Interpret limit points as subsequential limits, including a value repeated infinitely often. Let be the common nonempty compact cluster set. Boundedness implies and likewise for : otherwise a subsequence staying a fixed distance away would have a convergent further subsequence whose limit both belongs to and stays away from it.
Construct a matching of indices in alternating steps. At step choose the least unused index in the first sequence. Choose a nearest point to and an unused index in the second sequence with . Such indices exist infinitely often because is a subsequential limit. Pair with ; their values differ by at most . At step choose the least unused second-sequence index and match it to an unused first-sequence index within of a nearest point of . This gives the analogous bound with .
Every index is used exactly once: no step reuses an index, and by step at least the first indices in each sequence have been used. Hence the produced lists are permutations satisfying
The indices actively chosen as least unused tend to infinity on their respective odd and even steps. Their distances to therefore tend to zero, and . The two matching estimates prove . The cluster-set convention is important here; accumulation points of the set of distinct values would ignore infinite repetitions.
For the Fredholm convention, define
The equality follows from the Atkinson theorem. The quotient is a unital complex Banach algebra when is infinite-dimensional. Its invertible set is open by the Neumann series, so the spectrum is closed, and it lies in the disk .
For nonemptiness, suppose the quotient resolvent existed for every complex . It is analytic everywhere, and outside a sufficiently large disk its Neumann expansion gives . Continuity bounds it on the remaining compact disk. Every continuous linear functional applied to is therefore a bounded entire scalar function, constant by Liouville's theorem and zero because it tends to zero at infinity. The Hahn-Banach theorem, or separation by the dual, forces , contradicting its inverse identity. The essential spectrum is nonempty and closed.
Infinite dimension is necessary: every operator on a finite-dimensional Hilbert space is Fredholm, so its essential spectrum in this convention is empty. The proof works on nonseparable infinite-dimensional spaces as well; it needs neither self-adjointness nor a sequence basis.
The Weyl-von Neumann theorem states that on a separable Hilbert space a self-adjoint operator is a diagonal self-adjoint operator plus a compact self-adjoint operator. For a bounded operator , the perturbation can in fact be chosen Hilbert-Schmidt with arbitrarily small Hilbert-Schmidt norm. The separable-space formulation is also recorded in Kadison's Corollary B.
Here is a constructive proof. The bounded self-adjoint spectral theorem decomposes a separable into countably many cyclic reducing spaces, on each of which is multiplication by in for a finite measure supported in a bounded real interval. First work on one such space. Partition the interval into intervals of length at most , and repeatedly bisect them. Let be the finite-dimensional space of functions constant on the level- intervals. These spaces are nested and their union is dense, since dyadic step functions approximate continuous functions and continuous functions are dense in .
Choose normalized interval indicators as a basis of , and choose orthonormal bases of within each parent interval. This is possible because the refinement splits into orthogonal parent blocks. Each new basis vector at level is supported in an interval of length at most , and there are at most such vectors. For every basis vector choose a real in its supporting interval and define the diagonal operator . Then
so
One may take , so this bound tends to zero as . The error is Hilbert-Schmidt, hence compact; it is self-adjoint because both and are.
On the countably many cyclic summands choose error bounds with . Combining their orthonormal bases and diagonal operators gives with , proving the assertion. For an unbounded self-adjoint operator, first split by spectral intervals , apply the bounded construction on each, and choose summable error bounds. The resulting total is bounded compact, and has the same domain as ; the diagonal domain condition follows from the boundedness of .
Separability matters. Take an uncountable orthogonal sum of copies of multiplication by on . The range closure of a compact operator is separable, and a separable subspace of this direct sum is supported in only countably many summands: choose a countable dense set and unite its countable supports. A self-adjoint compact error therefore vanishes on every remaining summand, each of which reduces the proposed diagonal operator . The restriction of a diagonal self-adjoint operator to a reducing subspace has pure point spectrum, whereas multiplication by on has no eigenvectors. This is a contradiction, qualifying the theorem if the unspoken separability convention is dropped.
Apply part (c) to the commutative unital C-star algebra , commutative because is normal. A character is determined by , since generate the algebra and . The spectrum identity in part (c) shows that every occurs. Thus is a continuous bijection from a compact space to a Hausdorff space, hence a homeomorphism. Spectral permanence identifies this spectrum with the operator spectrum.
