Ocean-Atmosphere Interactions & Modelling: A General Overview 8 - - PowerPoint PPT Presentation

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Ocean-Atmosphere Interactions & Modelling: A General Overview

8th ICTP Workshop on the Theory and Use of Regional Climate Models May 27, 2016 RICCARDO FARNETI rfarneti@ictp.it

(ESP/ICTP)

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Outline

1

Motivation for using coupled models Ocean-atmosphere modelling

2

Posing the coupled model problem Foundations Resolving versus parameterizing: some numbers

3

Basics on (Low-Frequency) Variability A few examples

4

Mesoscale/regional examples Or how regional simulations can help overcome some issues

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Outline

1

Motivation for using coupled models

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

General motivational comments and challenges

Climate model fundamentals and the use of climate models as a tool for science involves some of the most difficult problems in classical and computational physics.

turbulence closures and subgrid scale parameterizations analysis and rationalization of massive datasets efficient methods for discretizing continuous media.

We are also touching on elements of the most important environmental and societal problem facing the planet.

Climate warming is happening and humans are the key reason. The ocean’s role in the earth climate is significant. Providing rational and robust models for understanding and predicting climate is a central element of climate science.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

General motivational comments and challenges

Climate model fundamentals and the use of climate models as a tool for science involves some of the most difficult problems in classical and computational physics.

turbulence closures and subgrid scale parameterizations analysis and rationalization of massive datasets efficient methods for discretizing continuous media.

We are also touching on elements of the most important environmental and societal problem facing the planet.

Climate warming is happening and humans are the key reason. The ocean’s role in the earth climate is significant. Providing rational and robust models for understanding and predicting climate is a central element of climate science.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Types of climate models

There are many types of ocean models...

conceptual or process models

integration time number of processes detail of description

Earth Models of Intermediate Complexity (EMICs) Global Climate Models or General Circulation Models (GCMs)

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Hierarchical approach

Hierarchical Ocean-Atmosphere Modelling A hierarchy of models and simulations to understand and simulate the physics and dynamical mechanisms of climate

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Hierarchical approach

Background

• @: I. Held (Science, 2014)

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Decadal Variability/Predictability lies in the Oceans

Internally generated potential predictability (From Boer et al, 2011) From Hawkins and Sutton, 2009

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Space-time diagram of motions

Broad range of space-time scales We see the absence of a clear spectral gap except for scales larger than 1000 km. We can use EMICs or Downscale to get information on smaller space-time scales.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Turbulent cascade of mechanical energy

Compliments of Baylor Fox-Kemper, Brown University, USA

3d turbulence: energy cascade to small scales 2d/QG turbulence: energy cascade to large scales (inverse cascade) Cascades act to couple space-time scales.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Outline

2

Posing the coupled model problem

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Theoretical foundations for ocean-atmosphere models

Continuum thermo-hydrodynamical equations

Seawater mass conservation Tracer mass conservation Momentum conservation Linear irreversible thermodynamics of seawater Typically assume hydrostatic balance

Boundary conditions

Air-sea interactions Sea ice-ocean interactions Ice shelf-ocean interactions Solid-earth-ocean interactions

Subgrid scale parameterizations

Momentum closure: frictional stress tensor Tracer closure: transport tensor Boundary layer parameterizations

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

A zoo of physical processes

The

• cean-atmosphere

interface contains a zoo of physical processes! Strong coupling between processes ⇔ no spectral gap. Coupling means it is generally better to resolve than parameterize. Yet we cannot resolve everything

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

A zoo of physical processes in the Ocean interior

What happens in the interior will affect the surface interacting with the atmosphere. ... The Ocean is not an SST ...

