Vietnam part 6 (optional)

That over the last 100 years, the increase in mean global temperature corresponds to increases in, or introduction of, certain trace gases that can strongly increase absorption of infrar

From the best estimates of temperature over the last millennium, the mean global surface

temperatures observed in the last decade are thought to be warmest, although, given the uncertainty in estimates of mean global surface temperature, warmer decades could well have occurred (the grey areas give an indication as to the uncertainty) More reliable estimates of mean global temperatures are thought to occur after about 1600 Certainly since 1900 a sharp increase in mean global surface temperature can be seen

The extent of anthropogenic influence on mean global temperature is still uncertain, as there also appears to be large natural temperature variations That over the last 100 years, the increase in mean global temperature corresponds to increases in, or introduction of, certain trace gases that can strongly increase absorption of infrared radiation is not in debate However, increase in absorption of infrared radiation is only the first step in a more complex interaction that apparently occurs in the Earth’s system The influence of many of these interactions is still uncertain and there are likely other important factors still to be uncovered Here we concentrate on what is known about the direct effect of trace gases on the radiative balance of the atmosphere

Most of the large studies undertaken to construct the global temperature record over the last

100 years are in relatively good agreement The main consensus is that, since around the year

1910 there has occurred a rather abrupt increase in the Earth’s lower atmosphere temperature

and that this increase was interrupted between 1940 and 1980, but since then has continued

until today

Abstracts from this publication:

ABSTRACT (Xoplaki et al 2005):

We evaluate variability, trends, uncertainties, and change of extremes

of reconstructed and observed European spring and autumn temperature

back to 1500 Spring and autumn temperature experienced systematic

century-scale cooling compared to present conditions The coldest

springs appeared during the Maunder Minimum (DT = -1 K wrt 1901–2000)

The amplitude of spring temperature variations at decadal and multidecadal

scales doubles that of autumn and is most expressed in northeastern Europe

The decade 1995–2004 was very likely the warmest of the last half millennium

Anomalously warm springs and autumns have generally become more extreme

in recent decades However, the recent changes are statistically not

fields for Europe back to 1500 show that the late 20th- and early

21st-century European climate is very likely (>95% confidence level)

warmer than that of any time during the past 500 years This agrees with

findings for the entire Northern Hemisphere European winter average

temperatures during the period 1500 to 1900 were reduced by ~0.5°C

(0.25°C for annual mean temperatures) compared to the 20th century

Summer temperatures did not experience systematic century-scale cooling

relative to present conditions The coldest European winter was 1708/1709;

2003 was by far the hottest summer

This slide is a reminder of an earlier slide presented in the course It shows that the average power input (think of this as Joules per second) to the Earth’s Atmosphere from the Sun is

1366 Wm-2, where the unit area is the cross-sectional area of the Earth This is a yearly average value that changes very little from decade to decade, implying that any small variations here cannot account for the observed change in Earth’s average surface temperature over the last 100 years

This slide shows some types of reflective surfaces on the Earth that contribute to the overall average surface albedo of 0.31 Noteworthy are the effects of snow and cloud How would the surface albedo be effected by a temperature change of the Earth’s surface?

Some of the electromagnetic radiation of the Sun that is intercepted by the Earth is reflected

back into space On average the fraction reflected is found to be 0.31 (31%) The fraction is

called the albedo,  Thus, taking this into account, we know the rate of energy absorption by

the Earth (which includes absorption by its atmosphere) One must also note that other sources

of energy are negligible (such as heating from the Earth’s interior and combustion of fossil

fuels) If we assume that a steady-state is reached for which the rate of absorption of

electromagnetic energy equals the rate of emitted electromagnetic energy, we can use the

Stefan-Boltzmann relation to calculate the temperature of the Earth’s surface When this is

done, the predicted average temperature of the Earth’s surface is 253 K, which is about 35 K

less than the experimentally-determined average value of about 288 K The explanation for

this discrepancy is quite simple and not at all surprising: the Earth’s atmosphere is insulating

