Some ways ocean circulation affects climate:

• biological productivity gas exchange• direct transport of gases (e.g., the carbon cycle)• ice albedo• heat exchange with atmosphere

“Heat Budget”

of ocean tells us• how ocean stores and moves heat • how ocean exchanges heat with atmosphere

Several possibilities: including1.

Q = 0 (thermally passive ocean)2.

D –

H small, Q = ∫Et

dV

(ocean is thermal flywheel)3.

D –

∫Et

dV

small, Q = H (ocean moves heat around)

1 good approx global and annual average2 good approx some places for timescales ≤ seasonal3 good approx annual average

What determines Q(x,y)?Need to consider both atmosphere dynamics andglobal oceanic velocity

and temperature field.

Similar expressions exist for transport of salt and other tracers.

100120

140160

180200

220240

260280

20

30

40

50

60

−6000

−4000

−2000

0

When calculating H, we often choose domain so that most of the lateralboundaries are “walls”, except for a 1 or 2 zonal sections (as shown). u●n

= 0 on bottom, and if we can neglect u●n

through the top, H is simply meridional heat transport.

Ocean and Atmospheric Meridional Heat Transport

Ocean clearly important in tropicsLatest observations smaller ocean role at high latitudes(still significant uncertainties)

Trenbirth

and Caron, 2001: J Clim, 3433-3443

oceanatm

osph

ere

tota

l1 PW = 1015

W

Trenberth

& Caron (2001)

Rules for Density1)

Density hardly changesρ

≈

1000 kg/m3

= 1 gm/cm3

everywhere in ocean2) Density increases with S:

At T=0 and p=0, ρ(S=0) = 999.8, ρ(S=35)=1028.1

3)

Density increases with p:At S = 35 psu, T = 0 C, p = 4000 dbar, ρ

= 1046.4

4)

Density decreases with T5)

Rate of change w.r.t. T depends on T,p

|dρ/dT| smaller if water is colder

or deeper6)

In ocean, variations due to p are greatest, butvariations due to T & S more important:

u

responds to horizontal

ρ

variations (indep

of p)

Equation of State

Accurate formula for ρ(T,S,p) can be found in Gill (1982)Rough approximation:

ρ

= 999.83 + C(p) + β(p)S

–

α(T,p)T

–

γ(T,p)(35-S)TWhere

C ≈

5.05(p/1000 dbar) kg/m3

β

≈

.8 kg m-3

psu-1

α

≈

.071[1+.351(p/1000 dbar) + (.068 C-1)T] kg/m3

γ

≈

.003 kg m-3

psu-1

C-1

0 5 10 15 20 25 30−8

−6

−4

−2

0

2

in situ temperature (C)

dens

ity (

kg/m

3 )

ρ(p,T,S) − ρ(p,0,S), Standard (solid) and Approx (dashed)

p = 0 dbar, S = 0 psup = 4000 dbar, S = 35 psu

p = 0 dbar, S = 35 psu

0 10 20 30 40−30

−25

−20

−15

−10

−5

0

5

salinity (psu)

dens

ity (

kg/m

3 )

ρ(p,S,T)−ρ(p,35,T)

p=0,T=0

p=0,T=30

p=4000,T=0

0 10 20−5

−4

−3

−2

−1

0

density (kg/m3)

pres

sure

(10

3 dba

r)

ρ(p,S,T)−ρ(0,S,T)

T=0,S=0T=0,S=35

T=30,S=35

30 32 34 36 38 40

0

5

10

15

20

25

30

salinity (psu)

pote

ntia

l tem

pera

ture

(C

)

σθ(θ,S)

2022

24

26

28

30

30 32 34 36 38 40

0

5

10

15

20

25

30

salinity (psu)

pote

ntia

l tem

pera

ture

(C

)

σ4(θ,S)

36

38

40

4244

46

48

Potential Density and Some Nomenclature

If water has given θ, S, and p, its potential density

= ρ(T=θ,S,p=0).

This is a way to filter out uninteresting dependence on p.

Because density is always about 1000, useful to define“sigma-t”: σt

= ρ(T,S,0) –

1000“sigma-theta”: σθ

= ρ(θ,S,0) –

1000Can define with reference to other depths, such as

σ1 = ρ(θ,S,1000 dbar) –

1000

Net Annual-Average Heat Flux into Ocean

General pattern:absorbs heat near equator, loses heat at high latitudes

(as expected)Sea’s heat transport comparable to atmosphere:

Significant role in moderating meridional T gradientIndicates importance of ocean circulation.

Typical values O(50 W/m2)Some interesting details:

heat gain largely in east, heat loss in westPacific heat gain especially bigAtlantic heat loss especially big

Annual Avg Net Heat Flux

−150 −100 −50 0 50 100 150

Evaporation/Precipitation Patterns

Evap. minus Precip

(and runoff) drives S variationsNo (significant) salt fluxes through surface,But freshwater fluxes can increase/decrease S dilution.General pattern:

Net precip

in tropics and high latitudesNet evap

in subtropics

Typical values Order(1 m/yr)

from COADS, cm/yr

Annual Avg Precipitation

0 50 100 150 200 250 300 350

from COADS, cm/yr

Annual Avg −Evaporation

−200 −150 −100 −50 0

from COADS, cm/yr

Annual Avg P−E

−150 −50 50 150 250

Surface Property Distributions

Temperature: general pattern as would be expectedwarm in tropics, cold at high latitudesup to about 6 C winter-summer differenceSST lags land temperature

