Midterm review - UCLA: Atmospheric and Oceanic Sciences

Download Report

Transcript Midterm review - UCLA: Atmospheric and Oceanic Sciences

Mid-course overview
(NB: overview a guide but not necessarily everything)
Chapter 1. Intro…. Review elements include:
 Basic definitions, terms, e.g., Climate, climatology,
anomaly, teleconnection, …
 Trace gases, anthropogenic increase
 Perspective from paleoclimate on where fossil fuels come
from, some idea of time scales, natural variability, are
recent changes unusual,..
 ENSO material, mostly as comes back in chpt 4.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
1.1 Climate dynamics, climate change and climate prediction
 Climate: average condition of the atmosphere, ocean, land
surfaces and the ecosystems in them (includes average
measures of weather-related variability, e.g. storm
frequency)
 Weather: state of atmosphere and ocean at given moment.
 Average taken over January of many different years to obtain a
climatological value for January, etc.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
 Climate change:
• occurring on many time scales, including those that affect
human activities.
• time period used in the average will affect the climate that one
defines.
• e.g., 1950-1970 will differ from the average from 1980-2000.
 Climate variability:
• essentially all the variability that is not just weather.
• e.g., ice ages, warm climate at the time of dinosaurs, drought in
African Sahel region, and El Niño.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
 Global warming: predicted warming, & associated changes in
the climate system in response to increases in "greenhouse
gases" emitted into atmosphere by human activities.
 Greenhouse gases: e.g., carbon dioxide, methane and
chlorofluorocarbons: trace gases that absorb infrared
radiation, affect the Earth's energy budget.
 warming tendency, known as the greenhouse effect
 Climate prediction endeavor to predict not only humaninduced changes but the natural variations. e.g., El Niño
 Climate models:
• Mathematical representations
of the climate system
• typically equations for temperature,
winds, ocean currents and other climate
variables solved numerically on computers.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
 El Niño:
• largest interannual (year-to-year) climate variation
interaction between the tropical Pacific ocean and the
atmosphere above it.
• a prime example of natural climate variability.
• first phenomenon for which the essential role of dynamical
interaction between atmosphere and ocean was
demonstrated.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
 Teleconnections: remote effects of El Niño (or other
regional climate variations).
 Anomaly: departure from normal climatological
conditions.
• calculated by difference between value of a variable at a given
time, e.g., pressure or temperature for a particular month,
and subtracting the climatology of that variable.
 Climatology includes the normal seasonal cycle.
• e.g., anomaly of summer rainfall for June, July and August
1997, = average of rainfall over that period minus averages of
all June, July and August values over a much longer period,
such as 1950-1998.
• To be precise, the averaging time period for the anomaly and
the averaging time period for the climatology should be
specified.
• e.g., monthly averaged SST anomalies relative to 1950-2000 mean.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Table 1.1
1.4 Global change in recent history
Air main constituents
Formula Concentration
Nitrogen
N2
78.08%
Oxygen
O2
20.95%
Water
H2O
0.1 to 2 %
Trace gas name
Concentration (in 2004)
Carbon dioxide
CO2
377 (parts per million) ppm, ~0.038%
Methane
CH4
1.75 ppm
Nitrous oxide
N2O
0.32 ppm
Ozone
O3
0.000251 ppm (atm. average)
(~10 ppm max in stratosphere)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.1
Carbon dioxide concentrations since 1958,
measured at Mauna Loa, Hawaii.
•Annual, interannual variations:
biological impacts on carbon cycle
•Trend: due to fossil fuel emissions.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.2
Concentration of various
trace gases
estimated since 1850
•Methane: Cattle, sheep, rice
paddies, fossil fuel byproduct; wetlands, termites
(parts per billion).
•Nitrous Oxide: biomass
burning, fertilizers?
•Chlorofluorocarbons: manmade, zero before 1950
(parts per trillion).
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.3
Global mean surface temperatures estimated since preindustrial times
• Anomalies relative to 1961-1990 mean
• Annual average values of combined near-surface air temperature
over continents and sea surface temperature over ocean.
• Curve: smoothing similar to a decadal running average.
• From University of East Anglia Climatic Research Unit, Brohan
et al. (2006).
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
1.5 El Niño: An example of natural climate variability
 ENSO: El Niño/Southern Oscillation.
 [Cont’d in Chpt 4]
• El Niño is associated with warm phase of a phenomenon that
is largely cyclic.
