Document Sample
alexander Powered By Docstoc
					                             Midlatitude Upper Ocean Processes

                                         Michael Alexander

                          NOAA-CIRES, Climate Diagnostics Center

   Beginning with the pioneering work of Namias (e.g. 1959, 1963, 1965, 1969) and Bjerknes

(1964), many studies have sought to understand how midlatitude sea surface temperature (SST)

anomalies form and the extent to which they influence the atmosphere. SST anomalies tend to

extend over a nearly uniform surface mixed layer, and variability in the mixed layer depth (h) can

influence the evolution of SSTs and exchanges between the surface and deeper layers in the ocean.

Here, we will develop the energy budget of the ocean surface layer and examine the physical

processes which influence SST anomalies and exchange of heat with the deeper ocean.

SST Characteristics
   As a first step, we examine the basic characteristics of SSTs over the Northern Hemisphere

oceans. The annual mean SST maxima are found in the western tropics of both ocean basins (Fig.

1a). The strong gradients in SST located between 35 and 45N correspond to the Gulf Stream and

the Kuroshio extension in the Atlantic and Pacific, respectively. The seasonal cycle, as indicated by

the August – March SST difference, exceeds 5C over most of the ocean north of 25N and 10C

on the far western side of both basins (Fig. 1b).      In midlatitudes, the departure of the mean

seasonal cycle from the annual mean is much greater than the interannual SST variability as

indicated by the ratio of standard deviation of monthly SST anomalies to that associated with the

mean seasonal cycle (Fig. 2). This ratio is less than 0.3 over much of the North Atlantic and Pacific
but exceeds 2.0 in the equatorial Pacific where interannual SST variability associated with El Niño/

Southern Oscillation (ENSO) dominates over the seasonal cycle.

   The surface layer over most of the world’s oceans is vertically well mixed and thus,

heating/cooling from the atmosphere is spread from the surface down to the base of the mixed

layer. The surface mixed layer (ML) has substantial heat capacity given that it is relatively thick (in

contrast, only the upper few centimeters of land absorb heat from the atmosphere), and that

seawater is dense and has a high specific heat. For example, a surface flux of ~100 W m -2 is

required to warm a 65 m mixed layer (which is close to the annual mean h over the northern

oceans) by 1C month-1. Due to the large thermal inertia of the surface layer, SSTs reach a

maximum in August-September and a minimum in March (Fig. 3), about 3 months after the

maximum and minimum in solar forcing, compared to a one-month lag for land temperatures.

   The mixed layer is deepest in late winter, when it ranges from 100 m over much of the North

Pacific to more than 500 m in the far North Atlantic, but shoals to around 20-30 m in late spring

and summer (Fig. 3, also see Monterey and Levitus 1997; Alexander et al. 2000a). Since, h is

approximately 5–20 times smaller in summer than in winter, less energy is required to generate

SST anomalies, leading to larger SST variability in summer compared with winter (Fig. 4).

Surface Heat Budget

   The first law of thermodynamics can be written as:

    T             T           T 
        v T  w     AT   K    ,                                                         (1)
    t             z         z  z 
where v and w are the horizontal and vertical velocities, and A and K are the horizontal and

vertical diffusion coefficients. Integrating Eq. (1) over the depth of the mixed layer and assuming

that A and K are constants, the heat budget for the ML, a la Frankignoul (1985), can be written as:

    Tm              w  we               Qnet  Qswh
         v Tm           Tb  Tm                A2Tm                                  (2)
     t              h                      o c p h

where Tm (= SST) is the mixed layer temperature. The vertical velocity in Eq. (2) is composed of

the mean vertical motion w and the entrainment velocity we, which is the turbulent flux through the

base of the mixed layer. Tb denotes temperature just below the ML base and  o and c p are the

density and specific heat of ocean water. The third term on the right hand side (RHS) of Eq. (2)

represents the net surface heat exchange together with the energy exiting the ML through

penetration of solar radiation, Qswh, into the deeper ocean. The final term represents horizontal

diffusion due to eddies, which acts to weakly damp SST anomalies.

   The horizontal temperature advection, the first term of RHS of Eq. 1, is primarily due to Ekman

( vek ) and geostrophic ( v g ) currents. The former represents the total transport through the mixed

layer and is given by vek = -k  /o f , i.e. it is 90º to the right of the surface wind stress in the

Northern Hemisphere. The latter, v g , is related to the sea surface and thermocline slope, and

depends on boundary waves and the local and domain integrated affects of Ekman pumping.

