Wallace J.M., Hobbs P.V. Atmospheric Science. An Introductory Survey

Подождите немного. Документ загружается.

414 The Atmospheric Boundary Layer

(hh) Other conditions being the same, alpine

glaciers and snow-fields lose more mass on

a humid summer day than on a dry summer

day.

(ii) On a clear, calm day, the surface sensible

heat flux into the air does not usually

become positive until 30 to 60 min after

sunrise.

(jj) In fair weather, the heat and momentum

fluxes at the top of the mixed layer due to

entrainment are usually downward, but the

moisture flux is positive.

(kk) You can estimate the static stability of

the boundary layer by looking at the

shape of the smoke plume from a smoke

stack.

(ll) In Fig. 9.40, why do the surface wind speed

and the cloudiness increase as the air flows

northward across the sharp front in the

sea-surface temperature field that lies

along 1 °N?

9.8 Estimate the temperature variance for the

velocity trace at the bottom of Fig. 9.6.

–101 –100 –99 –98 –97

–96 –95

–94 –93 –92 –91

–101

–100 –99

–98 –97

–96 –95

–94 –93 –92 –91

–7 –6 –5 –4 –3 –2 –1 01 3 456 7

–7–6–5–4–3–2–1

1234567

wind speed

m s

–1

]

East Pacific cross-equatorial cloud optical depth (MODIS),

SST (TMI) and wind stress (QUIKSCAT) (2000/09/11)

SST [C]

Fig. 9.40 Sea surface temperature (colored shading) surface winds (arrows) and clouds (gray shading) over the equatorial

Pacific at a time when the equatorial front in the sea surface temperature field is well defined along 1 °N. The scalloped appear-

ance of the front is due to the presence of tropical instability waves in the ocean. [Based on NASA QUIKSCAT, TMI, and MODIS

imagery. Courtesy of Robert Wood.]

P732951-Ch09.qxd 9/12/05 7:48 PM Page 414

Exercises 415

9.9 Prove that the definition of covariance reduces

to the definition of variance for the covariance

between any variable and itself.

9.10 Given the following variances in m

2

Where: Location A Location B

When (UTC): 1000 1100 1000 1100

0.50 0.50 0.70 0.50

0.25 0.50 0.25 0.25

0.70 0.50 0.70 0.25

When, where, and for which variables is the

turbulence (a) stationary, (b) isotropic, and

Some, but not all, of the answers Homogeneous

for u wind at 1100 UTC. Stationary for v wind at

location B. Isotropic at location A at 1100 UTC.

9.11 Given the following synchronous time series

for T(°C) and w(ms), find (a) mean

temperature, (b) mean velocity,

variance, and (e) kinematic heat flux.

T2122202525151823212416121922

w120322 33 00 1 42 3

9.12 Under what conditions is the Richardson

number not likely to be a reliable indicator for

the existence of turbulence? Why?

9.13 If the dissipation length scale is L



, what is the

e-folding time for the decay of turbulence (i.e.,

time for TKE



m to equal 1e of its initial

value), assuming you start with finite TKE but

there is no production, consumption, transport,

or advection?

9.14 The upward vertical heat flux in soil and rock

due to molecular conduction is given by

(9.34)

where K is the thermal conductivity of the

medium, in units of W m

1

, which ranges in

value from 0.1 for peat to 2.5 for wet sand. From

the first law of thermodynamics, we can write

(9.35)C











F K





z



where C is the heat capacity per unit volume in

3

1

(i.e., the product of the specific heat

times the density). If K is assumed to be inde-

pendent of depth

(9.36)

where D  K



C is called the thermal diffusivity.

Consider the response of the subsurface tem-

perature to a sinusoidal variation in surface

temperature with amplitude T

and period P.

(a) Show that the amplitude of the response

drops off exponentially with depth below the

surface with an e-folding depth of

(b) The e-folding depth of the annual cycle,

as estimated from Fig. 9.12, is 2 m. Estimate

the e-folding depth of the diurnal cycle at the

same site.

