ann-sofi smedman uppsala university uppsala, sweden · ann-sofi smedman uppsala university uppsala,...
Post on 29-Jun-2020
4 Views
Preview:
TRANSCRIPT
Effects of transfer processes on marine atmosphericboundary layer
orEffects of boundary layer processes on air-sea
exchange
Ann-Sofi SmedmanUppsala University
Uppsala, Sweden
Swell
Long wavescreateform drag
Upward momentum flux
Quasi-frictional decouplingat the surface
Turbulence resembles free convection conditions
Ocean
MABL
Effect of transfer process on MABL
Effect of MABL process on air-sea exchange
MABL
Detached eddiesimpinging on the surface
Increased small scaleheat flux
Increased CHNand CEN
Ocean
UVCN regime
In models air-sea exchangeis described through Monin-Obukhov similarity
theorya theory which is well tested over land
but
is it valid over the ocean?
Large heat capacity
Small diurnal variation
Roughness (waves)increasing with
increasing wind speed
What are the differences ?
10002200
2 buoys (temp, waveheight , dir. and CO2 Footprint area
Tower 30 mTemp and wind profiles 5 levels
Turbulence 3 levelsHumidity and CO2 at 2 levels
Long term measurements in the Baltic Sea
Turbulence, wind speed, temperature and waveparameters
Short term experiments with RV Aranda and ASIS buoy
In the mean, excellent agreement between tower and ASIS
Data set from Agile experiment in lake Ontario 1994-95
A 15- m research vessel Agile was equipped to measure waves and turbulence with
• Sonic R2A• Thermocouple• Licor• Wave staff array
Measurements were performed at 7.8 m during two autumn periods
Growing sea (slow waves)Unstable stratification
• Monin-Obukhov similarity theory is validbut
roughness length, zo, is a function of wave age
0.04 0.05 0.06 0.0710
−6
10−4
10−2
Growing sea
u*/c
p
z 0/σ
Swell (long waves traveling faster than the wind)
andunstable stratification
• A quite different type of boundary layer develops
The swell driven BL is characterized by
• Small or positive momentum flux• Low level jet• M-O scaling does not hold• The vertical extent can be global• Turbulence spectra resemble free convection conditions• Swell represents reverse atmospheric-ocean coupling
−0.04 −0.035 −0.03 −0.025 −0.02 −0.015 −0.01 −0.005 0 0.0050
5
10
15
20
25
30
u’w’ (m2s−2)
heig
ht (
m)
C2 C
p/U
8=1.79C3
CP/U
8=1.61Shearing stress
C1C
p/U
8=4.72
F1C
p/U
8=4.66
F2C
p/U
8=1.73
Small momentum flux
uw x 1000
positive uw
Aircraft measurements
Tower measurements
Positive momentum flux
LES simulations Peter Sullivan
0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 20
5
10
15
20
25
30
35
U (ms−1)
heig
ht (
m)
Mean wind profiles
Cp/U
8 = 4.72 C
p/U
8 = 4.60
Case C1 Case F1
Low level jet
2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 70
5
10
15
20
25
30
35
U (ms−1)
heig
ht (
m)
Mean wind profiles
Cp/U
8 = 1.79 C
p/U
8 = 1.61 C
p/U
8 = 1.73
Case C2 Case C3 Case F2
0 2 4 6 8 1010
0
101
102
103
104
U (m/s)
z (m
)
Pibal tracking 03090610−23 LST
The vertical extent is global
−1 0 1 2 3 4 5 6 70
5
10
15
20
25
30
φm
heig
ht (
m)
φm
= (1−15z/L)−1/4for L= −30 m
C1Eq.C2F1F2C3
Φm calculated for five swell cases
Cp/U8=4.7