Generally, SST variations due to vek are larger than those for v g on seasonal to interannual

timescales in midlatitudes (Gill and Niller 1973; Frankignoul 1985; Seager et al. 2000). The

standard deviation of the anomalous meridional Ekman advection, vek  yTm (overbar and prime

denote the monthly mean climatology and the departure from the mean, respectively), is shown for
January in Fig. 5. The variability maximum located between 35-45N is coincident with the

largest mean temperature gradients (Fig. 1). The term vek  yTm is the dominant component of the

total anomalous Ekman heat transport (Fig. 5, bottom) since  yTm is much greater than  xTm ,  xTm

and  yTm .

    In the open ocean of the extratropics the vertical mass flux into the mixed layer is primarily due

to entrainment (Frankignoul 1985; Alexander 1992a), i.e. we > w. The entrainment velocity is often

estimated from the turbulent kinetic energy equation (Niller and Kraus 1977; Gaspar 1988) and can

be broadly represented by:

     w e = M + B - D / ( - S)                                                                 (3)

where M is the mechanical turbulence generated by wind stirring (~ u*  (  air /o )3/2 ), B is the

buoyancy forcing which is primarily due to Qnet, D is the dissipation, and and S are the density

jump and current shear across h (the latter is not included in all ML models). The ML deepens via

entrainment; anomalies in we are primarily generated by wind stirring in summer and surface

cooling in fall and winter (Alexander et al. 2000a). The mixed layer shoals by reforming closer to

the surface, there is no entrainment at that time (we = 0), and h is the depth at which there is a

balance between surface heating (positive buoyancy flux), wind stirring and dissipation. In general,

deepening occurs gradually over the cooling season while the mixed layer shoals fairly abruptly in

the spring.

    On seasonal to interannual timescales, the time tendency of Tm is primarily due to the net

surface heat flux, Ekman advection, and entrainment heat flux (geostrophic advection may also be

important on decadal timescales, e.g. see notes by Marshall). The long-term mean values of these
terms in January are shown in Fig. 6. In the western basins of the North Pacific and the North

Atlantic, the magnitude of Qnet is more than 3 times as large as the Ekman and entrainment terms.

The anomalous net heat flux is also the dominant term that generates SST anomalies during winter

over most of the Northern Hemisphere Oceans (Fig. 7; Cayan 1992a&b; Battisti et al. 1995;

Delworth 1996). However, Ekman advection contributes to SST anomaly development in the

narrow band with strong temperature gradients: along 40N in the west Pacific and off the northeast

coast of North America (Fig. 7, middle) and as a result of changes in deep convection, entrainment

has a significant influence on SST' near Greenland. The relative contributions of the three terms is

quite different in summer: the interannual standard deviation of the temperature tendencies due to

Ekman advection and entrainment in August exceed those due to net heat flux over portions of the

midlatitude oceans (Fig. 8).

   The net surface heat exchange has four components: the shortwave (Qsw), longwave (Qlw),

sensible (Qsh) and latent (Qlh) heat fluxes. Qsw is a functions of cloudiness, humidity and the

surface albedo. Fluctuations in cloudiness, especially stratiform clouds, have a strong influence on

Qsw especially over the North Pacific in spring and summer (Weare 1994, Klein et al. 1995, Norris

and Leovy 1994). Increased clouds cool the ocean, while a colder ocean enhances the static

stability, which lead to an increase in stratiform clouds. As a result, there is the possibility for

positive feedback, as indicated by an inverse relationship between low clouds and SST anomalies.

Indeed, Norris et al. 1998 found that summer stratus cloud amount and SST anomalies are

negatively correlated on both interannual and decadal timescales over the central and western

Pacific at 35N where there are strong gradients in both SST and cloud amount. The positive SST-

low clouds feedback is likely to increase the persistence of SST anomalies from spring to summer,

however, it is difficult to envision how this relationship could be a fundamental cause of
persistence of SST anomalies over interannual and longer timescales, given the very different state

of the atmosphere-ocean system between summer and winter.

    Variability in the sensible and latent heat fluxes, which are functions of the near surface wind

speed, air temperature and humidity and the SST, dominate Qnet in winter, since the atmospheric

internal variability and mean air-sea temperature difference is much larger during the cold season.

Anomalies in Qlh and Qsh are about the same magnitude at high latitudes, while Qlh >> Qsh in the

tropics and subtropics, since warm air holds more moisture and small changes in temperature can

lead to large changes in specific humidity (the relative humidity is nearly constant at about 75-80%

over the ocean). Anomalies in Qsh and Qlh are primarily associated with wind speed anomalies in

the tropics and subtropics but are more dependent on temperature and humidity anomalies at mid to

high latitudes (Halliwell and Mayer 1996; Alexander and Scott 1997). In general, Qlw, which is a

function of SST, cloud cover and humidity, varies less than the other three components.