9.15 For the situation shown in Fig. 9.12 in the text:

(a) Making use of Eq. (9.36), derive a

relationship for the phase lag t between the

temperature wave at two depths differing by

height z, given P the wave period and D

the thermal diffusivity of the soil as defined

in the previous exercise. (b) Using the

annual cycle data given in Fig. 9.12 in the text,

estimate the phase lag between the temperature

curves at 1.5 and 6 m and then use it to find

the value of D.

9.16 Given the heat-flux profile of Fig. 9.22, extend

the method of Fig. 9.8 to estimate the sign of the

triple correlation in the middle of the

mixed layer, which is one of the unknowns in

Eq. (9.11).

9.17 Given: F

 0.2 K m s

1

, z

 1km,

 0.2 m s

1

, T  300 K, z

 0.01 m, find and

explain the significance of the values of the

(a) Deardorff velocity scale; (b) Obukhov length;

z  30 m.

9.18 The flux form of the Richardson number is



(





) w





wu



 wv



ww





h 

√





 D



P732951-Ch09.qxd 9/12/05 7:48 PM Page 415

416 The Atmospheric Boundary Layer

Use gradient transfer theory to show how R

related to R

9.19 If the wind speed is 5 m s

1

at z  10 m and

the air temperature is 20 °C at z  2 m, then

(a) what is the value of the sensible heat flux

at the surface of unirrigated grassland if the

skin temperature is 40 °C? (b) What is the

value of the latent heat flux? [Hint:



 1231

(W m

2

)(K m s

1

) for dry air at sea level.]

9.20 (a) Confirm that the following expression is a

solution to Eq. (9.26), and (b) that it satisfies

the boundary condition that V  0 at z  z

where

9.21 Use the expressions for surface-layer wind

speed from Eq. (9.22), from Exercise 9.4 in

Section 9.3.3, and from the previous exercise to

plot curves of V vs. z for (a) neutral (L ),

(b) stable (L  100 m), and (c) unstable

(L 10 m) stratifications and confirm that

their relative shapes are as plotted in Fig. 9.17a

and 9.17b. Use z

 0.1 m.

9.22 (a) If boundary-layer divergence is constant

during fair weather, the mixed layer can stop

growing during midafternoon even though

there are strong sensible heat fluxes into the

boundary layer at the ground.Why? (b) If

 constant, derive an equation showing the

growth of z

with time in a region where

divergence



is constant.

9.23 If F* is known, and if sF

s  0.1 sF*s, then

show how knowledge of time-averaged

temperature at two heights in the surface layer

and of mean humidity at the same two heights is

sufficient to estimate the sensible and latent

heat fluxes in the surface layer.

9.24 As a cold, continental air mass passes over the

Gulf Stream on a winter day, the temperature of

the air in the atmospheric boundary layer rises

x 



1  15

(z  z

)





 2 arctan x 











 2



1  x



 ln



1  x



by 10 K over a distance of 300 km.Within this

interval the average boundary layer depth is

1 km and the wind speed is 15 m s

1

.No

condensation is taking place within the

boundary layer and the radiative fluxes are

negligible. Calculate the sensible heat flux from

the sea surface.

9.25 (a) If drag at the ground represents a loss of

momentum from the mean wind, determine the

sign of if the mean wind is from the west.

Justify your result.

(b) Do the same for a wind from the east,

remembering that drag still represents a

momentum loss.

9.26 Given the following temperature profile,

determine and justify which layers of the

atmosphere are (a) statically stable, (b) neutral,

and (c) unstable.

z (km)



(°C)

221

1.8 23

1.6 19

1.4 19

1.2 13

1.0 16

0.2 16

010

9.27 If the total accumulated surface heat flux

from sunrise through sunset is 5100 km, then

use the sunrise sounding in the previous

exercise to estimate the depth and potential

temperature of the mixed layer just before

sunset.