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Equilibration time scale problem

Scaling argument for deep adjustment time H2/κ = (2000 m)2/(2 × 10−5 m2/s) (1) = O(5000 years) (2) Bottom line for global climate: Performing long (climate scale) simulations at eddy-resolving / permitting resolution are not practical Must live with deep ocean not being at equilibrium in most simulations

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Upper ocean boundary and wave interactions

• FIG. 1. A schematic view of the influence of waves on air–sea exchanges.
• From Cavaleri et al (2012)

New research activities in boundary layer param prompted by refined atmos and

• cean resolutions

that admit new dynamical regimes (e.g., mesoscale eddies, tropical cyclones). An increased awareness in the climate community of the importance of surface ocean gravity

• waves. See also

Ufuk’s talk in a few minutes.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

The marginal ice zone (MIZ)

From ONR Marginal Ice Zonal Project

Questions about processes at the marginal ice zone are of prime importance as Arctic sea ice melts.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

The marginal ice zone (MIZ)

• Fig. 1: Schematic of the Southern Ocean circulation, showing the numerous relevant physical and

biogeochemical processes [courtesy of L. Talley].

Southern Ocean upwelling:

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Resolving versus parameterizing: some numbers

Setting the model’s grid scale to the Kolmogorov length ∆ = 10−3m over a global (ocean) domain of volume 1.3 × 1018 m3 requires 1.3 × 1027 discrete grid cells. This is roughly 104 × Avogadro’s number! Each model grid point has a velocity vector and tracer fields to time integrate. Conclude:

We will be dust long before DNS of global climate simulations. We must use parameterizations to simulate, or regional simulations.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Spatial scale of mesocale and submesoscale eddies

MODIS satellite w/ inserts by A. Adcroft (GFDL)

Eddy size ∝ first baroclinic Rossby Radius λm = cm/|f|, where the phase speed is approximated by (Chelton et

• al. 1998)

cm ≈ 1 m π

−H

N dz. Global models are marginal at representing this scale; regional and process models can help reach into the submesoscale.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Spatial scale of mesocale and submesoscale eddies

−80 −70 −60 −50 −40 −30 −20 −10 10 20 10 20 30 40 50 60 70 80 90 100 [Km] Latitude CM2.4: Rossby Radius vs Grid Spacing 0.25o 1st Rossby Radius

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Resolution required to represent mesoscale eddies

From Hallberg (2013)

Hallberg (2013): 2∆ ≤ λ1 needed to resolve mesoscale eddies. Map indicates the necessary Mercator spacing for 2∆ = λ1.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Ocean resolution in IPCC-class climate models

The ocean is but one component amongst many within climate system models. Resolution refinement is painfully slow! This diagram is useful to target one’s career choices.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Nevertheless, progress is exciting!

Daily SST from the GFDL CM2.6, a 0.1◦ configuration for the ocean component, under a 50 km global atmosphere model But with a big limitation...

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Outline

3

Basics on (Low-Frequency) Variability

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Possible Mechanisms and sources of variability

Climate variability might arise primarily from the atmosphere, independent of varying boundary conditions such as SST. Climate variability might be enhanced by the presence of an

• cean with a large heat capacity, leading to a red spectrum. The

null hypothesis for climate variability. Climate variability might arise via coupled ocean-atmosphere modes (e.g. ENSO). Controversial in mid-latitudes. Climate variability might have primarily an oceanic origin. Ocean variability might affect the atmosphere without the need for coupled modes.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Possible Mechanisms and sources of variability

Climate variability might arise primarily from the atmosphere, independent of varying boundary conditions such as SST. Climate variability might be enhanced by the presence of an

• cean with a large heat capacity, leading to a red spectrum. The

null hypothesis for climate variability. Climate variability might arise via coupled ocean-atmosphere modes (e.g. ENSO). Controversial in mid-latitudes. Climate variability might have primarily an oceanic origin. Ocean variability might affect the atmosphere without the need for coupled modes.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Possible Mechanisms and sources of variability

Climate variability might arise primarily from the atmosphere, independent of varying boundary conditions such as SST. Climate variability might be enhanced by the presence of an