   The processes that influence SST anomaly development over the midlatitude oceans in winter

are organized by the large-scale atmospheric circulation. Cayan (1992b) and Iwasaka and Wallace

(1995) have used observations to show the link between the atmospheric circulation, surface fluxes

and SST; here we explore these relationships using an atmospheric general circulation model

(AGCM) coupled to a variable depth mixed layer ocean model. The dominant mode of the surface

atmospheric variability in the AGCM over the North Pacific during winter has one main center that

extends across the entire basin at around 45°N, as defined by the leading empirical orthogonal

function (EOF) of the sea level pressure (SLP) over the North Pacific (Fig. 9, top). The leading

EOF closely resembles observations (not shown) and is the surface manifestation of the

Pacific/North American (PNA) pattern. For a stronger Aleutian low, which occurs during the

negative phase of EOF 1, changes in the wind speed, air temperature and humidity create negative
Qnet anomalies that cool the ocean along ~35N that are ringed by Qnet’ of the opposite sign (Fig. 9,

middle). Since heat flux anomalies are mostly generated by the atmospheric internal variability in

midlatitudes, the dominant pattern of Qnet has a spatial scale set by the large-scale atmospheric

circulation. As a result, the pattern of correlations between the time series associated with SLP

EOF 1 and Qnet anomalies across the North Pacific (Fig. 9 middle) closely resembles the first EOF

of Qnet (Fig. 9 bottom).   Given that Qnet is the dominant term in the upper ocean heat budget in

winter, the anomalous SST field during DJF is also strongly related to the leading SLP pattern (Fig.

10). Similar results are obtained for the North Atlantic (not shown).

   Seager et al. (2000) have found that the dominant winter SST anomaly pattern over the North

Atlantic in a fixed depth ocean model, a variable depth model where h is specified as a function of

location and time of year and an ocean GCM are similar to each other and to observations when the

models are forced with by fluxes from an atmospheric mixed layer model using observed surface

winds (Fig. 11). These results provide further evidence that surface fluxes are the primary

mechanism for coupled air-sea variability in midlatitudes during winter, although Ekman transport

can contribute to the variability in regions of strong winds and SST gradients (compare Figs. 11

a&b near ~45°N in the western North Atlantic). However, mixed layer processes are important for

SST evolution in other seasons and are crucial for the winter-to-winter persistence of SST

anomalies, as discussed in the next section.

Mixed layer dynamics

   Variability in h and entrainment affect SSTs on time scales from hours to decades: entrainment

rapidly cools the mixed layer during the passage of storms (Camp and Elsberry 1978) and acted to
maintain the cool period which began in the central North Pacific during the mid 1970s (Miller et

al. 1994). Eq. 2 indicates that entrainment can impact SSTs directly through fluctuations in the

entrainment rate (we) and/or temperature jump at the base of the ML (Tm-Tb), or indirectly by

creating anomalies in h.    Using a mixed layer model, Alexander et al. (2000a) found that

w (Tm -Tb ) /h has the greatest impact on SST anomalies in summer, while w e (Tm -Tb ) /h is more

important in fall. Anomalies in h have a significant impact on the heat balance of the ML in spring

and summer. For example, if the ML shoals earlier than usual, the average net heat flux will heat

up the thinner surface layer more rapidly, creating positive SST anomalies (Elsberry and Garwood


   The mean seasonal cycle of mixed layer depth also has the potential to influence the evolution

of upper ocean thermal anomalies. Namias and Born (1970, 1974) were the first to note a tendency

for midlatitude SST anomalies to recur from one winter to the next without persisting through the

intervening summer. They speculated that temperature anomalies that form at the surface and

spread throughout the deep winter mixed layer, remain beneath the ML when it shoals in spring.