Cp/U8=1.6
Turbulent kinetic energy budget
ερ
θ +∂′∂
+∂
′∂⋅+′−
∂∂⋅′′=
zew
zwpw
Tg
zUwu
oo
10
P B Tp Tt D
−7 −6 −5 −4 −3 −2 −1 0 1 2 3 4 5
x 10−3
0
5
10
15
20
25
30
TKE (m2s−3)
heig
ht (
m)
P B Tp Tt D
Growing sea
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
Frequency − n (Hz)
Ene
rgy
− S
(n)
(m2 s−
1 )
Growing sea case
npeak
= 0.42 Hz
Turbulent energy budgetGrowing sea
−3 −2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 2 2.5 3
x 10−4
0
5
10
15
20
25
30
TKE (m2s−3)
heig
ht (
m)
Case F1
BTp P Tt ε
Turbulent energy budgetcp/U=4.7
0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 20
5
10
15
20
25
30
35
U (ms−1)
heig
ht (
m)
Mean wind profiles
Cp/U
8 = 4.72 C
p/U
8 = 4.60
Case C1 Case F1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
Frequency − n (Hz)
Energ
y −
S(n
) (m
2s
−1)
Case F1
nswell
= 0.22 Hz
−5 −4 −3 −2 −1 0 1 2 3 4
x 10−3
0
5
10
15
20
25
30
TKE (m2s−3)
heig
ht (
m)
Case C3
P B Tt Tpε
2 2.5 3 3.5 4 4.5 5 5.5 6 6.5 70
5
10
15
20
25
30
35
U (ms−1)
he
igh
t (m
)Mean wind profiles
Cp/U
8 = 1.79 C
p/U
8 = 1.61 C
p/U
8 = 1.73
Case C2 Case C3 Case F2
Turbulent energy budgetcp/U=1.6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10
0.005
0.01
0.015
0.02
0.025
0.03
Frequency − n (Hz)
Energ
y −
S(n
)(m
2s−
1)
Case C3
nswell
= 0.22 Hz
nmin
=0.35 Hz
)2
''''(''
1 2
εθ ++∂∂
+−−=∂∂
pv Tqwz
wTg
wuzU
'')(
wuT
zU p
Tp −=∂∂
dzuT
mUzUz
m
p∫=−5.2
2*
)5.2()(
From the turbulence energy eq,
The effect of Tp can be isolated
4.5 4.7 4.9 5.1 5.3 5.5 5.7 5.9 60
5
10
15
20
25
30
35
U (ms−1)
heig
ht
F2 (based on numerical expressions for Tp)
blue: measured U−curve
red: measured U−curve minus Tp/u*2
Heat fluxes
All turbulent fluxes in models must be expressed as mean variables
)( soH TTU
wC−′′
=θ
Wind speed Air-sea temperaturedifference
Stanton numberHeat flux
In the same way for latent heat, CE
)}/)}{ln(/{ln( 00
2
THN zzzz
C κ=
Reduced to neutral stability
Von Karman constant=0.4
Roughness length Roughness lengthfor temperature
In the same way for latent heat, CEN
Heat fluxes, unstable stratificationΔT>3o
• Traditional parametrization (COARE algorithm) is valid
• No variation with wave age
Heat fluxes, stable stratification
• Low-level jets cause shear suppression leading to decreased fluxes
Heat fluxes, unstable stratification, very close to neutral
Wind speed > 10 m/s and ΔT< 3o
Fluxes are underestimated with COARE algorithm
10−4
10−3
10−2
10−1
100
101
102
10−4
10−3
10−2
10−1
100
101
nSu,
w(n
)/u*2 Φ ε2/
3 nS T(n
)/T*2 Φ ε−1
/3Φ N
T
f=nz/U
Tuw
Unstable stratification -- horizontal rolls
U T
w
10−4
10−3
10−2
10−1
100
101
10−3
10−2
10−1
100
101
f=nz/U
nSu,
w(n
)/u *2 Φ
ε2/3 n
ST(n
)/T
*2 Φε−1
/3Φ
NT
Tuw
Unstable near neutral -- detached eddies
UT
w
10−4
10−3
10−2
10−1
100
101
102
0
0.05
0.1
0.15
0.2
0.25
f=nz/U
nCO
wt(n
)/w
t
Norm. heat flux Östergarnsholm
L=−10 m
L=−42 mL=−140 m
L=−250 m
Normalized heat fluxes
2 4 6 8 10 12 140
0.5
1
1.5
2
2.5
3x 10
−3
U10
CH
NMIUU, L > −150m
data2
R2, 1 m/s bins
MIUU, L <−150m; 0.5<(Tw − T10) <1.0K
MIUU, L < −150m; 1.0 < (Tw − T10) < 1.5K
MIUU, L < −150m; 1.5 < (Tw − T10) < 2.0 K
ΔT<3o
Sonic temp.