One is not referring here to conductive insulation (as one cannot conduct heat into space), one

refers to radiative insulation Though it is not surprising that the outer part of a body (the

atmosphere) provides some form of heat insulation, what is slightly unusual is that the part of

the body that is mainly heated by the Sun lies within the insulating sheet It is the Earth’s

surface and not the outer atmosphere that receives most of the Sun’s energy The Sun’s output

energy maximum happens to correspond to a spectral region (visible) in which atmospheric

molecules (especially O2, H2O, and N2) do not absorb significantly Much of the radiated

photons – at longer wavelengths - from the Earth’s surface however are absorbed by the

atmosphere

If one looks at the emission spectrum of the Earth from space one finds that it is equivalent to

a black body at about 253 K It is just that it is slightly warmer under the ‘blanket’ The more

The Sun has a surface temperature of about 5800 K but just below the surface it is warmer

than this Likewise the surface of the Earth/atmosphere system is 253 K, but it is difficult to

define exactly at which height this surface is situated – as it varies with time and position

Ref for Earth’s interior input H Pollack, S Hurter and J Johnson Rev Geophys., 31

(1993), p 267

The two most important concepts for the understanding of radiative forcing, the term given to

the absorption of Earth’s radiation by its atmospheric constituents, are that (a) filters of

electromagnetic radiation emit light themselves depending on their temperature, and (b) that

the amount of light absorbed by a filter is not necessarily linearly related to its thickness (or

concentration)

An isolated body that is heated continually (say the Earth by the Sun) shall eventually emit radiation according to its steady-state surface temperature If another (this time unheated) body

is placed in the vicinity of the first it will be heated by it Eventually this second body will emit radiation according to its surface temperature based on the energy it receives from the first body Now however the first body receives two sources of energy (one from the Sun say, and the other from the body it is heating via radiation) The overall effect it that the first body

reaches a new higher steady-state temperature This is a universal principle So if a CO2

molecule is placed near the Earth’s surface and absorbs radiation from it The temperature of the Earth’s surface must increase as a result of this process alone

This slide and the previous one illustrate that when viewing a body via its radiative emission one must take account that perhaps some of the detected radiation does not originate from the same surface This example looks at a spatial separation of energy, but it is entirely equivalent

to treat it as a spectral separation of energy where some wavelengths can penetrate unabsorbed through higher surfaces These wavelengths then have a partial black-body spectrum of the warmer surface beneath

This slide and the previous one illustrate that when viewing a body via its radiative emission

one must take account that perhaps some of the detected radiation does not originate from the

same surface This example looks at a spatial separation of energy, but it is entirely equivalent

to treat it as a spectral separation of energy where some wavelengths can penetrate unabsorbed

through higher surfaces These wavelengths then have a partial black-body spectrum of the

warmer surface beneath

Previously, the radiative transfer was discussed in terms of several blocks contained in a

perfectly insulating box with only the bottom block having an internal source of energy The

resulting equilibrium situation is given in the diagram on the left Now consider the three

upper blocks brought together so that thermal conduction between them is perfect The

equilibrium situation is the same as having a single block above the main block For the last

diagram, the three upper blocks are brought into contact with the main block In this situation

the power across the surface of the main block is 1000 W (per unit area) In other words, the

three blocks above now make no difference to the power emitted from the bottom surface

The key to radiative forcing of the Earth’s Troposphere is the cool absorber A molecule

absorbs radiation at altitude where it is cooler This absorbed energy is converted to heat and

re-distributed by a combination air motions and conduction throughout the troposphere The

cool gas emits radiation to space according to its temperature, which is less than that it has

absorbed The cooler the molecule, the more efficient the trapping of radiation Absorption

bands that are fully saturated will effectively trap radiation at the top of the troposphere, where

it is coldest The saturation of absorption may well go beyond the Tropopause but the

difference in energy will go into the stratosphere and by this mechanism the stratosphere will

receive less energy than in emits over the absorption band Radiative energy transfer in the

stratosphere will not significantly effect the tropospheric temperature unless a very large

change in the stratospheric temperature occurs

As far as atmospheric concentration is concerned, of all the greenhouse gases, CO2, has the

simplest profile because its mixing ratio is nearly constant up to 10-2 mb, corresponding to an

altitude of about 85 km This means that [CO2] decreases exponentially with altitude -

proportional to the air pressure change - throughout the troposphere and stratosphere ,

(neglecting those places effected by direct CO2 emissions and the temperature change with

altitude) The main CO2 absorption band that blocks IR emission from the Earth's surface

occurs at around 15 m (660 cm-1) This band corresponds to the ro-vibrational transitions

associated with the CO2 bending mode The other main CO2band lies outside the emission

spectrum of the Earth and also outside the main emission spectrum of the Sun (see next page)