The thermal anomalies are then incorporated into the stable summer seasonal thermocline (30-100

m) where they are insulated from surface fluxes which act to damp SST'. When the mixed layer

deepens again in the following fall, the anomalies are re-entrained into the surface layer and

influence the SST.    Using subsurface temperature data from ocean weather ships and one-

dimensional mixed layer model simulations, Alexander and Deser (1995) found that this

“reemergence mechanism” (shown schematically in Fig. 12) occurred at several locations away

from strong ocean currents. Alexander et al. (1999, 2000b), Bhatt et al. (1997) and Watanabe and

Kimoto (2000) found further evidence for the reemergence of SST anomalies over the North

Pacific and North Atlantic Oceans.
    The structure of the reemergence mechanism, as indicated by the regression values between

SSTs in spring and temperature anomalies as a function of month and depth, is shown in Fig. 13 for

regions in the east, central and west North Pacific. In all three areas, springtime temperature

anomalies decay at the surface but persist in the seasonal thermocline in summer and then return to

the surface in the following fall and winter.      The timing and strength of the reemergence

mechanism differs in the three regions due to the mean seasonal cycle of h, which is much deeper

in the west compared to the east Pacific. The period when the SST anomalies are first created and

how long they persist at the surface also influence the structure of the reemerging anomalies at a

given location. For example, in the east Pacific region, SST anomalies created in February or

March extend through a relatively deep mixed layer, persist at greater depths in summer, and are

reentrained later in the year compared with those initiated in April-May (see Fig. 14 in Alexander

et al. 1999).

The Atmospheric Bridge

    In the previous sections we have examined the physics for SST anomaly formation via local

processes, now we will consider how changes in the climate system in remote locations can

influence the upper ocean in midlatitudes.

    While air-sea interactions associated with El Niño and the Southern Oscillation are centered

over the equatorial Pacific Ocean, ENSO influences the atmosphere and ocean over the entire

globe. During El Niño events, the enhanced precipitation over the anomalously warm water in the

central and eastern Pacific affects the global atmospheric circulation.        These atmospheric

teleconnections associated with ENSO (see reviews by Lau 1997 and Trenberth et al. 1997 and the
colloquium notes of Branstator and Trenberth for more information) alter the near-surface

temperature, humidity and wind, and the distribution of clouds, which in turn, can influence the

ocean far from the ENSO region (shown schematically in Fig. 14). Thus, the atmosphere can act as

a “bridge” between SST anomalies in the equatorial Pacific and ocean conditions in the North

Pacific, Atlantic, Indian and Southern Hemisphere Oceans. In addition, the developing SST

anomalies in remote locations have the potential to feedback on the atmospheric circulation.

   The “atmospheric bridge” between SST anomalies in the equatorial and the North Pacific

Ocean has been studied with an ocean OGCM by Luksch and von Storch (1990, 92) and using an

AGCM coupled to a grid of 1-D ocean models with prescribed SSTs in the tropics by Alexander

(1990, 92a), Lau and Nath (1994, 96) and Blade (1999). During El Niño events the Aleutian low

strengthens and moves slightly south during winter, as a result the net heat flux and Ekman

transport results in cooling of the central North Pacific and warming along the coast of North

America. The evolution of SST and SLP anomalies associated with ENSO are shown in Fig. 15 for

observations and Fig. 16 for an AGCM with SSTs specified in the tropical Pacific and a ML model

over the remainder of the global oceans. The ability of the model to simulate many of the observed

El Niño-La Niña SST differences indicates the importance of the atmospheric bridge for creating

global SST anomalies.

   The extent to which the developing SST anomalies in the North Pacific associated with ENSO

feedback on the atmospheric circulation is uncertain: Hendon and Hartmann (1982) and Alexander

(1992b) have found a negative feedback, i.e. the Aleutian low is not as strong in model simulations

when the evolving midlatitude SSTs are allowed to affect the atmosphere, while (Lau & Nath 1996)

have found a positive feedback. Results from recent analyses suggest the feedback could be

complex: i.e. the atmospheric response to ENSO is damped in the PNA region early in winter
(Blade 1999; Newman et al. 2000) but slightly enhanced in March (Blade personal

communication), perhaps by enhancing the persistence of the circulation anomalies. However,

since most studies have found that the response to the ENSO-induced midlatitude SST anomalies is

modest it is difficult to isolate the true feedback given the large internal atmospheric variability.

   Some of the thermal anomalies created in the North Pacific via the atmospheric bridge, return to

the surface in the following fall and winter via the reemergence mechanism. But some, especially

the cold water in the vicinity of 35°N, 165°W leave the surface layer and flow towards the equator

within the permanent pycnocline. Gu and Philander (1997) have hypothesized that these anomalies

have the potential to influence ENSO on decadal timescales (see Deser’s notes for more details).

Thus, the atmospheric bridge coupled with subduction, described in more detail below, may lead to

decadal variability.