• La Niña for the cold phase.
 El Niño arises in tropical Pacific along the equator.
• Changes in sea surface temperature, ocean subsurface
temperatures down to a few hundred meters depth, rainfall,
and winds: ocean-atmos. interaction!
• Variations in the Pacific basin within about 10-15 degrees
latitude of the equator are the primary variables.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.12
Paleoclimate: A few climate notes with Geological time scale
 Very distant past--- Myr=millions
of years
 Key points:
 Climate can vary substantially, on
all timescales
 Long periods in deep past with
warmer climate than present (&
higher est. CO2 )
 deposition over 100s of millions of
years sequesters carbon dioxide as
fossil fuels (oil, coal, natural gas)
 return of this CO2 to atmosphere
occurring over very short period.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.13
Antarctic ice core records of CO2, deuterium isotope ratio
variations (dD), and Antarctic air temperature inferred from dD
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Chapter 2
Basics of Global Climate
Review elements include:
 Basic concepts and terminology that tend to recur later:
• E.g., albedo, moist convection, easterly winds
2.1 Components and phenomena in the climate system
 Climate processes
• solar radiation tends to get through the atmosphere; ocean
heated from above  stable to vertical motions.
• warm surface layer, colder deep waters
• mixing near ocean surface  upper mixed layer ~ 50 m
• mixing carries surface warming down as far as
thermocline, layer of rapid transition of temperature to
the colder abyssal waters below
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.2
Satellite image based
on visible light with
5°x5° grid overlay
Courtesy of NASA
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
The parameterization problem
 For each grid box in a climate model, only the average
across the grid box of wind, temperature, etc. is
represented.
 The average of smaller scale effects has important impacts
on large-scale climate.
• e.g., clouds primarily occur at small scales, yet the
average amount of sunlight reflected by clouds affects
the average solar heating of a whole grid box.
 Parameterization: representing average effects of scales
smaller than the grid scale (e.g., clouds) as a function of
the grid scale variables (e.g. temperature and moisture) in
a climate model.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
2.2 Basics of radiative forcing
 Solar radiation input: Infrared radiation (IR) is the only
way this heat input can be balanced by heat loss to space
 Since IR emissions depend on the Earth's temperature, the
planet tends to adjust to a temperature where IR energy
loss balances solar input:
 Blackbody radiation: approximation for how radiation
depends on temperature:
 Total energy flux integrated across all wavelengths of light
R = sT4
 Full climate models do detailed computation as a function
of wavelength, for every level in the atmosphere, …
 [& Trace gases absorb IR at wavelengths where O2, N2
ineffective….]
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
2.3 Globally averaged energy budget
Pathways of energy transfer in a global average
Figure 2.8
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Atmosphere emits IR downward  absorbed at surface.
Greenhouse effect:
•The upward IR from the surface is mostly
trapped in the atmosphere, rather than
escaping directly to space, so it tends to
heat the atmosphere.
•The atmosphere warms to a temperature
where it emits sufficient radiation to
balance the heat budget, but it emits both
upward and downward, so part of the
energy is returned back down to the
surface where it is absorbed.
•This results in additional warming of the
surface, compared to a case with no
atmospheric absorption of IR.
•Both gases and clouds contribute to absorption of IR and thus
to the greenhouse effect.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Pathways of energy transfer in a global average (cont.)
 At the top of the atmosphere, in the global average and for
a steady climate:
• IR emitted balances incoming solar.
 Global warming involves a slight imbalance:
• a change in the greenhouse effect slightly less IR
emitted from the top (chap. 6).
• small imbalance slow warming.
 Three roles for clouds and convection:
1. heating of the atmosphere (through a deep layer)
2. reflection of solar radiation (contributing to albedo)
3. trapping of infrared radiation (contributing to the
greenhouse effect)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
2.4 Gradients of rad. forcing and energy transport by atm.
 Differences in input of solar energy between latitudes 
temperature gradients.
 These gradients would be huge if it were not for heat transport
in ocean & atmosphere and heat storage in ocean.
Figure 2.9
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Latitude structure of the circulation (cont.)
 Hadley cell: thermally driven,
overturning circulation, rising in
the tropics and sinking at
slightly higher latitudes (the
subtropics).