Salinity: general pattern follows forcinghigh values in subtropical evaporation regionslow values in tropics and high latitudessome regional extreme values

Density: dominated by T variations2-D patterns: zonal and inter-basin differences

indications of effects of ocean currents?aT and S data from Levitus

et al. (1994), World Ocean Atlas

Mar SST

0 4 8 12 16 20 24 28 32

Sep SST

0 4 8 12 16 20 24 28 32

cold tongue

Mar SSS

28 32 33 34 35 36 37 38 42

Mar SS Density

20 21 22 23 24 25 26 27 28

Sep SS Density

20 21 22 23 24 25 26 27 28

High-Resolution SST SnapshotShows More Structure

Pacific Atlantic

Meridional-Depth Profiles

Warm tropical water is just a thin layer at top of oceanmost of ocean is < 5 Cwarm layer has “double bowl”

shape

structure very similar in Atlantic and PacificSalinity has more complicated structure

subsurface “tongues”

of fresh watercreate minima and maxima in S(z) in placesindications of deep flow?

T and S data from Levitus

et al. (1994), World Ocean Atlas (meridional sections)And World Ocean Atlas (2001) (global sea surface properties)

Surface Mixed Layer

Top of ocean relatively homogenous: “mixed layer”

Mixed layer grows thicker via:• wind stirring• convection due to cooling/salinification

Mixed layer grows thinner via:• solar heating• desalinification

(precipitation)

Hence ML depth varies greatly in space/time10 –

1000 m depth

Shallower at low latitudes, daytime, summerDeeper at high latitudes, night, winter

Near-global ocean model,

Hirst

et al. (1996, JPO)

NADW

AABW

CDW

Oceanic Overturning Circulation(s)

Sub-Tropical Cells

North Atlantic Deep Water

Antarctic Bottom Water

Circumpolar Deep Water

STC

Tomczak

and Godfrey, Regional Oceanography, ch

15, after Bainbridge (1980)

Pot.Temp.

Salinity

Oxygen

How do we know the DMOC exists?

Tracers hint at deepflows in Atlantic:

Relatively saltyNorth Atlantic Deep Water(NADW) moves southward at2 –

4 km depth

Relatively fresh Antarctic Bottom Water (AABW)moves northward near bottom

Relatively fresh Antarctic Intermediate Water(AAIW) moves northward atabout 1 km depth.

Smethie

et al. (2000, JGR)

More tracerEvidence(freons

fromAtmosphereSince 1940s)

Pickart

and Smethie

(1993, JPO)

Geostrophic velocity shows DWBC in N Atl(from density measurements + “Pogo”

floats)

Section 3

Schmitz (1996), cited in Ocean Circulation and Climate

Schematic of Global DeepMeridional Overturning Circulations

NADW FORMATION RATE13 Sv

(Rintoul

1991)15 Sv

(Ganachaud

Wunsch

2000)20 Sv

(Talley 2003)

Wyrtki

and Kilonsky, 1984 0 10 20-10latitude

dept

h (m

)

Equator has important vertical structure:

Surface flow: westwardSubsurface Flow: eastward

“Equatorial Undercurrent”Flow upward along thermocline

(see previous slide)Important part of Subtropical Cellcolder subtropical water flows to equator flows E’wardand upwells returns poleward as Ekman transport

El Nino makes Cold Tongue go away

Wind-Driven Gyres

Consider water

in basin forced by wind stress at surface:

Closed basin: watermust form GYRES.

Direction based onwind shear

Can get water flowing inopposite directionto wind

top view

Ocean Gyres (clockwise and counter-clockwise) follow the Wind Curl

Surface circulation and depth-averaged circulation shows gyres whichare generally aligned with the regions of positive and negative wind curl(see observations on following slides).

Reid (1997, Prog

Oceanogr)

Kuroshio

E. Australian Cur.

NoteStrongWesternBoundaryCurrents

Observed Pacific Surface Geostrophic Flow

subpolar

gyre

subtropical gyre

Peru

-Chi

le C

ur

subtropicalgyre

Antarctic Circumpolar Current

N. Eq. Counter Cur. S. Eq. Cur.

N. Eq. Cur.

Oyashio

−100 −80 −60 −40 −20 0 20

0

20

40

60

Smoothed Annual Mean Surface Velocity from Drifters

Arrow length proportional to (speed)1/2

based on Hansen and Poulain

(1996)

subpolar

gyre

subtropical gyre

Gulf Stream

N. Eq. Cur.

Eq. Counter Cur.

Braz

il Cu

r.

Falkland/Malvinas Cur.

BenguelaCur.

SEC

Reid (1989, Prog

Oceanogr)

subtropical gyre

Antarctic Circumpolar Current

Observed Surface South Atlantic Geostrophic Flow

Agul

has

Cur

subtropical gyre

Antarctic Circumpolar Current

Observed Surface Indian Ocean Geostrophic Flow

Eq. Counter Cur.

S. Eq. Cur.

How much heat transport is associated with Subtropical Cells?

latitude

dept

h (m

)

Pacific T at 210 E

−45 −30 −15 0 15 30 45−500

−400

−300

−200

−100

0

φ

−20

0

20