Importance of the Salinity Barrier Layer

1
Importance of the Salinity Barrier Layer for the
Buildup of El Nino Maes et al., J.Climate,
2005 Presented by Sindu Raj Parampil
2
Objective To explore the importance of BL in
maintaining heat buildup in Western Pacific
prior to El Nino Outline
El Nino events during 1993 2002 as
characterized by SST anomaly, Thermosteric
Anomalies BLT Results from Control
experiment of Coupled Model mean state
seasonal cycle Analyses of a strong El Nino
from model simulations a. Control run
b. 2 stage perturbation experiments
Physical processes involved during buildup prior
to El Nino
3
Bjerknes (1969) gt ENSO is the result of the
interaction between
surface winds SST Wyrtki
(1975,1985) -gt sea level observations Cane
Zebiak (1985) -gt simple coupled model gt Period
between 2 El Nino events correspond to the time
required to recharge equatorial band with
warm water Wyrtki (1975) gt Heat buildup in
Western Pacific is a precursor to El Nino Meinen
McPhaden (2000) -gt Observations for 1980 99
gt Variations in SST, zonal wind stress
East-West slope of thermocline in phase
Heat buildup lead by 7 months
Equatorial heat buildup is an intrinsic element
of El Nino !!

4
Why is the heat buildup not
dissipated? Observations show that Mixed Layer
is insulated from deep cold waters by strong,
salinity stratification within the warm
isothermal surface waters i.e., Barrier Layer
(BL) Maes et al (2002b) -gt coupled model gt
Removal of BL during the El Nino onset resulted
in reduced or aborted warm event
BL
5
Time-longitude sections (2ºN -2ºS)
black line 28.5ºC isotherm Contour interval
0.5ºC, 3 dyn-cm 10m
6
A detailed study of the relation between BLT
and heat buildup requires basinwide insitu
observations -gt currently impossible GCMs -gt
good alternatives, depending on its ability to
simulate observations Coupled
model Meteo-France AGCM -gt ARPEGE forecast
model (Deque et al.,'94) Spectral T31 triangular
horizontal truncation 3.75o resolution in
tropics LODYC OGCM -gt Ocean Parallelise code
(Madec et al 1998) domain 120oE 70oW 30oN
30oS ( tropical Pacific ) zonal resolution -
1o latitudinal resolution 0.5o at equator 2o
at meridional boundaries
7
Coupled Model Seasonal Cycle
Time-longitude sections (2ºN -2ºS) thick black
line 28.5ºC isotherm
8
SSTA Cold tongue development Aug Sep
SST gt 29oC not always present in Western
Pacific TauX Semiannual cycle in east (180o
120oW) annual
cycle to west 180o Westerly winds
in winter gt Westerly Wind Bursts (WWBs) BLT
significant only in Western Pacific
annual cycle largest values ( 20m) in boreal
winter Relationship between BLT warmest
waters is evident. In agreement with recent
observations made by Delcroix McPhaden
(2002) Model capable of capturing SST wind
interactions on seasonal time scales !!
9
Control Run years 14 - 17
Contour interval 0.5ºC, 3 dyn-cm 10m
10
Mar -Jun -gt warm water throughout the equatorial
band Warm SSTA in central eastern regions of
the basin associated with the arrival of
downwelling equatorial Kelvin Waves End of yr-15
-gt warm pool reaches its easternmost
extent. Development of ve thermosteric
anomalies traced back to western Pacific during
the middle of yr-14 but SSTA are not well
defined. Also there is the presence of a
significant BLT, as seen in observations
11
Is the relationship between BLT and heat buildup
a robust systematic feature of the model?
Interannual evolution of Nino 3.4 (5oN 5oS,
120oW 170oW) SST (black), 0-500 dbar
thermosteric (dashed), BLT (grey) Anomalies in
the western Pacific (4oN 4oS, 130oE 180o)
over 30-yr.
SST index (oC) left axis Thermosteric
anomalies (dyn cm) BLT (m) anomalies right
axis
12
The 2-stage perturbation experiment I stage
Vertical mixing scheme in the ocean model made
independent of salinity stratification
4oN 4oS, SST 28oC, period prior
1-yr to mature phase of El Nino (Oct
of year 14 ) gt Increase in vertical mixing
destroys / reduces salinity stratification
II stage Normal formulation of mixing scheme
restored run for 18 months gt
Impact of reduced BLT modified heat buildup on
El Nino
13
Temperature difference between buildup year
30-yr climatology
Thick white line 28.5oC ve anomaly in west
reduced by factor of 2 Shift of heat buildup
from west to east.
14
Thus, reduction in the BLT does reduce the heat
buildup prior to the onset of the strong El Nino

BUT Is
the reduction large enough to impact the El Nino
development? 6 member ensembles generated for
CTRL PERT runs - initial condition of Warm
pool SST were given perturbations
O(0.1oC) CTRL ensemble SST perturbations
applied once PERT ensemble SST perturbations
applied at start of II stage
15
CTRL ensemble (black lines) SSTA gt 1oC
westerly wind anomalies in Central Pacific
east-west tilt of thermosteric anomalies flatten
at edges of the basin PERT ensemble (gray
lines) ve SSTA disappear no westerly wind
anomalies except in far west WWBs
thermosteric anomalies ve but constant over
entire equatorial band Hence, reduction of
buildup due to removal of BL, a year prior to the
event can clearly be associated with annihilation
of the strong El Nino.
16
CTRL
PERT
Thick white line 34.25-psu SSS Contour interval
1ºC, 0.01 Nm-2 10m
17
CTRL Nov yr-14 to Apr yr-15 series of WWBs
over the warm pool
generate downwelling equatorial
Kelvin waves seen in
20oC depth anomaly I series Eastward
displacement of Warm Pool II series Jan Feb,
depress thermocline propagate east III series
Jul Aug, Kelvin waves arrive in east during
upwelling season PERT Oct
yr-14 to Jan yr-15 series of WWBs but with
extended fetch Jan thermocline depressed by 10
20m. Mar Apr few WWBs but no Kelvin waves
End of Apr heat buildup lost does not
recover. Jul upwelling season is more
pronounced.
18
Sensitivity of model is strong during Oct
Mar and dynamics of the Warm Pool has an
important influence during this period!!

Parameters averaged over
2oN 2oS band East gt 25o to east of SSS
West -gt western boundary to SSS value

Mixed layer heat budget
19
Conclusions 1. BL plays an important role in
the dynamics of El Nino. 2. In this region, heat
buildup is a direct consequence of the
suppression of entrainment cooling below the
mixed layer, due to the presence of the BL.
3. Relationship between the heat buildup and BL
is a necessary condition for sustained
accumulation of heat in the western Pacific.
4. The subsequent development of El Nino in
central eastern equatorial band is a remote
response to Kelvin waves generated by
SST-wind coupling over the warm pool.
20
5. The removal of the BL in the western Pacific
causes the reduction and discharge of the
heat buildup in the western Pacific before
El Nino onset and thus prevents the development
of El Nino
21
Reference Maes. C., J.Picaut and S. Belmari,
2005 Importance of the Salinity Barrier Layer
for the Buildup of El Nino. J.Climate
22
Thank You