 Explanations of this in chpt. 3
 Relation to observations
• rising branch assoc. with
convective heating and heavy
rainfall; subtropical descent
regions, little rain.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.13
Intertropical convergence
zones (ITZCs) or tropical
convection zones: heavy
precipitation features deep in
the tropics, (convergence refers
to the low level winds that
converge into these regions).
Monsoons: tropical convection
zones move northward in
northern summer, southward
in southern summer, especially
over continents.
Not just a function of latitude,
e.g., strong convection over
tropical western Pacific, little
over cold eastern Pacific:
Walker circulation along
equator
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.16
 Equatorial cold tongue:
along the equator in the
Pacific.
• maintained by upwelling
of cold water from below.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.14
Equatorial Walker circulation
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.18
Ocean vertical structure
 Ocean surface is
warmed from above 
lighter water over
denser water (“stable
stratification”).
 Deep waters tend to
remain cold
• on long time scales,
import of cold waters
from a few sinking
regions near the poles
maintains cold
temperatures.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.19
The thermohaline circulation
 Salinity (concentration of salt) affects ocean density in addition
to temperature.
 Waters dense enough to sink: cold and salty
 Thermohaline circulation: deep overturning circulation is
termed the (thermal for the temperature, haline from the greek
word for salt, hals).
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.21b
Values from Denman et al. , IPCC (2007);
format follows Sarmiento and Gruber , Physics
Today (2002).
The carbon cycle
 Fossil fuels 6.4 PgC/yr (1990s) (incl. 0.1 PgC/yr cement production)
 ~40% coal, 40% from oil and derivatives such as gasoline, 20% from natural gas
 1.6 PgC/yr land-use change; e.g., deforestation to agricultural (smaller C storage)
 1990s; anthropogenic emissions ~8 PgC/yr (6.4 fossil fuels + 1.6 land use change)
 Fortunately, less than half remains in the atmosphere
 2.2 PgC/yr increased flux into ocean + 2.5 PgC/yr taken up by land vegetation
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 2.22
Values are from Denman et al., IPCC (2007); format
follows Sarmiento and Gruber , Physics Today (2002).
Fossil fuel emissions and increases in atmospheric CO2 concentrations
 Fossil fuel emissions converted to CO2 concentration change if all remained in
the atmosphere (1 ppm for each 2.1 PgC); rising …
 Actual rate of accumulation (change in concentration each year; all positive =
rising concentration, but variable rate of increase)
 Variations primarily due to land biosphere, e.g., droughts assoc with ENSO
 Accumulation rate ~ 55% of fossil fuel emissions on average
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Chapter 3
Physical Processes in the Climate System
Review elements include:
 Where do we get the equations in climate models
(conservation of momentum, energy, mass,…+eq of state)
 Main balances from these equations *especially as we
apply them to explain important climate features*
• PGF vs Coriolis
• Thermal circulation, thermal expansion
• Upwelling,…
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
3.1 Conservation of Momentum
Only in vertical
d velocity = Coriolis+PGF+gravity+F
drag
dt
eqs. 3.4 & 3.5
• Coriolis force: due to rotation of earth (apparent force).
• PGF: pressure gradient force. Tends to move air from high
to low pressure.
• Fdrag: friction-like forces due to turbulent or surface drag.
Use force per unit mass for atm/oc.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Coriolis force (cont.)
• Turns a body or air/water parcel to the right in the northern
hemisphere; to the left in the southern hemisphere.
• Exactly on the equator, the horizontal component of the
Coriolis force is zero
• Acts only for bodies moving relative to the surface of the
Earth’s equator and is proportional to velocity.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 3.4
Schematic of geostrophic wind and wind with frictional effects
Geostrophic balance:
At large scales at mid-latitudes and approaching the
tropics the Coriolis force and the pressure gradient
force are the dominant forces (for horizontal motions)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Section 3.1 Overview
• An approximate balance between the Coriolis force and the
pressure gradient force holds for winds and currents in many
applications (geostrophic balance) (Fig. 3.4).
• The Coriolis force tends to turn a flow to the right of its
motion in the Northern Hemisphere (left in the Southern
Hemisphere); the pressure gradient force acts from high toward
low pressure.
• The Coriolis parameter f varies with latitude (zero at the
equator, increasing to the north, negative to the south); this is
called the beta-effect ( = rate of change of f with latitude).
