind with respect to the mean surface wind direction) produced one vertical slice of the boundary layer every 2 s.
The cross-correlation technique was applied to the RHI scans to measure the downwind component of wind-speed as a function of altitude and downstream distance. Correlation functions were summed over 600-m horizontal bands to reduce noise.
The wind-speeds from the RHI scans show the air near the surface increasing speed while the upper-levels decrease in speed. The vertical gradient of wind-speed decreases offshore because of strong vertical mixing caused by convection.
The wind-field derived from PPI scans of the VIL on 13 January 1998 show a strong wind-speed maximum in the upper right corner. This appears to be caused by Sheboygan Point which lies directly upwind along the shearline.
Photograph of the VIL in Sheboygan, WI, during Lake-ICE. This was taken looking west before dawn with the moon setting. The bright light on the beam steering unit was used to prevent frost from forming on the windows.
VISIBLE GOES-8 SATELLITE IMAGE
RADIOSONDE SOUNDINGS TAKEN 10-KM WEST OF LIDAR SITE
RADIOSONDE SOUNDINGS TAKEN 10-KM WEST OF LIDAR SITE
VISIBLE GOES-8 SATELLITE IMAGE
AEROSOL BACKSCATTER FROM ONE PPI SCAN OF VOLUME IMAGING LIDAR
AEROSOL BACKSCATTER FROM ONE PPI SCAN OF VOLUME IMAGING LIDAR
The 250-m resolution wind-field derived from PPI scans of the VIL on 10 January 1998 shows acceleration and veering of the offshore flow. The largest streak in the divergence and vorticity fields appears to be a building wake originating at the Edgewood power plant located 3-km south of the lidar.
Wind
Spee
d (m/
s)
120-degree component of wind-speed derived from 1290 lidar RHI-scans between 10:44 and 11:08 CST on 13 January 1998.
Divergence as a function of offshore distance. These north-south averages of the divergence show the effect of the acceleration near the shore.
Vorticity as a function of offshore distance. These north-south averages of the vorticity show the effect of veering near the shore.
2 4 6 8 103
4
5
6
7
8
9
10
11
wind speed
wind direction
2 4 6 8 10240
242
244
246
248
250
252
254
25610 January 1998, 14:16-14:57 UT
Distance east of VIL (offshore distance) km0
(degrees)
(m/s)
-40 -30 -20 -100
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Alt
itud
e (k
m)
Temperature C
15:00 Z T 15:00 Z Td16:30 Z T 16:30 Z Td
0 20 400
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Wind Speed (m/s)
15:00 Z16:30 Z
280 300 320 3400
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Wind Direction (degrees)
15:00 Z16:30 Z
-40 -30 -20 -100
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Alt
itud
e (k
m)
Temperature C
13:30 Z T 13:30 Z Td15:00 Z T 15:00 Z Td
0 20 400
0.5
1
1.5
2
2.5
3
3.5
4
Wind Speed (m/s)
13:30 Z15:00 Z
200 250 3000
0.5
1
1.5
2
2.5
3
3.5
4
4.5
Wind Direction (degrees)
13:30 Z15:00 Z
LIDAR HERE
INTRODUCTIONAerosol backscatter data from the University of Wisconsin Volume Imaging Lidar (VIL) are used to check the accuracy of large-eddy simulations (LES) of an internal convective boundary layer.
Wind speed and direction and eddy size and shape are obtained from cross-correlation of the aerosol backscatter data and simulated lidar aerosol backscatter in the LES.
The VIL was deployed in Sheboygan, Wisconsin, during the winter 1997-1998 Lake-Induced Convection Experiment (Lake-ICE).
Sheboygan is located on the western edge of Lake Michigan. The lake does not freeze during the winter.
LIDAR HERE
LIDAR HERE
LIDAR HERE
POWER PLANT
LIDARHERE
LIDAR HERE
LIDAR HERE
LIDARHERE
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
nort
h-so
uth
lag
(m)
1800 m
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
2700 m
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
3600 m
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
4500 m
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
nort
h-so
uth
lag
(m)
1800 m
0.1
0.10
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
2700 m
0
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
3600 m
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
4500 m
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
nort
h-so
uth
lag
(m)
1800 m
0
00
00
0
0.1
0.1
0.2
0
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
2700 m
0
0
0
0
0.1
0.1
0.2
0.2
0
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
3600 m
00
00
0
0
0.10.1
0.2
0.2
0
0
0
0
-200 0 200-300
-200
-100
0
100
200
300
east-west lag (m)
4500 m
0
0
ALT
ITU
DE
(km
)
DISTANCE (km)0 2 4 6 8 10
0.2
0.4
0.6
45 50 55 60 65 70 75 80 85 90
0 2 4 6 8 10
00.5
11.5
COLD OFFSHORE FLOW
RESTORING ZONE
RESTORING ZONE
LAND WATER
30 40 50 60 70 80 90 1001
1.5
2
2.5
3
3.5
4
Relative Humidity (%)
Sca
tter
ing
Fitzgerald et al. JAS 82, data 7-26-79Fitzgerald et al. JAS 82, data 7-27-79Fitzgerald et al. JAS 82, data 7-30-79Fitzgerald et al. JAS 82, data 7-31-79f(RH) = -2.5 + 8.4/(100-RH)2
-200 0 200 400 600 800 1000 12000
0.2
0.4
0.6
0.8
11800 m offshore24 s time-lags13 Jan 1998
Downwind Lag (m)C
orre
lati
on
-200 0 200 400 600 800 1000 12000
0.2
0.4
0.6
0.8
11800 m offshore15 s time-lagsLES #181
Downwind Lag (m)
Cor
rela
tion
LARGE-EDDY SIMULATIONCROSS-CORRELATION of aerosol backscatter data provide quantitative measurements of mean eddy shape, size, orientation, wind speed and direction.