• cean with a large heat capacity, leading to a red spectrum. The

null hypothesis for climate variability. Climate variability might arise via coupled ocean-atmosphere modes (e.g. ENSO). Controversial in mid-latitudes. Climate variability might have primarily an oceanic origin. Ocean variability might affect the atmosphere without the need for coupled modes.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Possible Mechanisms and sources of variability

Climate variability might arise primarily from the atmosphere, independent of varying boundary conditions such as SST. Climate variability might be enhanced by the presence of an

• cean with a large heat capacity, leading to a red spectrum. The

null hypothesis for climate variability. Climate variability might arise via coupled ocean-atmosphere modes (e.g. ENSO). Controversial in mid-latitudes. Climate variability might have primarily an oceanic origin. Ocean variability might affect the atmosphere without the need for coupled modes.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Hasselmann (1977)’s Stochastic Climate Model

The ocean mixed layer (the slow component), of much higher heat capacity, integrates atmospheric white noise (the fast component), giving rise to a red spectrum. ∂tT′ = −λ T′ + F(t) The variance spectrum is |T′(ω)|2 =

|F′|2 ω2+λ2

and so the slope is ω−2

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Barsugli and Battisti (1977)’s Stochastic Climate Model

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Does this work?

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Does this work? YES!

10

−3

10

−2

10

−1

5 10 15 20 25

Atmospheric Surface Temperature Spectra PSD Frequency [1/days]

Uncoupled Coupled

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Basic effects of ocean-atmosphere thermal coupling

increases variance in both media. decrease energy fluxes between them. prescribing mid-latitude SSTs does not lead to a correct simulation of low-frequency thermal variance in the atmosphere. We need a coupled ocean-atmosphere model ...

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Is this ‘all there is’? ...

Is the integration of atmospheric variability by the oceanic mixed layer producing a red spectrum all there is? dynamical process can indeed produce variance at long periods

10

!

10 10

!5

10

!

10

!

10

!2

10

!

variance (K

HADISST

10

!

10 10

!5

10

!

10

!3

10

!

10

!

KAPLAN

SST spectra AR!1 spectra AR!1 conf!95% SST spectra AR!1 spectra AR!1 conf!95%

frequency (1/years) frequency (1/years)

Figure 6. Mean spectra of midlatitude SST anomalies of the HADISST and Kaplan

SST data sets (thick lines), along with the best fit spectra from an AR(1) process (thin central line) with 95% confidence levels (thin outer lines). Adapted from Dommenget & Latif (2002).

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

We can add spatial coherence in atmosphere and a dynamical ocean: Regional Basin Modes/Oscillations

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Adding spatial coherence in atmosphere and a dynamical ocean: Regional Basin Modes/Oscillations

The Pacific Decadal Oscillation PDO North Pacific SST integrates weather noise SST anomalies provide reduced damping of atmospheric signals at low-frequency local and remote coupled feedbacks

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Adding spatial coherence in atmosphere and a dynamical ocean: Regional Basin Modes/Oscillations

The Atlantic Multidecadal Oscillation AMO AMOC variability forces AMO signal (most probably) AMO forces atmospheric response, e.g. negative NAO (maybe) trans-basin connections

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Adding spatial coherence in atmosphere and a dynamical ocean: Regional Basin Modes/Oscillations

The Indian Ocean Dipole IOD

• cean-atmosphere interaction causing interannual climate

variability Oscillations of SSTs due to variability in trade winds Tropical → shorter time scale (interannual)

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Outline

4

Mesoscale/regional examples

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Net Surface Heat Flux

Blue → Heat into the Ocean

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

SST bias in Coupled Models

“Models still show significant errors ... The ultimate ! source of most is that many important small-scale ! processes are not represented explicitly in models …”! ! Randal et al., 2007 .!

Too Warm! Too Cold!

