Title: Lecture 16: Atmospheric chemistry
1Lecture 16 Atmospheric chemistry
- Questions
- How do solar forcing, radiative and convective
transfer set the vertical temperature structure
of the atmosphere, the latitudinal heat transport
by the atmosphere, and the global wind patterns
that drive ocean circulation? - How does the greenhouse effect work?
- Whats up with the ozone layer?
- Tools
- Gas phase chemistry, radiative and convective
heat transfer, box models, photochemistry, etc. - Reading
- Not well-treated in either Albarède or Press et
al., but some issues are raised in Press et al.
chapter 23 - A good short book is Daniel Jacob, Introduction
to Atmospheric Chemistry
1
2Atmospheric structure 0-D
- Radiative forcing the atmosphere is heated from
above by UV absorption in stratosphere and from
below by IR absorption in troposphere. Most
sunlight (visible peak) gets through to the
ground. A significant fraction (75) of the IR
is absorbed and re-radiated at lower temperature.
Outgoing radiation
Incoming radiation
2
3Atmospheric structure 0-D
- Radiative balance incoming radiation outgoing
radiation - Incoming radiation FSpr2(1- a)
- Solar flux at 1 AU, FS 1380 W/m2
- Area receiving sunlight is area of Earth
projected as a disk, pr2, where r 6471 km. - Albedo of earth a 0.3 (where aice1)
- Outgoing radiation 4pr2sTE4
- Area radiating is surface area of sphere, 4pr2
- TE is the effective blackbody temperature, s is
the Stefan-Boltzmann constant - So TE FS(1- a) / 4s1/4 255 K 18 C
- So if the Surface temperature of the Earth were
the effective radiating temperature (i.e., no
atmosphere), all water would be frozen. - To raise TE to 273 K by lowering albedo alone
would require a 0.08!
3
4Atmospheric structure Greenhouse effect I
- Now imagine an atmospheric layer that is
transparent to incoming solar radiation but
absorbs a fraction f of outgoing infrared
radiation. - Now we write two independent radiative balance
equations, for the surface at temperature To and
for the absorbing layer at T1
- Per unit area, FS (1- a)/4 (1- f)sTo4 fsT14
from space - Per unit area, 2fsT14 fsTo4 at absorbing layer
(top and bottom both radiate, hence the 2 we
used Kirchhoffs law el al and a greybody
assumption) - So To FS(1- a) / 4s(1-f/2)1/4 and T1 To
21/4 - Hence actual mean ground temperature To 288 K
for the earth implies f 0.77 (in which case T1
242 K). The maximum effect from a single layer
would be at f 1 and To 304 K 31 C (in
which case, naturally T1 255 K). Of course
there could be different absorbing layers at
different wavelengths, etc.
4
5Atmospheric structure Greenhouse effect II
- Here is an actual outgoing radiation spectrum
measured over Africa at noon. The ground is
radiating at 320 K in the non-absorbing
atmospheric window. The tropopause (where CO2
becomes optically thin) is radiating at 215 K,
the lower troposphere is radiating at 270 K (H2O
is thin above 5 km). The stratosphere is
radiating at 280 K (where O3 becomes optically
thin)
Of course, this is neither a steady-state nor a
0-dimensional situation, but in some sense the
ground-atmosphere system must adjust itself to
match the integral under this curve to incoming
solar radiation
5
6Atmospheric Structure 1-D
- Pressure structure
- Hydrostatic equilibrium between pressure gradient
and gravity - Ideal gas law (Ma 28.96 g/mol)
- Or, assuming constant T
- The logP-z curve in the figure is not quite
linear because the temperature is not actually
constant
- Let H RT/gMa be the scale height of the
atmosphere (7.4 km at T250 K) - Thus, e.g. a supersonic jet flying at 20 km is a
full scale height above a normal jet flying at 12
km and sees 1/e times the air density in its path
6
7Atmospheric Structure 1-D
- Temperature structure
- There are three reversals in the average
temperature profile of the atmosphere that divide
it into four layers - The Thermosphere, above 80 km (not shown in
figure), gets very hot due to UV absorption by
O2, but the density is so low it hardly matters - The Mesosphere is heated from below and has
decreasing T with altitude - The stratosphere is heated from above by UV
absorption by ozone. It is stably stratified. - The troposphere is heated by IR absorption by CO2
and H2O and may become convectively unstable.