The University of Wisconsin Nonhydrostatic Modeling System
Solves the non-Bousinesq, Quasi-compressible, enstrophy conserving form of the Navier-Stokes equation.
Finite-difference with dynamic conservation principles enforced for enstrophy (vorticity squared), vorticity, kinetic energy, and entropy. (A 3-D extension of Arakawa & Lamb, 1981.)
Subgrid turbulence parameterization uses a prognostic TKE equation that is equivalent to Lilly buoyancy enhanced formulation when TKE applied as a diagnostic. Equivalent to Smagorinsky formulation when TKE is applied diagnostically and shear is the only source of TKE.
Rayleigh damping layer at top of domain to prevent gravity wave reflection. Rayleigh restoring zones at east and west ends of the domain acting on U, V, W, T, and Q.
Periodic lateral boundary conditions
Bulk surface parameterization
5-layer soil model (snow-covered land and water surface in our simulations)
We simulate lidar aerosol backscatter by using the LES relative humidity (RH) and a passive tracer. First, the RH is used in a function to obtain an optical scattering factor. The scattering is multiplied by the tracer and then we take the log of their product to simulate the log amplifer used in the VIL.
USING THE LIDAR DATA TO VERIFY LARGE-EDDY SIMULATIONS
LIDARDATA
MODELRUN A
MODELRUN B
Distance east of lidar (km)
Dis
tanc
e so
uth
of li
dar
(km
)
LIDAR
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0
0.5
1
1.5
2
2.5
3
LES
Offshore distance (km)
Dis
tanc
e al
ong
coas
t (k
m)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0
0.5
1
1.5
2
2.5
3
LES
Offshore distance (km)
Dis
tanc
e al
ong
coas
t (k
m)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
0
0.5
1
1.5
2
2.5
3
LIDAR
Distance offshore (km)
Alt
itud
e (k
m)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
LES
Distance offshore (km)
Alt
itud
e (k
m)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
LES
Distance offshore (km)
Alt
itud
e (k
m)
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 50
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
w in les181 at 7.5 m AGL, 1800 s
distance from shore (km)
dist
ance
(km
)
-6 -4 -2 0 2 4 6
0
0.5
1
1.5
-0.5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
dist
ance
(km
)
distance from shore (km)
w in les195 at 7.5 m AGL, 1800 s
-6 -4 -2 0 2 4 6
0
0.5
1
1.5
-200 0 200 400 600 800 1000 12000
0.2
0.4
0.6
0.8
11800 m offshore12 s time-lagsLES #195
Downwind Lag (m)
Cor
rela
tion
-300 -200 -100 0 100 200 3000
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Downwind Lag (m)
Cor
rela
tion
LIDAR LES A (#181)LES B (#195)
0 20 40 60 80 100 120 1400
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Time (s)
Cor
rela
tion
LIDAR LES A (#181)LES B (#195)
-1 0 1 2 3 44
5
6
7
8
9
10
11
Offshore Distance (km)
Win
d S
peed
(m
/s)
LIDAR LES A (#181)LES B (#195)
Acknowledgements: This work was made possible by NSF ATM9707165 and ARO DAAH-04-94-G-0195. Simulations were run on a J50 donated by IBM. Lidar measurements were made possible by Jim Hedrick, Patrick Ponsardin and Ralph Kuehn.