For any complex polynomial in two commuting variables, evaluation gives . Again using the spectrum identity and spectral permanence,
This proves the requested spectral mapping theorem, including constants, and explains why conjugation in the second coordinate is required.
Let be the character space of an algebra, consisting of nonzero multiplicative complex linear functionals on . We construct the required Gelfand transform and prove its properties.
Characters exist. A maximal proper ideal exists by the axiom of choice and is closed: if it were dense it would contain an element within distance one of the identity, hence an invertible element, contradicting properness. The quotient is a complex unital Banach division algebra, and the Gelfand-Mazur theorem identifies it with , producing a character. The same argument applied to the proper ideal generated by proves
Every character has norm one, since its value belongs to the spectrum and is bounded by , while . The character equations are weak-star closed in the dual unit ball. Thus Banach-Alaoglu theorem makes compact Hausdorff; separability of makes it metrizable.
The algebra is inverse-closed in . If is invertible there, for some . For , the Neumann series for converges in , since the positive contraction in brackets has norm at most . Therefore . This proves spectral permanence for C-star algebras in the present concrete setting. In particular a self-adjoint element has real spectrum in , so characters take real values on it. Decomposing into self-adjoint real and imaginary parts gives .
Define . This is a unital star-homomorphism to . Every element of the abelian star-algebra is normal, and the C-star identity gives . The spectral radius formula therefore gives . The displayed spectrum identity now yields , proving isometry and injectivity. Its image is closed because is complete, contains the constants, is closed under conjugation, and separates distinct characters. The Stone-Weierstrass theorem makes that image dense in , hence equal to it. isometrically as a star-algebra.
For a bounded self-adjoint , nonnegative spectrum is equivalent to . Here is a useful elementary justification of the nontrivial direction. Let and , so has nonnegative quadratic form. Cauchy-Schwarz for that positive form gives . Taking unit vectors approaching the infimum makes , hence . Thus nonnegative spectrum forces . The reverse direction follows from the lower bound and adjoint argument for when , together with reality of the self-adjoint spectrum.
If its positive square root is zero. Otherwise set and , so and . Write the scalar binomial series
The last equality follows by taking in this positive-coefficient series. Consequently
converges absolutely in operator norm. Each partial sum is positive, because and the sum of its coefficients is at most one. The limit is positive and self-adjoint. Multiplying the absolutely convergent series gives .
For uniqueness let be another positive square root. It commutes with , hence with the polynomial-norm-limit . Thus . The kernel of is : zero quadratic form for a positive operator implies it annihilates the vector, by positive-form Cauchy-Schwarz. The difference vanishes there and on ; the latter is dense in the kernel complement because is self-adjoint. Therefore . The positive square root of an operator exists and is unique.
The finite-dimensional Krein-Milman theorem says that a nonempty compact convex set is the convex hull of its extreme points. A point is extreme if a proper convex combination of two points of equals it only when both points equal it.
Prove the theorem by induction on the dimension of the affine hull. Dimension zero is immediate. In positive dimension work inside that affine hull, where has interior. Every boundary point lies on a supporting hyperplane. To see this, choose exterior points and nearest points . Convexity and the minimum-distance condition give for . A subsequence of the normalized nonzero vectors converges to a unit vector , while , giving . The resulting face has smaller affine dimension. Its extreme points are extreme in , since equality in the supporting inequality forces both endpoints of any decomposition to lie in . Induction expresses as a convex combination of extreme points of .
Every interior point is a convex combination of the two endpoints of a line segment through it maximal in ; compactness gives the endpoints, and both are boundary points. Combining their convex decompositions completes the proof. In the generalization requested, a nonempty weak-star compact convex subset of the dual unit ball of a separable Banach space satisfies
The closure is now essential; separability makes this compact weak-star setting metrizable but the general Krein-Milman theorem holds more broadly in locally convex spaces.
Start with for some real . The equation becomes
Smooth belongs to every periodic Sobolev space, and multiplication by the smooth coefficient preserves by part (c), so the right side lies in . The Fourier multiplier has coefficient and maps isometrically to . Thus . Repeat this argument to get for every nonnegative integer . For any derivative order , choose with and apply part (b). Therefore . This is elliptic bootstrapping directly from the Fourier multiplier, with no assumption that the solution initially has nonnegative Sobolev order.