9.8 K/km
- Convective stability is defined by the
temperature gradient relative to the adiabatic
lapse rate. For dry air
a 1/T for ideal gas
- For saturated air, the moist lapse rate is more
like 6 K/km
7
8Atmospheric structure 2-D
8
- The Earth is unevenly heated by sunlight the
equator receives much more radiation per unit
area than the poles - It is the job of the atmosphere and oceans to try
to eliminate the resulting temperature gradient
by zonal heat transport - The resulting transport is of two types ocean
transport is dominantly sensible heat transport
(advection of warm water polewards), atmospheric
transport is dominantly latent heat transport
(low-latitude evaporation, high-latitude
condensation)
9Atmospheric structure 2-D
- Total zonal heat transport is obtained from
radiative balance calculations based on solar
forcing and measured outgoing IR as a function of
latitude (see Problem Set 6) - Atmospheric heat transport is obtained from
Radiosonde data that give abundant regular
measurements of temperature, winds, and humidity - Oceanic heat transport is obtained by difference,
but shows important features such as Western
Boundary currents in North
9
10Atmospheric structure 3-D
- In the absence of Coriolis force, solar forcing
would drive single Hadley cells in each
hemisphere, which we can understand using the
sea-breeze circulation
10
11Atmospheric structure 3-D
- But by 30 latitude, the Coriolis force gets
strong enough to break up the Hadley circulation,
resulting in subtropical oceanic gyres, tropical
rainfall, the 30 desert band, trade winds, etc.
Remember the geostrophic equation?
11
12Atmospheric structure 3-D
- But by 30 latitude, the Coriolis force gets
strong enough to break up the Hadley circulation,
resulting in subtropical oceanic gyres, tropical
rainfall, the 30 desert band, trade winds, etc.
Remember the geostrophic equation?
12
13Bulk chemistry of atmosphere
- To first order, the modern atmosphere originated
by degassing of volatile compounds from the
earths interior. This process continues, as
demonstrated for example by the 3He flux at
mid-ocean ridges
13
14Bulk chemistry of atmosphere
- What comes out of the Earth CO2, H2O, S, N2,
noble gases - What is now in the atmosphere
- 78.08 N2
- 20.05 O2
- 0.9 Ar
- 275 380 ppm CO2
- 0.0005 He
- 0.00005 H2
- Why are they different?
- H2O condenses. CO2 dissolves in oceans (60x more
than atmosphere) and precipitates as carbonates. - Noble gas in atmosphere is dominantly radiogenic
(40Ar, 4He) - H2 is lost from exosphere (He/H2 ratio 10 is
1000x primordial ratio) - O2 is produced and maintained by biology
14
15Geochemical cycles Nitrogen
- Here are the basic elements from which we might
construct a box model to understand the cycling
of Nitrogen in the surface reservoirs of the
Earth
15
16Geochemical cycles Nitrogen
- Here is a steady-state quantification of the N
box model
t 13 Ma
t 0.03 a
t 27 a
t 0.6 a
t 4 a
t 50 a
t 200 Ma!
16
17Geochemical cycles Oxygen and Carbon
- To make atmospheric oxygen, it is not enough to
have photosynthesis, because respiration and
decay of organic carbon take the oxygen back to
CO2. - Rather, each mole of oxygen in the atmosphere
must be compensated by a mole of buried organic C
in sediments - But the total inventory of sedimentary organic C,
about 107 Pg, is enough to account for 30 times
the atmospheric inventory of O2! - Think about this next time you burn fossil fuel,
but dont think too hardthe industrial increase
in CO2 from 280 to 380 ppm represents a decrease
of O2 from 20 to 19.98 - The balance is accounted for by burial and
storage of SO42- and Fe2O3, since the mantle
provides mostly S2- and FeO.
17
18Geochemical cycles Carbon
- The important greenhouse gases are CO2, CH4, and
H2O (but H2O is a passive amplifier, not a
cause), so global climate is intimately tied to
the carbon cycle
About this time Wally Broecker sent an alarmist
letter to President Nixon about global cooling
18
19Geochemical cycles Carbon
- Proxy records allow longer reconstructions than
instrumental data...
19
20Carbon
- We have accurate measurements of the increase in
atmospheric CO2 concentrations. - We can estimate the effect on climate forcing.
- CO2 is the biggest climate forcing, but many
others are significant. This is the 2001
assessment by the IPCC
20
21Carbon
- We also know from economic records the total
amount of fossil fuel burned, and only about half
the resulting CO2 has accumulated in the
atmospherewhere is the rest?
- Taken up by ocean and by terrestrial biosphere,
but how much of each? - One good way to tell is from simultaneous
high-precision data on O2 and CO2 - Fossil fuel, a mix of coal, gas, and oil,
consumes 1.38 mol O2 for every 1 mol CO2 released - Land uptake by photosynthesis is 11,
CO2H2OCH2OO2 - Ocean uptake by solubility and pH adjustment has
no effect on O2 - Ocean warming lowers O2 solubility, though
21
22Stratospheric ozone production and loss
- The existence of ozone in the stratosphere
determines the temperature structure of the upper
atmosphere and, by the way, is essential for life
at the Earths surface. - It is therefore worthwhile to understand the
chemical kinetics of production and loss and the
effects of anthropogenic gases.