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Two-way nesting in the Agulhas region

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

The Benguela Upwelling problem

Of all the major coastal upwelling systems in the World’s ocean, the Benguela, located off south-west Africa, is the one which climate models find hardest to simulate well. Increasing both oceanic and atmospheric resolutions (and shifting winds towards the coast) improves the simulation.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

The double-ITCZ problem

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Air-Sea interaction at basin (slow and large) scales

Stronger wind speed → lower SST via mixing and turbulent flux Negative Correlation → Atmosphere drives the Ocean

Kushnir et al. 2002 SST and wind anomaly patten related to NAO

rtime (Dec–Mar), anomalous SST, ocean–atmosphere turbulent heat flux (

Mean wind is westerly ➜ Mean wind is easterly ←

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Air-Sea interaction at mesoscales (fast and short)

Enhanced (Reduced) wind speed over warm (cold) SST Positive Correlation → Ocean drives the Atmosphere

QSCAT WIND STRESS TRMM SST TRMM SST and QuikSCAT wind stress on 3 September 1999

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Effects on the Atmosphere

Enhanced (Reduced) wind speed over warm (cold) SST Positive Correlation → Ocean drives the Atmosphere case from the observations

1 1.5 2 2.5 3 3.5

mm d–1

4 4.5 5 5.5 6 40° W 80° W 70° W 60° W 50° W 40° W 80° W

a

Wind convergence, satellite (10−6 s−1)

50° N 45° N 40° N 35° N 30° N 25° N 80° W –8 –6 –4 –2 2 4 6 8 70° W 60° W 50° W 40° W 5 4 4 3 3 2

a

Observed rain rate, satellite

50° N 45° N 40° N 35° N 30° N 25° N 5 4 4 3 3 2 80° W 70° W 60° W 50° W 40° W

∙u Satellites rain rate: ERA-I

Upward wind (10−2 Pa s−1) Pressure (hPa)

200 300 400 500 600 700 800 900 1,000 32° N 34° N 36° N –2 –1.5 –1 –0.5 0.5 1 1.5 2 2.5 3 38° N 40° N 42° N

a

50° N 45° N 40° N 35° N 30° N 25° N

upward wind convergence divergence

Minobe et al. 2008

tropopause! westerly

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

How does it work?

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Discretizing a column of ocean fluid

z = η Qpbl, τ surface, Qm

surface fluxes

J(x) J(x) J(s) J(s)

fluxes crossing grid cell faces

xi xi+1 sk+1 sk

penetrative shortwave

z = −H

bottom fluxes

Qbottom,τ bottom

From Griffies and Treguier (2013)

Boundary fluxes through surface and bottom. Transport convergence (advective and subgrid scale), body forces (gravity, Coriolis), contact forces (pressure, friction), and penetrative radiation render time tendency for mass, tracer, and momentum.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Conservation is important

Take the vertically-integrated Temperature budget ∂t η

−H dz θ

• =

−∇ · η

−H dz (uθ + Fsgs)

• + Qheat/(ρCp)

Assuming steady state and a basin: ρCp

• dx

η

−H dz(vθ + Fy) =

yn

ys dy

• dx Qheat

A meridional ocean heat transport is thus implied by the net surface forcing.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Lateral BCs for regional ocean models

Near-global observations are pushing models to improve. Argo + satellites provide high quality near-global information. These data sources are now assimilated into global ocean models. These products could generate the BC’s for our coupled regional models.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions

Motivation for using coupled models Posing the coupled model problem Basics on (Low-Frequency) Variability Mesoscale/regional examples

Summary

Where the envelope can be pushed

1

Role of resolution in climate

How/will climate sensitivity, variability, predictability be modified with eddying ocean simulations and higher atmospheric resolution? Coupled ocean-atmosphere models are still too coarse to resolve mesoscale SST influence on the atmosphere. This can readily be achieved with regional coupled models.

2

Regional and Global Coupled Models

There is a clear motivation for the development of both regional and global coupled models. and for a comparison and feedback between the two.

RFARNETI@ICTP.IT

Ocean-Atmosphere Interactions