For every real , the elementary weight estimate
follows from , using the reverse rearrangement for negative . If is smooth its Fourier coefficients decay faster than every power, so is summable. The Fourier convolution formula and Young convolution inequality give
Initially this holds for trigonometric polynomials, and density extends it to every , agreeing with distributional multiplication. Hence smooth multiplication preserves every order.
If , the stronger estimate gives
using the summability estimate from part (b) with . Approximation extends the product, and the continuous embedding identifies it with the pointwise product. Completeness from part (a), this estimate and the constant unit make a commutative Sobolev algebra for . Rescaling the norm by if necessary gives an equivalent strictly submultiplicative Banach-algebra norm.
Write . There are lattice points in the dyadic shell , while each contributes at most a constant times . The geometric series converges when , proving . For and , the Cauchy-Schwarz inequality yields
Thus every derivative Fourier series through order converges absolutely and uniformly. Starting with the continuous sum for , uniform convergence of successive derivative series justifies differentiation, so these are its continuous classical derivatives. , with a continuous norm bound.
Use with normalized Haar measure and Fourier basis , . For any real , the periodic Sobolev space is
Weighted Fourier coefficients identify it isometrically with , so it is a Hilbert space and complete. This definition includes negative orders and distributions. For nonnegative integer the norm is equivalent to the sum of squared L2 norms of weak derivatives through order .
Pass to . Closed invariant subspaces and intertwining maps are unchanged by this closure, and is still dense. Because is a star-algebra, every closed invariant subspace reduces it.
Every nonzero nondegenerate compact-operator representation contains an irreducible reducing subspace. Indeed choose a nonzero positive compact element ; an isolated nonzero eigenvalue gives a nonzero finite-rank spectral projection by continuous functional calculus, using a function vanishing at zero. The finite-dimensional corner has a minimal nonzero projection , with . Choose a unit vector and set . Then , since is scalar times . If is the projection onto a reducing subspace of , it commutes with , and . Its projection property makes either zero or . Cyclicity then makes or . Thus is irreducible.
Take a maximal orthogonal family of such irreducible reducing subspaces, by the axiom of choice. If their orthogonal complement were nonzero, its restricted representation would be nondegenerate: a vector annihilated by all restricted would, by taking adjoints, be orthogonal to and hence zero. The image of on this complement is a norm-closed C-star algebra of compact operators, so the preceding construction gives another irreducible summand, contradicting maximality. The family therefore exhausts .
Finally, suppose one irreducible equivalence class occurred infinitely often. Choose some acting nontrivially on that class, and a unit vector with . Transport by unitary intertwiners into distinct equivalent summands. These vectors are orthonormal; their images under are orthogonal and all have the same nonzero norm. That contradicts compactness of . The decomposition has finite multiplicity for every irreducible type, as claimed in nondegenerate representations by compact operators.
The Von Neumann double commutant theorem states that for a unital star-subalgebra ,
The commutant of an operator algebra consists of operators commuting with every element. Each commutation equation is closed in the weak operator topology, so is weakly closed and contains both indicated closures.
Let and fix . On let act diagonally and put . The subspace reduces the diagonal star-algebra. Its orthogonal projection therefore commutes with every diagonal . Writing as an operator matrix shows that each entry belongs to . Since , the diagonal commutes with . Unitality gives , so and . Thus . By definition of , for any a single satisfies . These are precisely the finite-vector neighbourhoods of the strong operator topology. Hence belongs to the strong closure, completing the proof. Unitality, or the corresponding nondegeneracy hypothesis, is essential to the theorem's stated form.
Apply part (a) to the compact positive operator . Its positive square root is defined by multiplying each eigenvector of eigenvalue by and vanishing on the kernel. The eigenvalues tend to zero, so is compact. For every , .
Consequently is a well-defined isometry on . Extend it to the closure of that range and set it to zero on its orthogonal complement. This gives a partial isometry with initial space , final space , and
For uniqueness, the positive factor in any such polar decomposition of a bounded operator satisfies . The positive-square-root uniqueness proved in 4(b) fixes . The relation then fixes on the dense subset of its initial space and the kernel convention fixes it elsewhere. This proves uniqueness with the canonical initial-space condition. Without that condition the word “unique” would be false: can be written with many different unitary .
