Title: Ch. 5
1Ch. 5 2nd Law of Thermodynamics
- Entropy and Potential Temperature
- The right hand sides of our equations for
potential temperature and specific entropy are
the same, so we have - Leading to the final result
- Changes in entropy (for a reversible process) are
directly proportional to changes in potential
temperature. - Adiabatic (reversible) processes are isentropic.
2Ch. 5 2nd Law of Thermodynamics
- Entropy and Potential Temperature
- We could write a similar equation for the
extensive variable, S, as - What about an irreversible adiabatic process with
ds gt 0? - For any adiabatic process, d? 0 ? ds for an
irreversible process. - Entropy increase due to irreversible work
(uncompensated heat), such as frictional
dissipation. - Isentropic equals adiabatic, but not always the
other way.
3Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
(following Bohren and Albrecht, 1998) - On a flat earth (an approximation), the total
mass per unit area (kg m-2) between altitudes z1
and z2 is - Assuming hydrostatic equilibrium we can
substitute for the density, ?, to get - Considering an atmospheric layer between 2
constant pressure surfaces, the total mass of the
layer is constant.
4Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Assume the layer is
- Isolated from its surroundings
- Not heated or cooled by direct radiation (solar
or terrestrial) - Not heated by conduction from surroundings or
surface (if in contact) - QUESTION What is the equilibrium temperature
profile in this layer? - Total enthalpy per unit area of layer is
5Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Aside from the influence of the gravitational
field, the total enthalpy would be conserved. - Because of gravitational field, the layer has
potential energy, given by - Recalling that the units on cpdT are (J
kg-1K-1)(K), which J kg-1 m2 s-2, and the
units on gz are (m s-2)(m) m2 s-2, also J kg-1,
the enthalpy is an energy equivalent.
6Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Given the assumptions concerning isolation, the
enthalpy and the potential energy are the only
forms of energy the layer can have. - We can add them up and realize that their sum
must be constant, yielding, - The quantity cpdT ( h) gz is called the dry
static energy (see equation (4.31) in the text).
7Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- We can also calculate the total entropy of the
layer as, - Note that the additive constant has been left off
and we are playing fast and loose by taking the
natural log of something that has dimensions. - The maximization of entropy is addressed by
asking, what temperature and pressure profiles
maximize the layer entropy subject to the
constancy of the dry static energy?
8Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Since we are talking temperature and pressure, we
convert our equations to pressure from height
coordinates. - The dry static energy becomes (using the
hydrostatic equation) - We have reversed the order of integration to get
rid of the minus sign introduced by the
hydrostatic equation, and the extra g has been
included in the constant.
9Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- The equation for layer entropy becomes
- We can also write the relationship between
entropy and potential temperature as
10Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- From the potential temperature equation we have
- Factoring out ?/T (p0/p)?, we have
11Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Simplifying a little further, we have
- Substituting this into our expression for ds/dz,
we get - We can multiply thru by T, take cpd inside the ()
and put the last term in terms of a derivative,
giving
12Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- We can put two equations of interest in
p-coordinates as - From the potential temperature equation we have
an expression for the ratio of T/? (p/p0)0.286. - At p 1000 mb, T/? 1.0.
- At p 200 mb (80 of the atmosphere), T/?
0.631. - If we restrict consideration of atmospheric
layers to those a few hundred meters thick, a few
10s of mb, we can use T/? ? 1.0.
13Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Then we have
- We can integrate this to get
- We had earlier that
- So we can write
14Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- We are almost there.
- We apply a change of variables p to x where
- This transforms the coordinates so that x 0 is
in the middle of the layer, since ½ (p1 p2)
defines the arithmetic midpoint of the layer, and
dx dp. - We can rewrite the last integral as
15Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Lets play with this a little.
- When p p1 (at the bottom of the layer), what is
x? - x gt 0 since p1 gt p2 (pressure decreasing with z).
- When p p2 (at the top of the layer), what is x?
- Here x lt 0, so the sign sense is reversed, but
intervals are preserved.
16Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Now the limits of integration are symmetric about
x, and ? the limit at p1 is given by - So now we want to find the function T for which
the entropy, S (divided by the constant cpd/g),
is a maximum, subject to the constraint indicated
by the previous equation, i.e.,
17Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- For simplicity we will consider only the set of
linear potential temperature profiles, such that - Here a and b are arbitrary.
- Putting this in our constraint equation yields
- Therefore a is a constant. Since S is related to
ln?, once a is fixed, S can only depend on b.
18Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- The condition for S to be an extremum (maximum or
minimum) is that its derivative vanish - Looking at the derivative of the integral (taking
the derivative operator inside the integral), we
have - Since d(lnu)/dx 1/u du/dx. Now we carry out
the integral
19Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- We have to solve
- Solving this yields
20Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Since a is a constant, the only solution for this
equation is if b 0. - We determine whether this is a maximum or minimum
by taking the 2nd derivative with respect to b.
Doing this we get, - Because the integrand is positive definite, the
result is negative, indicating that this is a
maximum.
21Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Of all linear potential temperature profiles, b
0 indicates a constant ?, since ? a and a was
determined to be a constant. - Therefore, ? const maximizes the entropy of an
isolated layer of the atmosphere in hydrostatic
equilibrium. - The dry adiabat delineates between stable and
unstable air. - Entropy maximization requires the equilibrium
temperature of an isolated atmospheric layer
subjected to mixing, but with no condensation or
evaporation, to decrease with height at the dry
adiabatic lapse rate.
22Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- This is not intuitive.
- We might expect the equilibrium profile to be
isothermal. - A solid isolated from the environment with an
initial non-uniform temperature gradient, reaches
a constant temperature by conduction eliminating
gradients. - In the atmosphere, energy transfer is dominated
by convection rather than conduction. - Thus, convective equilibrium is established,
rather than conductive or radiative equilibrium. - Fluids are inherently different from solids.
23Ch. 5 2nd Law of Thermodynamics
- Maximization of Entropy in the Atmosphere
- Mass motion in solids is absent, but it dominates
fluids. - Suppose a parcel moves to another level within
the isolated layer and then mixes with its
surroundings. - Temperature of parcels is not conserved during
vertical motion. - If a parcel moves up (cools) and mixes with air
at that level, the layer is destabilized the
net effect is cooling. - Similarly for a downward moving (warming) parcel.
- Mixing increases the entropy of the layer even
though it is stable. Well mixed layers have an
adiabatic profile.