   Subduction occurs when water exits the ML and enters the permanent pycnocline, it can then

flow far from where it was last in contact with the surface. The annual subduction rate, Sann, can be

estimated following Marshall et al. (1993):

                        0
    Sann  wek 
                    f     vdz  v
                                 h   h ,                                                         (3)

where w Ek is the Ekman pumping velocity (positive upward), f the coriolis parameter,  the

meridional derivative of f, hmax is the winter mixed layer base, and here the overbar denotes annual

means. The vertical velocity at h is less than w Ek , since some of the wind stress curl forcing is used
to drive meridional flow near the surface (the second term on the RHS of Eq. 3). The third term is

lateral induction - flow through the sloping base of the mixed layer. Subduction is intimately

related to the mean seasonal cycle of h, since it is only fluid leaving the ML when it is deep that

irreversibly enters (ventilates) the pychnocline. Thus, the temperature and salinity properties of the

main pycnocline are mainly determined by conditions in the mixed layer during winter.

   Several studies have analyzed the subduction rate over various regions of the ocean. Huang and

Qiu (1994) calculated the subduction rate for the North Pacific (see their Fig. 5) using Levitus

(1982) climatology data and Hellerman and Rosenstein wind stress data. In contrast to the results of

Marshall et al. (1993) in the North Atlantic, Huang and Qiu found that, over most of the subtropical

North Pacific, the subduction rate is just slightly larger than the Ekman pumping rate. Qiu and

Huang (1995) extended this study to examine subduction and obduction (where water from the

permanent pycnocline is irreversibly transferred into the mixed layer) in the North Pacific and the

North Atlantic. They found that subduction rates in the two oceans are comparable but obduction

is much stronger in the North Atlantic than in the North Pacific. This is related to the stronger

thermohaline circulation of the North Atlantic where Gulf Stream water is drawn into the subpolar

basin in the form of the North Atlantic Current, resulting in large heat fluxes and deep mixed

layers. The combination of the fast flowing North Atlantic Current and the gradient in winter

mixed layer depth results in strong lateral induction, which is the dominant component in the

obduction rate in the subpolar North Atlantic.

   Most studies that have examined climate variability associated subduction have analyzed the

advection of temperature anomalies created in the surface mixed layer by the mean flow within the

subtropical gyre (e.g. Deser et al. 1996, Schneider et al. 1999; Zhang and Liu 1999; also see

Deser’s notes). However, changes in the strength of the surface currents and/or gradient of h can
also affect lateral induction and thus lead to thermal and potential vorticity anomalies in the

subduction region (Inui et al. 1999; Xie et al. 2000). Both of these studies find that in an ocean

GCM forced with variable winds the location of subduction maxima can vary by as much as 10 in

longitude, which could explain much of the subsurface variability in the subduction region.

   I thank Carol Ladd and Masahiro Watanabe for the notes they provided from my lecturer at the

NCAR summer colloquium that provided the framework for this manuscript. I would also like to

thank James Scott for his assistance with some of the calculations and graphics.

Figure Captions

Fig.1. Annual average SST (top) and long term monthly mean August – March SST (bottom) in
ºC, computed using the Reynolds Reconstructed dataset on a 2ºx2º latitude-longitude grid for the
years 1950-1999.

Fig. 2. The relative magnitude of interannual anomalies compared to the mean seasonal cycle of
SST as indicated by the standard deviation of monthly SST anomalies divided by the standard
deviation of the long term monthly means from the annual mean.

Fig. 3. Mean seasonal cycle of mixed layer depth (heavy line) and the temperatures (color) in the
upper 200 m of the central North Pacific (34ºN-46ºN 160ºW-145ºW). The data is based on the
XBT analyses of Warren White for the period 1955-1990. The temperature ranges from 9ºC (blue)
to 19ºC (red), however the basic temperature structure is representative of much of the North

Fig. 4. Standard deviation of interannual SST anomalies (ºC) in March and August, which indicates
the departure of monthly values from the long-term monthly mean.

Fig. 5. (top) The interannual standard deviation of the anomalous heat flux due to meridional
Ekman advection of the mean SST gradient, vek  yTm , for January. The field has been multiplied by
   cph to give units of W m-2. (bottom) Ratio of the standard deviation of vek  yTm to the standard
deviation of the total anomalous Ekman heat transport, (vek  yTm ) . vek  yTm is associated with
anomalies in the zonal wind stress.
Fig. 6. (top) The January mean net surface heat flux (Qnet), (middle) Ekman heat transport
(Qek= oc p h{v Tm } ) and (bottom) entrainment heat flux through the base of the mixed layer
(Qwe= ocp hwe{Tb  Tm}), where the latter two have been multiplied by   cph to give units of W m-
  . The values are obtained from a 50-year integration of an atmospheric GCM coupled to a grid of
independent mixed layer models over the global oceans.