The spectral theorem for compact self-adjoint operators says that a compact self-adjoint has an orthogonal decomposition
with real nonzero eigenvalues of finite multiplicity. Only finitely many eigenvalues, counting multiplicity, have absolute value at least any given . Thus the nonzero eigenvalues form a finite list or a sequence tending to zero, and in operator norm. The kernel may have arbitrary Hilbert dimension.
For existence of an extremal eigenvector, self-adjointness gives . One elementary proof of the nontrivial inequality uses polarization: if the quadratic form is bounded in absolute value by , its polarized real part is bounded by on two unit vectors; adjusting a phase bounds by . Choose unit with quadratic values tending to or . Then
If , compactness supplies a convergent subsequence of , and the last estimate supplies a convergent unit subsequence of , with eigenvalue .
Eigenvectors for distinct eigenvalues are orthogonal, since . An infinite orthonormal sequence in one nonzero eigenspace would violate compactness. More generally an infinite orthonormal family with eigenvalues bounded away from zero has images separated in norm, again impossible. Take all nonzero eigenspaces and let be their closed span. Its complement reduces ; if the restriction there were nonzero, the extremal-eigenvector argument would produce another nonzero eigenvector, a contradiction. Thus . The finite truncations of the eigenvalue sum have remainder norm tending to zero, proving the asserted norm expansion.
On the Hardy space of the circle , let be the projection onto nonnegative Fourier modes and . The Toeplitz index theorem for continuous symbols states: for continuous , is Fredholm exactly when never vanishes, and then
with counterclockwise winding positive.
First, is compact. For Laurent-polynomial symbols the negative-mode crossing involves only finitely many modes and is finite rank. Uniform approximation and extend this to continuous symbols. If is nonvanishing, is therefore a two-sided parametrix, and part (b) gives the Fredholm property.
Writing , a continuous argument along the parametrized circle increases by . Removing this increase gives a periodic continuous logarithm of , so . The symbols give a continuous Fredholm path from to . Part (c) makes their indices equal. For , is the th unilateral shift, with zero kernel and -dimensional cokernel. For negative it is the corresponding backward shift, with kernel dimension and zero cokernel. In both cases the index is .
For necessity, if use the normalized Hardy kernels
They have norm one and converge weakly to zero, as their inner products with each Fourier basis vector tend to zero. Their squared moduli are Poisson kernels concentrating at , so continuity gives and hence . A Fredholm parametrix would imply in norm, because a compact operator takes this weakly null bounded family to zero in norm. This contradicts unit norm, proving the converse.
Fix a Fredholm operator . Split its domain as and its target as . For a sufficiently small norm perturbation , write
where remains invertible by the Neumann series. Invertible triangular row and column operations reduce this matrix to
The second block maps the finite-dimensional space into . Hence the perturbed operator has closed range, finite kernel and cokernel, and
since the rank of that finite block cancels from the difference. Thus the Fredholm set is open and its index is locally constant, hence norm continuous.
If are Fredholm, their product is Fredholm by multiplying their invertible images in the Calkin algebra. For index additivity use the following finite-dimensional exact sequence:
Here is reduction modulo , the induced -map sends to , and the last map is the quotient by the larger range . These maps are well-defined. For example, is the image of the preceding map. If , subtracting a suitable makes , proving exactness at . The other positions follow directly from the definitions. Alternating dimensions in this exact sequence gives
For a Fredholm operator , define to invert and to vanish on the range complement. The bounded inverse theorem makes bounded, and
The errors are finite rank, so is invertible in the Calkin algebra.
Conversely suppose and for bounded and compact . On , is the identity. An infinite-dimensional closed kernel would contain an orthonormal sequence, whose images under have no convergent subsequence, contradicting compactness. Thus the kernel is finite-dimensional.
There is a positive lower bound for on . Otherwise choose unit vectors there with . Since , compactness gives a norm-convergent subsequence with unit limit . The limit lies in the kernel complement and has , a contradiction. This lower bound proves closed range: if converges, remove each kernel component; the remaining form a Cauchy sequence and converge to a preimage of the limit.
Finally adjointing gives . The same kernel argument makes finite-dimensional. Therefore is Fredholm. , the Atkinson theorem. The finite-dimensional case also satisfies the parametrix formulation directly, though its quotient algebra is the zero algebra.