40
50
60
70
80
90
100
distance from shore (km)
alti
tude
(km
)
Relative humidity
-6 -4 -2 0 2 4 60
0.2
0.4
0.6
0.8
1
1
1.5
2
2.5
distance from shore (km)
alti
tude
(km
)
Scattering Enhancement Factor
-6 -4 -2 0 2 4 60
0.2
0.4
0.6
0.8
1
0
0.5
1
1.5
distance from shore (km)
alti
tude
(km
)
Passive Tracer (aerosol concentration)
-6 -4 -2 0 2 4 60
0.2
0.4
0.6
0.8
1
-1
-0.5
0
0.5
distance from shore (km)
alti
tude
(km
)
Simulated Lidar Backscatter
-6 -4 -2 0 2 4 60
0.2
0.4
0.6
0.8
1
For the simulations presented here:DOMAIN SIZE: 800x120x69=6.6 million pointsRESOLUTION: 15 m in all directionsTIME-STEP: 0.5 sINITIALIZATION: horizontally homogeneousSURFACE TEMPERATURE: -20 C (land) 5 C (water)
LESs explicitly simulate the eddies & plumes in a turbulent boundary layer that are responsible fortransporting heat, moisture, trace gases and momentum.
Autocorrelation using 40-scans
Cross-correlation using 24-secondscan separation
Cross-correlation using 48-secondscan separation
Cross-correlation using 72-secondscan separation
SUMMARY OF COMPARISON OF OBSERVATIONS WITH SIMULATIONS
RELATIVE HUMIDITY (%)
ONE RHI SCAN FROM LIDAR(EAST-WEST VERTICAL SLICE)
ONE EAST-WEST VERTICAL SLICE OF SIMULATED AEROSOL BACKSCATTER FROM MODEL RUN A
ONE EAST-WEST VERTICAL SLICE OF SIMULATED AEROSOL BACKSCATTER FROM MODEL RUN B
ONE PPI SCAN FROM LIDAR(HORIZONTAL SLICE AT 5-m ABOVE SURFACE)
1.8 km 4.5 km3.6 km2.7 km
ONE HORIZONTAL SLICE AT 7.5-m OF SIMULATED AEROSOL BACKSCATTER FROM MODEL RUN A
ONE HORIZONTAL SLICE AT 7.5-m OF SIMULATED AEROSOL BACKSCATTER FROM MODEL RUN B
TWO-DIMENSIONAL CROSS-CORRELATION FUNCTIONS OF AEROSOL BACKSCATTER ON HORIZONTAL PLANES AT GIVEN OFFSHORE DISTANCES
1800-m OFFSHORE CROSS-SECTION OF THE 2-D CORRELATION FUNCTION ALONG A
DOWNWIND ORIENTED AXIS
LIDAR
MODEL RUN B
MODELRUN A
-500 0 500 1000
0
0.2
0.4
0.6
0.8
1
Downwind distance (m)
Cor
rela
tion
-500 0 500 1000
0
0.2
0.4
0.6
0.8
1
Downwind distance (m)
Cor
rela
tion
-500 0 500 1000
0
0.2
0.4
0.6
0.8
1
Downwind distance (m)
Cor
rela
tion
-500 0 500 1000
0
0.2
0.4
0.6
0.8
1
Downwind distance (m)
Cor
rela
tion
0
0.2
0.4
0.6
0.8
1
Nor
th-s
outh
lag
(m)
East-west lag (m)-200 0 200 400 600
-400
-200
0
200
400
0
0.2
0.4
0.6
0.8
1
Nor
th-s
outh
lag
(m)
East-west lag (m)-200 0 200 400 600
-400
-200
0
200
400
0
0.2
0.4
0.6
0.8
1
Nor
th-s
outh
lag
(m)
East-west lag (m)-200 0 200 400 600
-400
-200
0
200
400
0
0.2
0.4
0.6
0.8
1
Nor
th-s
outh
lag
(m)
East-west lag (m)-200 0 200 400 600
-400
-200
0
200
400
The fundamental advantage of LES is its ability to explicitly calculate fluxes due to coherent structures. However, the modeled eddies can be sensitive to numerical methods. Conventional data lack the spatial and temporal resolution to evaluate the fidelity of the simulated structures.
Here we present two simulations that differ only by the numerical advection scheme.
In general, both simulations show spatial and temporal eddy-structure which is remarkably similar to the lidar data. However, these two simulations differ markedly near the lake-edge. In particular, run B shows more elongated structures than the lidar data. The structures in both simulations are tend to orientations aligned with the domain rather than the wind direction as shown in the lidar data.
The wind accelerates too rapidly as it leaves the shoreline in both simulations. This may be related to lack of shoreline topography in the model or errors in the initialization of the pressure-gradient. Our current simulations show large-eddies of only about half the mixed-layer depth developing over the 6-km fetch of land in the model.
WIND DIRECTION
0.0
0.1
0.2
VERTICAL VELOCITY (m/s)
VERTICAL VELOCITY (m/s)
The images above show vertical velocity across the entire model domain at 7.5 m above the surface. The advection scheme used in Model Run A encouraged fine-scale roll-circulations to develop farther upstream than the advection scheme used in Model Run B.