The average radiative lifetime of CO2 excited in its bending vibration is reasonably well

known It's about one second Normally though not many excited CO2 molecules have the

opportunity to loose energy by radiation Energy is lost instead by collision with O2, N2, H2O,

and CO2 (and O atoms in the upper atmosphere) In the lower atmosphere the frequency of

collisions is so great (and the rate constant for the deactivation process is sufficiently large)

that essentially all absorbed photons end up as heat Absorbed IR photons are definitely not

reflected back to the Earths surface This does not mean that emissions do not occur CO2

molecules are constantly excited and de-excited by collisions Occasionally, a photon is

emitted The emission rate being proportional to the steady-state population of excited levels

according to the Boltzmann distribution

The figure above shows the main absorption/emission lines of CO2 in the bending transition

If you are not familiar with the form of this spectrum, I suggest you look it up in a standard

infrared spectroscopy text book

Only a very high resolution spectrum of the Earth's emission can reveal the true nature of

radiative forcing In fact, the spectrum above is still not sufficiently high, it therefore does not

reveal many important details The first thing to notice, is that the spectrum appears similar a

Planck spectrum but with various portions diminished Each diminished portion corresponding

to a combination of reduced transmission from the Earths surface (sometimes reduced to zero)

and emission from cooler gases from various altitudes In other words, this is not really an

absorption spectrum at all (despite what is mentioned in many explanations of global warming

that use such a spectrum)

There are two broad regions in the plot above though that do follow the Planck curve - though

intermittently - for the temperature of the Earth's surface (in this case 278 K) These regions

are collectively known as "the atmospheric window", through which radiative cooling via

emission from Earth's surface directly to space occurs This accounts for only 40 Wm-2

compared to the 390 Wm-2 that are emitted from the Earth's surface

As you can see, water vapour, CO2, and O3 account for most of the spectral features The

saturated lines of H2O also appear to follow a Planck function corresponding to 215 K, the

temperature of the Tropopause This is, in fact, quite accurate since the concentration of water

vapour in the stratosphere is so low that insignificant absorption takes place, so it appears that

most emission originates from the Tropopause Contrary to appearances above, emission from

CO2 occurs from all heights of the atmosphere, as will be discussed later The saturated bands

of ozone are at a higher temperature than those of water vapour since the main ozone layer is

in the stratosphere at a higher temperature than that of the Tropopause

Radiative forcing is concerned with both absorption of infrared radiation and emission As we

have already seen, a solid filter also emits radiation according to it temperature Solids are

normally good black-body radiators because they have a very high density of states and

transitions between these states cover effectively a continuous energy range A gas though has

a relatively few energy levels per unit energy range and it is therefore cannot produce a

continuous black-body emission spectrum But how much power (J s-1) is emitted by a

particular transition? According to what has been mentioned before about gasses in the

atmosphere we would expect the power to be the same as that corresponding to a black body

radiator, but this time only at a single wavelength (see the diagram above) If this is really the

case, then another transition at a different wavelength should have a different power But if

considering a single molecule only, this picture does not make sense It is more likely in the

above picture that the longer wavelength transition is the most intense owing to its greater

thermal population (it has a lower energy), if all other factors are equal So why does one

expect a body of gas to emit less power at the longer wavelength than at the shorter

wavelength in this case? This will only happen if the lines are saturated The phenomenon of

spectroscopic line saturation is dealt with on the next page

Here we look at what line-saturation means, from the perspective of absorption cross-section

Remember, absorption cross-section is related to absorption coefficient and also the transition

probability between to states

Imagine that a gas is placed inside a long square tube that has a cross-sectional area of 1 m2

The length of the tube is not specified If you look down the tube using one particular

wavelength you would in principle (but not in practice) observe the individual molecules

moving around and having an area that is not their physical (collisional) area, but an area

corresponding to the absorption cross-section (at that particular wavelength) of the molecule

Is this picture correct ? Let us see Suppose we have 1 x 1010 molecules per cm3 in a box that

that is 1m long and 1 m2 in area If you were to look through the length of the box, how much

of the area would the molecules occupy if each molecule had an absorption cross-section of 1

x 10-21 cm2 ? To answer this we initial assume that the change of one molecule being in front

of another molecule is negligible In other word, we can see all of the molecules if we look

down the box How many are there? There are 1 x 1016 cm-3 x 100 cm (length) x 1002 cm2