A bounded operator is a Fredholm operator when its range is closed and both and have finite dimension. Its Fredholm index is
For index zero, choose an isomorphism and extend it by zero on . The restricted map is a bounded isomorphism, by the bounded inverse theorem. The finite-rank operator fills exactly the missing kernel-to-cokernel block. Thus is invertible and is an invertible operator plus a compact operator.
Conversely suppose with invertible and compact. The Atkinson theorem proved in part (b) makes every , , Fredholm: its quotient image is always . The block argument in part (c) proves local constancy of the Fredholm index, so its index is constant along this interval and equals . This invokes the actual parametrix and block proofs below, not an unproved perturbation assertion. The claimed equivalence holds.
A hydraulic control of a zero-PV rotating channel is a critical section separating the upstream branch, where long-wave disturbances can propagate upstream, from the downstream branch, where they cannot. In the steady specific-energy description it is a stationary point of with respect to depth at fixed volume flux and local width. The control therefore fixes the discharge for a given upstream Bernoulli function and channel geometry; an arbitrary downstream condition cannot generally be imposed independently on the upstream reservoir.
Evaluate all quantities at and keep fixed while differentiating. With , the preceding branch gives
At the hydraulic control of a zero-PV rotating channel, this derivative vanishes. Using reduces it to , where . Equivalently,
Combining this with cancels the quadratic term and leaves
At control and . Both walls must remain wet. Since , the second depth is positive precisely when , giving
Equality is the limiting case , with the second wall just dry. For a wider channel, a dry lateral region requires a different solution and the two-wet-wall hydraulic control of a zero-PV rotating channel formula cannot be continued. The channel must also remain sufficiently long and slowly varying for the transverse geostrophic balance approximation to hold. As a check, gives , the familiar rectangular-channel critical discharge.

Pinned article: Introduction to the OurBigBook Project

Welcome to the OurBigBook Project! Our goal is to create the perfect publishing platform for STEM subjects, and get university-level students to write the best free STEM tutorials ever.
Everyone is welcome to create an account and play with the site: ourbigbook.com/go/register. We belive that students themselves can write amazing tutorials, but teachers are welcome too. You can write about anything you want, it doesn't have to be STEM or even educational. Silly test content is very welcome and you won't be penalized in any way. Just keep it legal!
We have two killer features:
  1. topics: topics group articles by different users with the same title, e.g. here is the topic for the "Fundamental Theorem of Calculus" ourbigbook.com/go/topic/fundamental-theorem-of-calculus
    Articles of different users are sorted by upvote within each article page. This feature is a bit like:
    • a Wikipedia where each user can have their own version of each article
    • a Q&A website like Stack Overflow, where multiple people can give their views on a given topic, and the best ones are sorted by upvote. Except you don't need to wait for someone to ask first, and any topic goes, no matter how narrow or broad
    This feature makes it possible for readers to find better explanations of any topic created by other writers. And it allows writers to create an explanation in a place that readers might actually find it.
    Figure 1.
    Screenshot of the "Derivative" topic page
    . View it live at: ourbigbook.com/go/topic/derivative
  2. local editing: you can store all your personal knowledge base content locally in a plaintext markup format that can be edited locally and published either:
    This way you can be sure that even if OurBigBook.com were to go down one day (which we have no plans to do as it is quite cheap to host!), your content will still be perfectly readable as a static site.
    Figure 5. . You can also edit articles on the Web editor without installing anything locally.
    Video 3.
    Edit locally and publish demo
    . Source. This shows editing OurBigBook Markup and publishing it using the Visual Studio Code extension.
  3. https://raw.githubusercontent.com/ourbigbook/ourbigbook-media/master/feature/x/hilbert-space-arrow.png
  4. Infinitely deep tables of contents:
    Figure 6.
    Dynamic article tree with infinitely deep table of contents
    .
    Descendant pages can also show up as toplevel e.g.: ourbigbook.com/cirosantilli/chordate-subclade
All our software is open source and hosted at: github.com/ourbigbook/ourbigbook
Further documentation can be found at: docs.ourbigbook.com
Feel free to reach our to us for any help or suggestions: docs.ourbigbook.com/#contact