subharmonic and superharmonic functions
First let’s look at the most general definition.
Definition.
Let G⊂ℝn and let φ:G→ℝ∪{-∞} be an upper semi-continuous function,
then φ is subharmonic if for every x∈G and r>0 such that
¯B(x,r)⊂G (the closure of the open ball of radius r around x is still in G) and every real valued continuous function h on
¯B(x,r) that is harmonic in B(x,r) and satisfies φ(x)≤h(x)
for all x∈∂B(x,r) (boundary of B(x,r)) we have that
φ(x)≤h(x) holds for all x∈B(x,r).
Note that by the above, the function which is identically -∞ is subharmonic, but some authors exclude this function by definition. We can define superharmonic functions in a similar fashion to get that φ is superharmonic if and only if -φ is subharmonic.
If we restrict our domain to the complex plane we can get the following definition.
Definition.
Let G⊂ℂ be a region and let φ:G→ℝ be a continuous function. φ is said to be subharmonic if whenever D(z,r)⊂G (where D(z,r) is a closed disc around z of radius r) we have
φ(z)≤12π∫2π0φ(z+reiθ)𝑑θ, |
and φ is said to be superharmonic if whenever D(z,r)⊂G we have
φ(z)≥12π∫2π0φ(z+reiθ)𝑑θ. |
Intuitively what this means is that a subharmonic function is at any point
no greater than the average of the values in a circle around that point. This implies that a non-constant subharmonic function does not achieve its maximum
in a region G (it would achieve it at the boundary if it is continuous there). Similarly for a superharmonic
function, but then a non-constant superharmonic function does not achieve its
minumum in G. It is also easy to see that φ is subharmonic if and only if -φ is superharmonic.
Note that when equality always holds in the above equation then φ would in fact be a harmonic function. That is, when φ is both subharmonic and superharmonic, then φ is harmonic.
It is possible to relax the continuity statement above to take φ only upper semi-continuous in the subharmonic case and lower semi-continuous in the superharmonic case. The integral will then however need to be the Lebesgue integral (http://planetmath.org/Integral2) rather than the Riemann integral which may not be defined for such a function. Another thing to note here is that we may take ℝ2 instead of ℂ since we never did use complex multiplication. In that case however we must rewrite the expression z+reiθ in of the real and imaginary parts to get an expression in ℝ2.
It is also possible generalize the range of the functions as well. A subharmonic function could have a range of ℝ∪{-∞}
and a superharmonic function could have a range of ℝ∪{∞}. With this generalization, if f is a holomorphic function
then φ(z):= is a subharmonic function if we
define the value of at the zeros of as .
Again it is important to note that with this
generalization we again must use the Lebesgue integral.
References
- 1 John B. Conway. . Springer-Verlag, New York, New York, 1978.
- 2 Steven G. Krantz. , AMS Chelsea Publishing, Providence, Rhode Island, 1992.
Title | subharmonic and superharmonic functions |
Canonical name | SubharmonicAndSuperharmonicFunctions |
Date of creation | 2013-03-22 14:19:39 |
Last modified on | 2013-03-22 14:19:39 |
Owner | jirka (4157) |
Last modified by | jirka (4157) |
Numerical id | 12 |
Author | jirka (4157) |
Entry type | Definition |
Classification | msc 31C05 |
Classification | msc 31A05 |
Classification | msc 31B05 |
Related topic | HarmonicFunction |
Defines | subharmonic |
Defines | subharmonic function |
Defines | superharmonic |
Defines | superharmonic function |