(area) = 1 x 1020 molecules Each of these has an area, according to the photon, of 5 x 10-20

cm2 So the total area covered by the molecules is 5 x 10-20 cm2 x 1 x 1020 = 500 cm2 The total

area of the box is 1002 cm2 = 1 x 104 cm2 So the fraction of light that can pass through is

(1-500/1 x 104) = 0.95 exactly Know let us use Beer-Lambert Law : Itr = Io x exp (-5 x 10-20 x 1 x

1016 x 100) = 1 x exp(-0.05) = 0.951 So this picture is quite accurate (For large cross-sections

or large concentrations the assumption that molecules do not 'hide' behind each other is not

correct)

line, the stronger is the absorption line

Image now that the molecules displayed above have only a single transition spread over the 10

- 11 micron range (assume the line shape is rectangular) According to the black-body curve

on the previous page, the total emission from this gas (assuming it to be a perfect black-body

radiator) should be 25 Watts Actually it is only half this value at 12.5 Watts becuase the

observed surface area of the gas is only 0.5m2 That is, only half of the box appears to be

filled This means that a layer of gas that absorbs 50% of the light passing though will only

emit 50% of the power of a black body at its emitting wavelength If the concentration of the

gas increases, or the length of the tube is increased (second box above) all of the radiation

passing through is absorbed The power emitted from this surface is now exactly the same as a

black-body radiator But two pages previously, an example was given of a molecule that had

two emission lines According to what has just been said the intensity of these two emission

lines from the body of gas as a whole must now be in a ratio dictated by Planck equation and

independent of their molecular emission intensity How does this actually occur? The

argument goes as follows: suppose the longer wavelength emission intensity on the molecular

level is greater than the short wavelength emission intensity (opposite to what is seen in the

Planck equation) If the emission line is strong, so is the absorption line This means that

emission from molecules further back in the box will be absorbed by the ones near the front so

only relatively few molecules contribute to the emission (such as depicted in the third box)

For weaker emissions, the light from molecules further back will make it through to the front

with out being absorbed, so the overall effect end up being the same That is, you see the same

amount of red from the second box and the third box independent of the size of the absorption

cross-sections (emission probabilities) On that basis then all emissions under saturation

conditions should have same intensity But one also need to consider two other factors: the

first is the Boltzmann population that gives lower levels a larger population than higher ones

This causes the rise in this Planck function on the left The second factor is more complex and

will be here just stated The overall transition intensity per unit wavelength for a fully allowed

transition decreases with increasing wavelength This causes the fall off at larger wavelengths

These two factors are responsible for the shape of the Planck spectrum

So, under saturation conditions, a body of gas (at a single temperature) will emit spectral lines

in the ratio dictated by the Planck equation

The emission spectrum of the Earth as observed from space has two components (remember also that there is reflection of 31 % of the incoming Sun's radiation, which is observed mainly

in the visible and UV spectral region) The first is emission originating at the Earth's surface The second is emission originating from various levels of the atmosphere This page shows a very simple picture in which there is a single layer of absorbing gas at 220 K that has rectangle absorption lines The gas possesses two absorption bands, one band is fully saturated, which means it absorbs all emission from the Earth's surface The other band only absorbs 50% of the emission from the Earth's surface This situation is given in the top most figure that shows only that component of emission from the Earth's surface as would be seen from space

The layer of gas also emits radiation The saturated band emits according to a black-body at

220 K over its band range The second band emits 50% of a black body spectrum, as discussed

in the previous pages The combined effect is given in the figure on the right This would be the overall emission to space

As you will see, the real situation as somewhat more complicated because the single absorbing layer must be replaced by a continuum of absorbers varying in concentration and temperature throughout the atmosphere, but the essential mechanism is the same

For these cases, the temperature of the earth would not change at all as the rate of emission

from the top of the CO2 layer (over the CO2 absorption bands) would be equal to the rate of

emission from the Earth’s surface You would not be able to observe a molecular spectrum

from space

Here is a more detailed picture of the balance of radiation On the left, in yellow, is represented

the incoming radiation from the Sun (mainly in the visible and uv spectral regions) Some 31

% is reflected to space; the greater part being reflected from the atmosphere (clouds and

aerosols mainly) Though a significant fraction of the remaining light (67 Wm-2) is absorbed in

the atmosphere – mainly by O3and O2below350 mn– the greatest flux is absorbed by the