• In the vertical direction, the pressure gradient force balances
gravity (hydrostatic balance). This allows us to use pressure as a
vertical coordinate. Pressure is proportional to the mass above
in the atmospheric or oceanic column.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 3.5
Application: thermal circulation
e.g.:
Tropics
West Pacific
(Hadley circ)
(Walker circ.)
subtropics
East Pacific
• relatively low pressure (at given height) at low levels in
warm region; PGF toward warm region (near surface)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Section 3.2 Overview
• Atmos: relationship of density to pressure and temperature
from ideal gas law
• Ocean: density depends on temperature (warmer= less dense,
e.g. sea level rise by warming) & salinity (saltier= more dense).
• Thermal circulations (Fig. 3.5): warm atmospheric column
has low pressure near the surface and high pressure aloft
relative to pressure at same height in a neighboring cold region.
Reason: see Fig. 3.5
• PGF near surface toward warm region; Coriolis force may
affect circulation but warm region tends to have convergence &
rising. e.g.: Walker, Hadley circulations
• sea level rise by thermal expansion: density changes by given
fraction (thermal expansion coefficient) so sea level rise proportional
to depth of column that warms*dT; e.g., deep vs upper ocean
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Section 3.3 Overview
• Ocean: time rate of change of temperature of water parcel
given by heating
• for a surface layer: net surface heat flux from the atm. minus
the flux out the bottom by mixing
• Atmosphere: Temperature eqn. similar to ocean but…
• when an air parcel rises, temperature decreases as parcel
expands towards lower pressure.
• Quickly rising air parcel (e.g. in thermals): little heat is
exchanged
• temperature decreases at 10 C/km (the dry adiabatic lapse
rate).
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Section 3.3 Overview (cont.)
• Time derivatives following parcel hide complexity of the
system : the parcels themselves tend to deform in complex ways
if followed for a long time.
•Results in the loss of predictability for weather.
• The time derivative for temperature at a fixed point is
obtained by expanding the time derivative for the parcel in
terms of velocity times the gradients of temperature (advection).
• Similar procedure applies in other equations.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 3.9
Coastal upwelling: e.g., Peru; northward wind
component along a north-south coast
• Drag of wind stress tends to accelerate currents northward
• Coriolis force turns current to left in S. Hem
drag
[momentum eqn. fu ≈ Fy ]
• u away from coast
horizontal divergence
upwelling
from below [thru bottom of surface layer ≈ 50m]
∂w
[Continuity eqn. D = _
]
∂z
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.7
Processes leading to equatorial upwelling
• Wind stress accelerates currents westward
[wind speed fast relative to currents, so frictional drag at surface
slows the wind but accelerates the currents]
• Just north of Equator small Coriolis force turns current
slightly to right (south of Equator to the left)
divergence in
surface layer
balanced by upwelling from below
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 3.11
3.4e Conservation of warm water mass in idealized layer above thermocline
warm less dense
cold dense
• Warm light water above thermocline at depth h
• Horizontal divergence/convergence in upper layer
movement of
thermocline
^
∂h
[approx.
+ HD = 0 H = mean thermocline depth,
∂t
^
D vertical avg. thru layer ]
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Section 3.5 Overview
• Conservation of mass gives equations for water vapor
(atmosphere) and salinity (ocean); and other things, e.g. ice/snow
•water vapor main sinks moist convection &precipitation; source
surface evaporation (transport in between)
•Salinity at the ocean surface is increased by evaporation and
decreased by precipitation.
• latent heat of condensation: water vapor sink gives heating in
clouds (connecting mass and energy equations)
•Latent heat of melting important to surface mass balance of ice
(or snow), e.g., application to time needed to melt an ice sheet by
surface heat flux imbalance [W/m2 vs kg/m2 *J/kg]
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Section 3.6 Overview
• Saturation of moist air depends on temperature according to
Figure 3.12. Relative humidity gives the water vapor relative to the
saturation value.
• A rising parcel in moist convection decreases in temperate
according to the dry adiabatic lapse rate until it saturates, then has
a smaller moist adiabatic lapse rate. The temperature curve in
Figure 3.13 (the moist adiabat) depends on only the surface
temperature and humidity where the parcel started.
• If this curve is warmer than the temperature at upper levels,
convection typically occurs.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
3.7 Wave Processes in the Atmosphere and Ocean: Overview
[Skim for qualitative background for chpt 4.]
• Waves play an important role in communicating effects from one
part of the atmosphere to another.