Fig. 7. As in Fig. 6 but for the interannual standard deviation in January for Qnet (top), Qek (middle)
and Qwe (bottom) in W m-2.

Fig. 8. As in Fig. 7 but for the interannual standard deviation of Qnet, Qek, and Qwe in August in W

Fig. 9. EOF 1 of SLP (top) and Qnet (bottom) over the North Pacific during December-February,
units are non-dimensional where the sum of the grid values squared is 1. The SLP and Qnet EOFs
explain 50% and 34% of the variance, respectively. The middle panel shows the correlation
between the leading SLP principal component (PC 1 - the time series of the amplitude and polarity
of EOF 1) with Qnet at individual grid squares. The SLP and Qnet fields are obtained from a 50-year
integration in which an atmospheric GCM is coupled to a grid of mixed layer ocean models. The
ocean model predicts SST and h based on vertical but not horizontal processes.

Fig 10. The correlation between PC 1 of SLP with SST anomalies at individual grid points (top)
and the EOF 1 of SST (bottom), which explains 39% of the total variance. The fields are from the
same coupled integration used in Fig. 9.

Fig. 11. Singular value decomposition (SVD) between surface winds from the NCEP reanalysis
with SSTs for the January-March season. The results are computed using (a) a variable depth
mixed layer ocean model and (b) a full ocean GCM. The SVD time series for the two models and
from observations are shown in (c); note that the time variations of the models are very similar to
observations even on decadal time scales. The SVD between winds and SST from observations
and a 75 m slab ocean (see Fig. 6 in Seager et al. 2000) looks very similar to the results shown here.

Fig. 12. Schematic depiction of the reemergence mechanism: mixed layer temperature anomalies
created by surface fluxes in winter, remain below the mixed layer when it shoals in summer (1st red
arrow) and then entrainment (blue arrow) brings the thermal anomalies back into the mixed layer
(2nd read arrow) when it deepens again in fall. Air-sea heat exchange damps the SST anomalies
over the thin mixed layer in summer.

Fig. 13. Monthly SST anomalies from January through April of the following year regressed on
SST anomalies averaged over the first February-March-April (FMA) period in the eastern (26N°-
42°N, 132°W-116°W), central (26N°-42°N, 164°W-148°W), and western (38°N-42°N, 160°E-
180°) Pacific. Values are computed from the NCEP ocean assimilation system for the years 1980-
1998. The contour interval is 0.1°C per 1°C in FMA, values greater than 0.7 are shaded red.

Fig. 14. Schematic of the “atmospheric bridge” shown as a meridional cross-section in the central
Pacific. The atmosphere can act like a bridge between parts of the ocean since changes in the
atmospheric circulation associated with ENSO force SST anomalies in regions far from the tropical
Pacific through changes in cloudiness, winds, etc. These SST anomalies then have the potential to
feed back upon the atmospheric circulation. In the North and South Pacific SST anomalies
generated by the bridge mechanism, can also enter the permanent pycnocline via subduction and
then flow towards the equator where they have the potential to influence tropical SSTs on decadal
time scales (e.g. Gu and Philander 1997).

Fig. 15. Maps of the difference in SST (°C, color shading) and SLP (mb, contours) between the
composite average of 9 El Niño (warm) and 9 La Niña (cold) events for Sep-Oct-Nov, Dec-Jan-
Feb, Mar-Apr-Mar, and Jun-Jul-Aug. On average, the events peak in December. The results are
based on NCEP reanalysis for the period 1950-1999.

Fig 16. As in Fig 15 but where the SLP and SST composite differences are based on ensemble
average of sixteen 50-year AGCM simulations where observed SSTs are specified between 15°N-
15°S and east of 172°E in the Pacific for the period 1950-1999 and a grid of predictive mixed layer
models over the remainder of the world’s oceans.


Alexander, M. A. 1990: Simulation of the response of the North Pacific Ocean to the anomalous
   atmospheric circulation associated with El Niño. Climate Dyn., 5, 53-65.

Alexander, M. A., 1992a: Midlatitude atmosphere ocean interaction during El Niño. Part 1. The
   North Pacific Ocean. J. Climate, 9, 944-958.

Alexander, M. A., 1992b: Midlatitude atmosphere ocean interaction during El Niño. Part 2. The
  Northern Hemisphere atmosphere. J. Climate, 9, 959-972.

Alexander, M. A., and C. Deser, 1995: A mechanism for the recurrence of wintertime midlatitude
   SST anomalies. J. Phys. Oceanogr., 25, 122-137.