O3 photolysis
O2 photolysis
23Stratospheric ozone production and loss
- Production of ozone in the stratosphere is well
understood the mechanism was defined by Chapman
in 1930
(activation reaction, l 240 nm, both oxygens
are O(3P) )
(1)
(M is any 3rd body reactions 2 and 3 make a
rapid cycle that defines the odd oxygen or Ox
family, l 320 nm)
(2)
(3)
(This O is O(1D) until some collision)
(4)
(quenching reaction)
We will see that k2 and k3 are much greater than
k1 and k4, so that at steady state there is a
significant abundance of Ox and it does not
matter whether it is O3 or O.
24Stratospheric ozone production and loss
- Steady-state solution for ozone abundance
- Ox steady state means setting rate of reaction 2
equal to 3 - where CO2 is the mixing ratio of O2 (0.2) and na
is the number density of all air molecules
(altitude dependent) - Then steady-state for entry and exit to Ox cycle
means setting rate of reaction 1 equal to
reaction 4
Note the photolysis rate constants k1 and k3
include a term for the ultraviolet flux, so they
increase upwards as the column depth of O2 and O3
above z decreases. On the other hand, the number
density of the atmosphere falls off exponentially
with increasing altitude.
25Stratospheric ozone production and loss
Up here no O2 to react
Down here no UV flux
- Steady-state abundance of O3 depends on product
of k1 and na3/2, so there is a maximum at 30 km.
The general shape of the prediction is a good
match to abundance data. - But the Chapman mechanism predicts a factor of 2
too much O3the source is certain so there must
be another sink!
26Stratospheric ozone production and loss
- The missing sinks for ozone come from catalytic
loss cycles, i.e. reaction cycles where the ozone
destruction agent is regenerated and can destroy
many ozone molecules before it exits the cycle - Good catalysts are generally radical species with
an odd number of electrons such as the hydroxyl
radical OH (9 e) - The OH loss cycle must be initiated by O(1D),
normally produced by k3 photolysis of O3
Activation steps removes one Ox, makes 2 OH,
requires deep UV and H2O
Catalytic cycle net reaction is 2O3 -gt 3O2
(OH and OH2 are the HOx radical family)
Termination step, slow
27Stratospheric ozone production and loss
- The OH loss cycle is efficient in principle but
does not account for enough O3 loss in the middle
and upper stratosphere - Limited at low altitude by low UV flux
- Limited at high altitude by very low H2O mixing
ratio - A more important (but more complicated) natural
catalytic loss cycle (whose discovery earned Paul
Crutzen a Nobel prize) is the NOx radical system
NOx radical species NO and NO2 NOy nonradical
reservoir species N2O5 and HNO3
28Stratospheric ozone production and loss
- When reaction of NO with O3 produces NO2, it has
several possible fates - Photolysis cycles it back to NO with no net
effect - Reaction with O catalytically destroys two Ox
species - Reaction with OH radical or O3 inactivates one NOx
NOx cycle no net effect, but rapidly cycles NO
NO2
Catalytic cycle branch net reaction consumes 2 Ox
Termination step, daytime
Termination step, nighttime
29Stratospheric ozone production and loss
- Because N2O from the biosphere is stable and
non-condensable, it reaches upper stratosphere
and meets enough O(1D) to form NOx and initiate
O3-loss catalysis - The other O3-loss mechanism is mostly
anthropogenic and involves sources of Cl and Br
stable enough to reach stratosphere
Together, the Chapman source roughly balances
these four loss mechanisms and explains the O3
abundance at all heights in the normal
stratosphere Chapman (O3O), HOx, NOx, and ClOx
30Polar Stratospheric ozone the Antarctic Ozone
Hole
- The total disappearance of the ozone layer in the
mid-stratosphere over Antarctica provides a
challenge to the standard gas-phase theory of
ozone balance, since in winter there is not
enough light to drive the HOx, NOx, or ClOx losses
October 2000
October 2002 ?
31Polar Stratospheric ozone the Antarctic Ozone
Hole
32Polar Stratospheric ozone the Antarctic Ozone
Hole
- The story is complicated but here is its essence
- 1) When temperature drops below 197 K Polar
Stratospheric Clouds (PSC) of HNO33H2O can form
even though H2O is very rare. - 2) PSC surfaces provide rapid total conversion of
inactive Cl species HCl and ClNO3 to active ClOx
and HNO3. - 3) When temperatures rise again in September,
the HNO3 would scavenge all the ClOx back to
ClNO3, except that the PSC particles grow big
enough to sediment out of the stratosphere,
removing HNO3 and leaving behind active ClOx. - 4) When light returns in Southern Spring, at high
ClO concentrations a catalytic photolysis
mechanism can run that consumes O3 without O(1D). - (More Nobel-quality chemistry, this time to
Molina and Rowland)
33Tropospheric Ozone
- Yes, the air in Pasadena really is getting better!
33