• Rossby waves depend on the beta-effect [change of coriolis force
with latitude]. Their inherent phase speed is westward. In a
westerly mean flow, stationary Rossby waves can occur in which
the eastward motion of the flow balances the westward
propagation. Stationary perturbations, such as convective heating
anomalies during El Nino, tend to excite wavetrains of stationary
Rossby waves.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Chapter 4
El Nino…
Review elements include:
 Tropical Pacific climatology as it sets the stage for ENSO
 Schematics of ENSO (but with sense of how these connect
to observations seen earlier)
 Feedbacks that strengthen El Nino/La Nina (Bjerknes
hypothesis feedbacks)
 Linkage between sea surface height, thermocline depth,
and pressure gradients in the upper ocean
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.7
December 1997 Anomalies of sea surface temperature
during the fully developed warm phase of ENSO
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 4.2 (Chapter 4 preview)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 4.3
The Bjerknes feedbacks (warm phase)
• Positive feedback loop reinforces initial anomaly
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.5
Commonly used index regions for ENSO SST anomalies
 When SST in the Niño-3 region is warm during El Niño, the SOI tends to
be negative, i.e., pressure is low in the eastern Pacific relative to the west.
 Pressure gradient tends to produce anomalous winds blowing from west
to east along the equator.
 Reverse during periods of cold equatorial Pacific SST (La Niña).
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.6
Nino-3 index of equatorial Pacific sea surface temperature anomalies
and the Southern Oscillation Index of atmospheric pressure anomalies
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 1.10
December 1997 Anomalies of sea level height
during the fully developed warm phase of ENSO
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
4.4 Pressure gradients in an idealized upper layer
Figure 4.4
Idealized upper
ocean layer
• Sea surface height  thermocline depth [cm vs. m]
• PGF from regions of deep thermocline (high sea surface=high
pressure above thermocline) toward regions of shallower
thermocline
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Response of the ocean to a westerly wind anomaly
Re: Onset and demise of El
Figure 4.13
Wind anoms
currents
Niño warm phase
• To east of the wind anomalies,
equatorial jet (Kelvin wave) extends
east, deepening thermocline (H)
• (recall: warms SST…)
• To west, inflow of water to jet (in
oc. upper layer) comes from off
the equator (but little effect on SST)
• shallow thermocline in west
extends westward (Rossby wave),
as mass transferred to east by jet
• when reaches western boundary,
can no longer supply mass by
extending shallow region
• Weakening of jet extends eastward,
ending warm phase
• As wind anomalies weaken,
shallow thermocline extends
eastward: transition to cold phase
Deep Thermocline anoms
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 4.5
Two positions of the thermocline, indicating
region of thermocline anomalies
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 4.8b
The transition into the 1997-98 El Niño warm phase (Apr. 1997)
Slowly evolving thermocline depth anomalies (= subsurface
temperature anoms) affect eastern Pacific SST later. Key to
predictability.
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 4.8c
The transition into the 1997-98 El Niño warm phase (Sep. 1997)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Figure 4.8d
The transition into the 1997-98 El Niño warm phase (Jan. 1998)
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
Basis for ENSO forecasts and Limits to skill
• weather unpredictable beyond ~two weeks
• But interaction with the slowly evolving ocean (both in Bjerknes
feedbacks and delayed effects of Western Pacific thermocline anomalies)
permits prediction of ENSO Eastern Pacific SST anomalies
(and associated atmospheric anomalies) up to ~9 months ahead
• Forecast skill tends to degrade with longer lead time for forecast
• Loss of skill in ENSO forecasts:
(i) Imperfections in the forecast system
-e.g., model errors, scarcity of input data (can be improved, if $)
(ii) Fundamental limits to predictability
- “weather noise”: acts like a random forcing on slow oceanatmosphere interaction
e.g., SST gradient determines average strength of Tradewinds.
But in a particular month, transient weather events can cause
equatorial Easterlies to differ from this
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP
An ensemble of forecasts during
the onset of the 1998-99 La Niña
•Start coupled model
Figure 4.18
from different ocean
initial conditions (leading
also to changes in atm. )
•Initial differences grow
 ensemble of prediction
runs
•Ensemble spread gives
estimate of uncertainty
• Spread tends to grow
with time (due to weather
noise & coupled
feedbacks)
• Ensemble mean gives
Courtesy of the European Centre for Medium-range Weather Forecasting.
best estimate
Neelin, 2011. Climate Change and Climate Modeling, Cambridge UP