Earth’s surface (168 Wm-2), which comprises direct radiation and scattered radiation from

clouds and particles as well as Rayleigh scattering from molecules This absorption causes air

motion (expending about 24 2) and also evaporation of water (expending a huge 78

Wm-2), but both of these are internal mechanisms and therefore not lost from the troposphere, but

expended eventually as heat (to the troposphere) So, in fact, only 66 Wm-2 from the sun goes

into heating the Earth’s surface, but (66 + 78 + 24 =138) Wm-2 is deposited in the troposphere

by the sun - all of this causes heating Additionally, the troposphere will receive (67/2 = 33.5)

Wm-2 of ‘band’ IR radiation from the stratosphere That is, half of the deposited energy going

up and half going down Thus, initially, the Earth + Troposphere system would radiated 171.5

W m-2 in the infrared

As you will see on the following pages, absorption of this radiation by cold infrared – active

molecules in the upper troposphere will lead to a net deposit of energy in the troposphere

resulting in further warming A balance is eventually reached when the Earth’s surface is at

288 K, implying an emission rate of 390 Wm-2 So, in equilibrium, only 40 W m-2 makes it

from the Earths surface directly to space in spectral regions where it is not attenuated (i.e in

regions of the atmospheric window)

less than 195 Wm-2 originates from the atmospheric molecules What these two numbers give

is the proportion of the energy spectrum that is effectively unaffected by molecular absorption,

to the proportion that is effected, as discussed 5 pages previously

In this picture, the downward arrow representing 234 Wm-2 is redistribution of energy by

convection and conduction, not by radiation Is important to realise that IR radiation

transmitted from the upper troposphere to the Earth’s surface has no net heating effect and is

only a consequence of the troposphere being heated

Ice creams don’t melt faster under a room temperature roof than if your standing under cold

clouds It’s the air temperature that counts if there is no direct sunlight

The height of the emission from gasses in the atmosphere is not actually very well defined, since at each wavelength there is always a contribution to the observed intensity in space from various layers

Calculating the height at which a molecule essentially emits light to space is critical in

calculating the radiative balance of the Earth This information can be partly gained from

satellite IR spectra, which indicates the temperature at which an absorption occurs

Previously it was demonstrated that one could visualize the altitudes at which emission occurs

to space by applying a step-wise procedure in which the amount of light transmitted from each

altitude is calculated separately A simpler, but entirely equivalent, method is to concentrate on

the amount of light absorbed from above This shows directly how much light can be

transmitted from beneath So, for the example given (middle plot) no light is expected to reach

space emitted directly from 30 km or below About 50% of the light emitted from 70 km is

expected to make it too space (100 km, say) Notice, that even though nearly all photons

emitted above 80 km are expected to make it to space, one must consider the how many

emitting species there are This really follows the reasoning on page 24-25 The place were

most emission occurs must correspond to the place where most absorption occurs (in the

downward direction) So a simple integration of the downward absorption profile reveals the

height at which emission occurs For each wavelength, the height and distribution will be

different

This page shows the main sources and sinks of CO2 in the atmosphere Background

concentrations of CO2 are relatively high: over the last several thousand years it has remained fairly constant at around 270 ppm according to ice core data Most CO2 is thought to be emitted from the oceans and from soils and vegetation Over periods of hundreds of thousands

of years, [CO2] undergoes some oscillation which seems to correlated with temperatures changes: the greater the temperature the greater is the CO2 concentration Evidence so far shows that these changes of CO2 are in response to temperature rather than the other way around, and that these changes mostly occur due to changes in uptake and emission of CO2 over the oceans and not over land Carbon-isotope evidence from marine sediments, and terrestrial carbon, based on pollen data from terrestrial sediments, suggest that terrestrial carbon storage was negligible during some major glacial periods, implying that the extra stored carbon during these cold periods was in the ocean and not on land Numerous oceanic

processes have been identified that could have contributed to the low glacial concentrations of CO2 One explanation conceived in the 1980s, and recently supported by nitrogen-isotope data, invokes a more efficient utilization of macronutrients in the southern ocean during glacial times, leading to higher rates of carbon export from the surface and thus to increased carbon storage at depth, reducing the equilibrium concentration of CO2 at the ocean surface