Alexander, M. A. and J. D. Scott 1997: Surface flux variability over the North Pacific and North
   Atlantic Oceans. J. Climate, 10, 2963-2978.

Alexander, M. A., C. Deser, and M. S. Timlin 1999: The reemergence of SST anomalies in the
   North Pacific Ocean. J. Climate, 12, 2419-2433.

Alexander, M. A., J. D. Scott, and C. Deser, 2000a: Processes that influence sea surface
   temperature and mixed layer depth variability in a coupled model. J. Geophys. Res., 105,

Alexander, M. A., J. D. Scott, and C. Deser, 2000b: Winter-to-winter recurrence of sea surface
   temperature salinity, and mixed layer depth anomalies. Progress in Oceanogr., submitted,
   [available from:].
Battisti, D. S., U. S. Bhatt and M. A. Alexander 1995: A modeling study of the interannual
   variability in the wintertime North Atlantic Ocean. J. Climate, 8, 3067-3083.

Blade, I. 1999: The influence of midlatitude ocean-atmosphere coupling on the low-frequency
   variability of a GCM. Part II: Interannual variability induced by tropical SST forcing. J.
   Climate, 12, 21-45.

Bjerknes, J. 1964: Atlantic air-sea interaction. Adv. in Geophys, 20,1-82.

Camp, N. T. and R. L. Elsberry 1978: Oceanic thermal response to strong atmospheric forcing, II,
   the role of one-dimensional processes. J. Phys. Oceanogr., 8, 215-224.

Cayan, D. R., 1992a: Latent and sensible heat flux anomalies over the northern oceans: Driving the
   sea surface temperature. J. Phys. Oceanogr., 22, 859-881.

Cayan, D. R., 1992b: Latent and sensible heat flux anomalies over the northern oceans: The
   connection to monthly atmospheric circulation. J. Climate, 5, 354-369.

Delworth, T. 1996: North Atlantic interannual variability in a coupled ocean-atmosphere model. J.
   Climate, 9, 2356-2375.

Deser, C., M. A. Alexander and M. S. Timlin 1996: Upper ocean thermal variations in the North
   Pacific during 1970 - 1991. J. Climate, 9, 1841-1855.

Elsberry, R. L. and R. W. Garwood 1980: Numerical ocean prediction models-goal for the 1980s.
   Bull. Amer. Meteor. Soc, 61, 1556-1566.

Frankignoul, C., 1985: Sea surface temperature anomalies, planetary waves, and air-sea feedback in
   the middle latitudes. Rev. Geophys., 23, 357-390.

Gaspar, P., 1988: Modeling the seasonal cycle of the upper ocean. J.Phys.Oceanogr., 18, 161-180.

Gill, A. E. and P. P. Niller, 1973: The theory of the seasonal variability in the ocean. Deep Sea
    Res., 20, 141-177.

Gu, D. and S. G. H. Philander 1997: Interdecadal climate fluctuations that depend on exchanges
   between the tropics and extratropics. Science, 275, 805-807.

Halliwell, G. R. and D. A. Mayer 1996: Frequency response properties of forced climatic SST
   anomaly variability in the North Atlantic. J. Climate, 9, 3575-3587.

Hendon, H. H. and D. L. Hartmann 1982: Stationary waves on a sphere: sensitivity to thermal
   feedback. J. Atmos. Sci., 39,1906-1920.

Huang, R. X., and B. Qiu, 1994: Three-dimensional structure of the wind driven circulation in the
   subtropical North Pacific. J.Phys.Oceanogr., 24, 1608-1622.
Inui, T., K. Takeuchi, and K. Hanawa, 1999: A numerical investigation of the subduction process in
    response to an abrupt intensification of the westerlies. J.Phys.Oceanogr., 29, 1993-2015.

Klein, S. A., D. L. Hartmann and J. R. Norris 1995: On the relationships among low-cloud
   structure, sea surface temperature and atmospheric circulation in the summertime northeast
   Pacific. J. Climate, 8, 1140-1155.

Lau, N.-C., and M. J. Nath, 1994: A modeling study of the relative roles of tropical and
   extratropical SST anomalies in the variability of the global atmosphere-ocean system.
   J.Climate, 7, 1184-1207.

Lau, N.-C., and M. J. Nath, 1996: The role of the "atmospheric bridge" in linking tropical Pacific
   ENSO events to extratropical SST anomalies. J.Climate, 9, 2036-2057.

Lau, N.-C., 1997: Interactions between global SST anomalies and the midlatitude atmospheric
   circulation. Bull.Amer.Met.Soc., 78, 21-33.