Platinum sensor ( o ***)
Heat flux from Östergarnsholm
Latent heat flux, Östergarnsholm
COARE
ΔΤ<2ο
2 4 6 8 10 12 14 16 180.5
1
1.5
2
2.5
3
3.5
4x 10
−3
CH
N
U10
SonicTC
ΔT>3
ΔT>3
0.8
0.9
0.8
0.6 0.7
2.18
0.752.2
2.7
2.0 3.0
2.3 2.0
..
Agile heat flux
thermocouple
sonic
ΔT<3o
2 4 6 8 10 12 140
0.5
1
1.5
2
2.5
3x 10
−3
U (ms−1)
CE
N
2.98 3.04
3.1
3.13.17
3.13
3.2
2.06
2.05
2.32
6.19
6.14
6.22
5.562.26
Agile Licor, ΔT<3
Agile latent heat flux
Solid curves data from Östergarnsholm
2 4 6 8 10 12 14 16 18−0.01
−0.005
0
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04Comparison of heat fluxes, differences, Agile−94,95
w´t
´ TC −
w´t
´ S (
m s
−1 K
)
U10
(m s−1)
Agile heat flux: (wt)TC- ( wt) sonic
The unstable very close to neutral regime
• The UVCN regime is defined for ΔT<3o and U>10m/s
• It is found both over land and sea in the lowest region (z<15 m) of the surface layer
• Small scale heat flux is a result of detached eddies originating from the upper part of the surface. This is in contrast to a convective surface layer ruled by horizontal rolls.
Frequency (%) of occurrence of UVCN conditions
ERA-40 reanalysis
Increase (W/m2) in latent heat flux with UVCN regime included
ERA-40 reanalysis
Global mean values of the increase of heat fluxes with UVCN regime included
• Sensible heat flux= 0.8 W/m2
• Latent heat flux (evaporation)= 2.4 W/m 2
• The effect of increased anthropogenic greenhouse gases (2006) on net radiation = 2.63±0.26 W/m 2
Conclusions
Unstable stratification
• M-O similarity theory is only valid for growing waves
During swell the MABL is characterized by
• Low-level jet (LLJ) at 5-10 m height• Constant wind speed above LLJ• Small or positive momentum flux varying with height• Energy is transported from waves to atmosphere by
pressure transport term Tp
Mixed seas
As soon as a small portion of the waves are travelling faster than the wind (mixed seas) MABL is slowly changing from an ordinary boundary layer to a ‘swell dominated ‘ boundary layer and
Monin-Obukhov similarity theory is not valid
During stable stratification wave effects are small
Heat fluxes
• The dynamics of the MABL influence the air-sea exchange of sensible and latent heat.
• When the convective boundary layer is ruled by horizontal rolls the well known COARE algorithm is valid.
• When detached eddies originating from the upper part of the surface layer is present (UVCN regime) the heat fluxes are increased and the Φh-function goes to zero
Why has not the UVCN regime been observed earlier?
• sonic temperature fails to measure temperature fluctuations correctly when σT is small (high wind speed, small ΔT)
• data with small ΔT are often removed from the data set
• measurements are taken at heights above the UVCN layer.
Thanks to• Ulf Högström, Erik Sahlee, Anna Rutgersson,
Hans Bergström, Cecilia Johansson, MIUU
• Will Drennan, RSMAS, Miami, Kimmo Kahmaand Heidi Pettersson, FMRI, Helsinki
• Phd-students: Andreas Karelid , Björn Carlsson, Maria Norman, Alvaro Semedo
• Former Phd-students: Cecilia Johansson Birgitta Källstrand, Anna Rutgersson, Anna Sjöblom, Xiaoli Guo Larsen
top related