9.28 Given a smoke stack half the height of a

valley: (a) describe the path of the centerline

of the smoke plume during day and night

during fair weather and (b) describe the

centerline path of the smoke on a strongly

windy day.

9.29 If you know the temperature and humidity

jumps across the top of the mixed layer and if

you know only the surface heat flux (but not

the surface moisture flux), show how you can

calculate the entrained heat and moisture

fluxes at the top of the mixed layer.

uw

P732951-Ch09.qxd 9/12/05 7:48 PM Page 416

Exercises 417

9.30 Show that over flat terrain the large-scale

vertical velocity at the top of the boundary layer

is approximately equal to

(9.28)

where

 z

{  V}

is the mass-weighted divergence in the boundary

layer and p

and p

are the pressures at the Earth’s

surface and at the top of the boundary layer.

{  V} 



(  V) dp

 p

)

P732951-Ch09.qxd 9/12/05 7:48 PM Page 417

P732951-Ch09.qxd 9/12/05 7:48 PM Page 418

The term climate refers to the mean state of the

atmosphere and related components of the Earth

system and it is also used in reference to atmospheric

variability on timescales longer than the 2- to 3-week

limit of most deterministic atmospheric predictability.

The mean state, including diurnal and seasonal varia-

tions, as defined by some prescribed averaging period,

is referred to as the climatological mean,

and depar-

tures from this mean state or normal are referred to as

climate anomalies.For example, a 4 °C temperature

anomaly denotes a temperature that is 4 °C above the

climatological mean for that particular location and

time of year. The term climate variability refers to

long-term variations or changes in the mean state, i.e.,

• intraseasonal climate variability denotes

month-to-month variations about the seasonally

varying climatological mean that occur within

the same season (e.g., the distinction between

an abnormally warm January and an abnormally

cold February),

• interannual variability denotes year-to-year

variations of annual or seasonal averages

(e.g., between the mean temperatures observed

in successive winter seasons), and

• decadal, century-scale, etc., denotes decade-to-

decade, century-to-century, etc. scale variations.

The distinction between climate variability and climate

change is largely semantic: if the variations of interest

take place within some specified interval (e.g., the 20th

century), they are referred to as the variability of the

climate within that interval, whereas if the variations

involve differences between two successive epochs

(e.g., the first and second halves of the 20th century),

they are referred to as the change in climate from one

epoch to the next.

The next section describes the present-day climate

and the processes that shape it. The nature and causes

of climate variability on various timescales is then

discussed and related to the history of past climate. The

section on climate variability is followed by a discussion

of the climate sensitivity and climate feedbacks. The

chapter concludes with a section summarizing some of

the scientific issues related to greenhouse warming.

10.1 The Present-Day Climate

The mean state of the climate system is determined by

the emission of radiation from the sun, the Earth’s

rotation rate and orbital characteristics, the composi-

tion of the atmosphere, and the interactions between

the atmosphere and the other components of the

Earth system that determine the fluxes of mass, energy,

and momentum at the Earth’s surface. Some of the

characteristics of that mean state were described in

Chapters 1 and 2. This section provides further

specifics concerning the diurnally and seasonally vary-

ing mean climate and the processes that maintain it.

10.1.1 Annual Mean Conditions

The global mean surface air temperature T

 15 °C

or 288 K and the Earth’s equivalent blackbody tem-

perature T

 255 K are equilibrium values, deter-

mined by the balance requirements that the globally

averaged net radiation through the top of the atmos-

phere, as well as the net energy flux (i.e., the sum of

419

Climate Dynamics

Conventional climatological means are based on prescribed sets of sequential 30-year periods, the most recent of which ended in

December 1990.

P732951-Ch10.qxd 12/09/2005 09:12 PM Page 419

420 Climate Dynamics

the short and longwave radiative fluxes and the fluxes

of latent and sensible heat), at the Earth’s surface, as

prescribed by (9.18) must vanish.

If these balance requirements were not fulfilled, the

temperatures would change until an equilibrium was

achieved. For example, if the net radiation at the top

of the atmosphere were downward, the Earth’s equiv-

alent blackbody temperature would have to rise until

the outgoing longwave radiation increased enough

to eliminate the imbalance. Based on global measure-

ments, such as those shown in Figs. 4.35 and 9.14, it is

possible to construct the global energy balance shown

in Fig. 10.1.

The 100 units of insolation (a synonym for the sea-

sonally varying, climatological-mean solar radiation)

incident on the top of the atmosphere represent a

flux density of 342 W m

2

integrated over the

Earth’s surface as shown in Fig. 4.8. Of the 100 units

incident on the top of the atmosphere, 3 are

absorbed by stratospheric ozone and 17 by water

vapor and clouds in the troposphere. A total of 30

units are reflected back to space: 20 from clouds and

aerosols, 6 from air molecules, and 4 from the Earth’s

surface. The total reflection (30 out of 100 units of

incoming radiation) is the Earth’s albedo.The

remaining 50 units are the net downward shortwave

flux through the Earth’s surface.

The Earth disposes of the energy that it absorbs

by a combination of longwave radiation and latent

and sensible heat fluxes as indicated by the red and

blue arrows in Fig. 10.1. The 110 units of longwave

radiation emitted from the Earth’s surface is equal to

effective emissivity of the surface, which ranges from

0.92 to 0.97 locally, times averaged over the sur-

face of the Earth, where T

is the surface temperature.

The net upward longwave flux through the Earth’s

surface (i.e., the difference between the upward emis-

sion from the surface and the downward emission

from clouds and greenhouse gases in the atmosphere)

amounts to only 21 units. Were it not for the presence

of the downward longwave fluxes (i.e., the greenhouse

effect) the Earth’s surface would be in equilibrium

with the incident solar radiation at a temperature

close to T

and the latent and sensible heat fluxes

would be much smaller than observed.

Exercise 10.1 On the basis of the data given in Fig.

10.1, describe the energy balance of the troposphere.

Solution: The troposphere is heated by the absorp-

tion of solar radiation, mostly by water vapor and

clouds (17 units), the absorption of longwave radiation

emitted by the Earth’s surface (110  12  98 units),

the absorption of longwave radiation emitted by

the stratosphere (5 units), the sensible heat transfer

through the Earth’s surface (5 units), and latent heat

release (24 units). Hence, the troposphere must emit a

total of 17  98  5  5  24  149 units of radiation

in the longwave part of the spectrum in order to

achieve a balance. It emits 89 units in the downward

direction and 60 units in the upward direction. ■

The annual mean latitudinal distribution of net radi-

ation at the top of the atmosphere, shown in Fig. 10.2,

was generated by zonally averaging the distributions

shown in Fig. 4.35. Because the temperature of the

Earth system is changing only very slowly, the globally

averaged insolation must be in balance with the

outgoing longwave radiation. Hence, there must be a

surplus of insolation relative to outgoing longwave

radiation at low latitudes and a deficit at high lati-

tudes. It is this imbalance that maintains the observed

equator-to-pole temperature contrast in the presence

of the strong poleward heat fluxes produced by baro-

clinic waves, as explained in Section 7.3 and 7.4.



Fig. 10.1 The global energy balance. The 100 units of incom-

ing energy represent the 342 W m

2

of incident solar radiation

averaged over the area of the Earth. Black arrows represent

shortwave radiation, red arrows represent longwave radiation,

and blue arrows represent the (nonradiative) fluxes of sensible

heat (SH) and latent heat (LH). The absorption and emission

by the Earth’s surface, the troposphere, and the stratosphere

each sum to zero, and the net flux through each of the inter-

faces sums to zero. [Adapted from Dennis L. Hartmann, Global

from Elsevier.]

100

110

S.H.

L.H.

TROPOSPHERE

EARTH’S SURFACE

STRATOSPHERE

SPACE

P732951-Ch10.qxd 12/09/2005 09:12 PM Page 420

10.1 The Present-Day Climate 421

The vertical distribution of temperature within

the troposphere is determined by the interplay

among radiative transfer, convection and large-

scale motions. The radiative equilibrium temperature

profile is unstable with respect to the dry adiabatic

lapse rate. The pronounced temperature minimum

near the 10-km level that defines the tropopause

(Fig. 1.9) corresponds roughly to the level of unit

optical depth for outgoing longwave radiation. Below

this level, the repeated absorption and reemission of

outgoing longwave radiation render radiative transfer

a relatively inefficient mechanism for disposing of the

energy absorbed at the Earth’s surface. Convection

and large-scale motions conspire to maintain the

observed lapse rate near a value of 6.5 K km

1

The observed lapse rate is stable, even with respect

to the saturated adiabatic lapse rate, because most of

the volume of the troposphere is filled with slowly sub-

siding air, which loses energy by emitting longwave

radiation as it sinks, as documented in Fig. 4.29. As the

air loses energy during its descent, its equivalent poten-

tial temperature and moist static energy decrease, cre-

ating a stable lapse rate, as shown schematically in

Fig. 10.3. It is only air parcels that have resided for

some time within the boundary layer, absorbing

sensible and latent heat from the underlying surface,

that are potentially capable of rising through this stably

stratified layer. Thermally direct large-scale motions,

which are characterized by the rising of warm air and

the sinking of cold air, also contribute to the stable

stratification. It is possible to mimic these effects in

simple radiative-convective equilibrium models by arti-

ficially limiting the lapse rate, as shown in Fig. 10.4. The

tropospheres of Mars and Venus and the photosphere

of the sun

can be modeled in a similar manner.

The concept of radiative-convective equilibrium is

helpful in resolving the apparent paradox that green-

house gases produce radiative cooling of the atmos-

phere (Fig. 4.29), yet their presence in the atmosphere

renders the Earth’s surface warmer than it would be in

their absence. An atmosphere entirely transparent to

solar radiation and in pure radiative equilibrium would

neither gain nor lose energy by radiative transfer in the

longwave part of the spectrum. However, it is apparent

from Fig. 10.4 that under conditions of radiative-

convective equilibrium, temperatures throughout most

of the depth of the troposphere are above radiative

equilibrium. It is because of their relative warmth

(maintained mainly by latent heat release and, to a

lesser extent, by the absorption of solar radiation and

the upward transport of sensible heat by atmospheric

motions) that greenhouse gases in the troposphere

emit more longwave radiation than they absorb.

Because the tropospheric lapse rate is determined not

Fig. 10.2 Annual average flux density of absorbed solar

radiation, outgoing terrestrial radiation, and net (absorbed

solar minus outgoing) radiation as a function of latitude in

units of W m

2

. Pink (blue) shading indicates a surplus (deficit)

of incoming radiation over outgoing radiation. [Adapted from

Dennis L. Hartmann, Global Physical Climatology, p. 31 (Copyright

1994), with permission from Elsevier.]

200

300

–100

100

0–30–60–90

°S

°N

Absorbed Solar

Emitted Longwave

Net

Radiation

Latitude

Flux Density (W m

–2

)

The sun’s photosphere is defined as the layer from which sunlight appears to be emitted; i.e., the level of unit optical depth for visible

radiation. The photosphere marks the transition from a lower, optically dense layer in which convection is the dominant mechanism

for transferring heat outward from the nuclear furnace in the sun’s core to a higher, optically thin layer in which radiative transfer is the

dominant energy transfer mechanism. Like the tropopause in planetary atmospheres, the photosphere is marked by a decrease in the lapse

rate from the convectively controlled layer below to the radiatively controlled layer above.

Fig. 10.3 Schematic of air parcels circulating in the atmos-

phere. The Colored shading represents potential temperature

or moist static energy, with pink indicating higher values and

blue lower values. Air parcels acquire latent and sensible heat

during the time that they reside within the boundary layer, rais-

ing their moist static energy. They conserve moist static energy

as they ascend rapidly in updrafts in clouds, and they cool by

radiative transfer as they descend much more slowly in clear air.

Tropopause

P732951-Ch10.qxd 12/09/2005 09:12 PM Page 421

422 Climate Dynamics

by radiative transfer, but by convection, it follows

that greenhouse gases warm not only the Earth’s sur-

face, but the entire troposphere.

In contrast to the troposphere, the stratosphere

is close to radiative equilibrium. Heating due to the

absorption of ultraviolet solar radiation by ozone

is balanced by the emission of longwave radiation

by greenhouse gases (mainly CO

O, and O

) so

that the net heating rate (Fig. 4.29) is very close to

zero. Raising the concentration of atmospheric CO

increases the emissivity of stratospheric air, thereby

enabling it to dispose of the solar energy absorbed by

ozone while emitting longwave radiation at a lower

temperature. Hence, while the troposphere is warmed

by the presence of CO

, the stratosphere is cooled.

10.1.2 Dependence on Time of Day

As the Earth rotates on its axis, fixed points on its

surface experience large imbalances in incoming and

outgoing radiative fluxes as they move in and out of

its shadow. As a point rotates through the sunlit, day

hemisphere, the atmosphere above it and the under-

lying surface are heated more strongly by the absorp-

tion of solar radiation than they are cooled by the

emission of longwave radiation. The energy gained

during the daylight hours is lost as the point rotates

through the shaded night hemisphere.

Over land, the response to the alternating heating

and cooling of the underlying surface produces diur-

nal variations in temperature, wind, cloudiness, pre-

cipitation, and boundary-layer structure, as discussed

in Chapter 9. Here we briefly discuss the direct

atmospheric response to the hour-to-hour changes in

the radiation balance that would occur, even in the

absence of the interactions with the underlying sur-

face. This response is often referred to as the thermal

(i.e., thermally driven) atmospheric tide.

Because of the atmosphere’s large “thermal iner-

tia,” diurnal temperature variations within the free

atmosphere are quite small, as illustrated in the

following exercise. The thin Martian atmosphere

reacts much more strongly to the diurnal cycle in

insolation (Table 2.5).

Exercise 10.2 If the Earth’s atmosphere emitted radi-

ation to space as a blackbody and if it were completely

insulated from the underlying surfaces, at what mass-

averaged rate would it cool during the night?

Solution: The cooling rate in degrees K per unit time

is equal to the rate of energy loss divided by the heat

capacity per unit area of the free atmosphere. During

the night the atmosphere continues to emit infrared

radiation to space at its equivalent blackbody tempera-

ture of 255 K; hence it loses energy at a rate of

The heat capacity of the atmosphere (per m

) is

equal to the specific heat of dry air c

times the mass

per unit area ( pg) or

1004 J kg

1

 10

9.8 m s

2

 10

J K

1

2

E 



 5.67  10

8

 (255)

 239 W m

2

Throughout the tropics the observed lapse-rate  is close to the saturated adiabatic value 

.As the Earth’s surface warms, the latent heat

released in moist adiabatic ascent increases, and so the numerical value of 

decreases. Hence, if  remains close to 

as the tropical tropo-

sphere warms, temperatures in the upper troposphere will warm more rapidly than surface air temperature, as demonstrated in Exercise 3.45.

The gravitational attraction of the moon and sun also induce atmospheric tides, but these gravitational tides are much weaker than the

thermal tides.

180

220

260 300

340

100

1000

Pure radiative equilibrium

Dry adiabatic adjustment

6.5

°C/km adjustment

Pressure (hPa)

Altitude (km)

Temperature (K)

Fig. 10.4 Calculated temperature profiles for the Earth’s

atmosphere assuming pure radiative equilibrium (red curve) and

radiative convective equilibrium, in which the lapse rate is artificially

constrained not to exceed the dry adiabatic value (dashed black

curve) and the observed global-mean tropospheric lapse rate

(blue curve). [Adapted from J. Atmos. Sci., 21, p. 370 (1964).]

P732951-Ch10.qxd 12/09/2005 09:12 PM Page 422

10.1 The Present-Day Climate 423

The cooling rate is therefore

Night lasts 12 h or 4.32  10

s. Hence, the overnight

cooling is only 1K. ■

The diurnal variation in incident solar radiation is

not a pure sine wave: it is proportional to the cosine

of the solar zenith angle during the day and zero at

night (see Fig. 9.9). Because of the nonsinusoidal time

dependence of the forcing, atmospheric tides are

made up of an ensemble of frequencies that are

integral multiples of one cycle per day. The most

important of these are the diurnal (one cycle per day)

and semidiurnal (two cycles per day) tides. The semi-

diurnal component, a response to the periodic heating

and cooling of the stratospheric ozone layer, is clearly

evident in tropical sea-level pressure records. The

corresponding diurnal and semidiurnal variations

in the horizontal wind field increase with height

from 1–2 m s

1

at tropospheric levels to more than

10 m s

1

in the mesosphere and thermosphere.

10.1.3 Seasonal Dependence

Each year, as Earth revolves around the sun, the extra-

tropical continents experience large temperature swings

and many regions of the tropics experience dramatic

changes in rainfall. These periodic climate fluctuations

are largely a response to the obliquity of the Earth’s

axis of rotation relative to the plane of the ecliptic.

The seasonally varying latitudinal distribution of

insolation incident upon the top of the atmosphere is

shown in Fig. 10.5. The equator-to-pole gradient is

strongest during winter, when the polar cap is in

darkness. In the summer hemisphere the increasing

length of daylight with latitude more than offsets the

increasing solar zenith angle so that insolation

increases slightly with latitude. The large seasonal

variations in insolation give rise to seasonally varying

imbalances in net radiation at the top of the atmos-

phere (Fig. 10.6), which range up to 100 W m

2

over

the subtropical and midlatitude oceans. Net radiation

at the surface of the ocean (not shown) exhibits a

similar distribution: upward in the winter hemisphere

and downward in the summer hemisphere. Most of

the excess insolation in the summer hemisphere is

stored in the ocean mixed layer and the cryosphere

and is released during the following winter, thereby



239 Wm

2

1

2

 2.39  10

5

1

moderating the seasonal contrast in atmospheric

temperature.

The storage of heat in the ocean mixed layer is

reflected in the warming of a relatively shallow layer of

water near the surface during late spring and early

summer, which leads to the formation of the seasonal

thermocline, as shown in Fig. 10.7. During autumn and

early winter reduced insolation and enhanced latent

and sensible heat fluxes cool the mixed layer, releasing

the heat that was stored during the previous summer.

The warming of high latitude regions during winter by

the release of this stored heat is larger than the pole-

ward heat transport by the western boundary currents

and (in the North Atlantic) by the thermohaline

circulation. As the seasonal thermocline cools during

autumn and winter, it also deepens by entraining water

from below. The ocean mixed layer reaches its mini-

mum temperature and maximum depth in spring, set-

ting the stage for the redevelopment of the thermocline

much closer to the ocean surface a few months later.

The large seasonal variations in the extent of

polar northern hemisphere pack ice (Fig. 2.12) and

other elements of the cryosphere also contribute to

10 10

JFMAMJJASOND

Calendar month

90 °S

°N

Latitude

0°

Fig. 10.5 Insolation incident on a unit horizontal surface

at the top of the atmosphere, expressed in units of MJ m

2

integrated over the 24-h day, as a function of latitude and

calendar month. Within the shaded regions the insolation

is zero. To convert to units of W m

2

multiply the contour

labels by 11.57. Solar declination refers to the latitude at which

the sun is overhead at noon. [Adapted from Meteorological Tables

(R. J. List, Ed.), 6th Ed., Smithsonian Institute (1951), p. 417.]

P732951-Ch10.qxd 12/09/2005 09:12 PM Page 423