Luksch, U., H. v. Storch and E. Maier-Reimer 1990: Modeling North Pacific SST anomalies as a
   response to anomalous atmospheric forcing. J. Mar. Syst., 1, 51-60.

Luksch, U., and von Storch, 1992: Modeling the low-frequency sea surface temperature variability
   in the North Pacific. J.Climate, 5, 893-906.

Levitus, S., 1982: Climatological Atlas of the World Ocean, NOAA Prof. Paper 13, U.S.
   Government Printing Office, Washington D.C., 173 pp.

Marshall, J. C. A. J. C. Nurser, and R. G. Williams, 1993: Inferring the subduction rate and period
   over the North Atlantic. J.Phys.Oceanogr., 23, 1315-1329.

Miller, A. J., D. R. Cayan, T. P. Barnett, N. E. Graham and J. M. Oberhuber 1994: Interdecadal
   variability of the Pacific Ocean: model response to observed heat flux and wind stress
   anomalies. Climate Dyn., 10, 287-302.

Monterey, G. I. and S. Levitus 1997: Climatological Cycle of Mixed Layer Depth in the World
  Ocean. U.S. Gov. Printing Office, NOAA NESDIS,5 pp., 87 pp.

Newman, M., M. A. Alexander, C. R. Winkler, J. D. Scott, and J. J. Barsugli, 2000: A linear
  diagnosis of the coupled extratropical ocean-atmosphere system in the GFDL GCM.
  Atmospheric Sciences Letters, 1.

Namias, J. 1959: Recent seasonal interactions between North Pacific waters and the overlying
  atmospheric circulation. J. Geophys. Res., 64, 631-646.

Namias, J. 1963: Large-scale air-sea interactions over the North Pacific from summer (1962)
  through the subsequent winter. J. Geophys. Res., 68, 6171-6186.
Namais, J. 1965: Macroscopic association between mean monthly sea surface temperature and the
  overlying winds. J. Geophys. Res., 70, 2307-2318.

Namias, J. 1969: Seasonal interactions between the North Pacific Ocean and the atmosphere during
  the 1960's. Mon. Wea. Rev., 97,173-192.

Namias, J. and R. M. Born, 1970: Temporal coherence in North Pacific sea-surface temperature
  patterns. J.Geophys.Res., 75, 5952-5955.

Namias, J. and R.M. Born, 1974: Further studies of temporal coherence in North Pacific sea surface
  temperatures. J.Geophys.Res., 79, 797-798.

Niller, P. P. and E. B. Kraus, 1977: One dimension models of the upper ocean. Modeling and
    Prediction of Upper Layers of the Ocean, ed., E.B. Kraus, Pergamon Press, 143-172.

Norris, J. R. and C. Leovy 1994: Interannual variability in stratiform cloudiness and sea surface
   temperature. J. Climate, 7, 1915-1925.

Norris, J. R., Y. Zhang and J. M. Wallace, 1998: Role of clouds in summertime atmosphere-ocean
   interactions over the North Pacific. J.Climate, 11, 2482-2490.

Qiu, B. and R. X. Huang, 1995: Ventilation of the North Atlantic and North Pacific: Subduction
   versus obduction. J.Phys.Oceanogr., 25, 2374-2390.

Schneider, N., A. J. Miller, M. A. Alexander and C. Deser 1999: Subduction of decadal north
   Pacific temperature anomalies: observations and dynamics. J. Phys. Oceanogr., 29, `056-1070.

Seager, R. Y., Y. Kushnir, M. Visveck, N. Naik, J. Miller, G. Krahmann and H. Cullen, 2000:
   Causes of Atlantic Ocean climate variability between 1958 and 1988. J.Climate, in press.

Trenberth, K. E., G. W. Branstator, D. Karoly, A. Kumar, N-C. Lau and C. Ropelewski 1997:
   Progress during TOGA in understanding and modeling global teleconnections associated with
   tropical sea surface temperatures. J. Geophys. Res., 103, 14,291-14324.

Weare, B. 1994: Interrelationships between cloud properties and SSTs on seasonal and interannual
  timescales. J. Climate, 7, 248-260.

Xie, S.-P., T. Kunitani, A. Kubokawa, M. Nonaka, and S. Hosoda, 2000: Interdecadal thermocline
   variability in the North Pacific for 1958-1997: A GCM simulation. J.Phys.Oceanogr., in press.

Zhang, R.-H. and Z. Liu 1999: Decadal thermocline variability in the North Pacific Ocean: two
   pathways around the subtropical gyre. J. Climate, 12, 3273-3296.

Shared By: