How statement and proof provenance work
The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.
- Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
- AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
- AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.
These labels describe origin, not correctness: citations and verification chips remain separate evidence.
Quasilinear Characteristics and Cauchy Kovalevskaya
1 · Prerequisites
- Absolute and Conditional Convergence; Rearrangement; Products
- Analytic Majorants and the Cauchy–Kovalevskaya Theorem
- Arc Length and Rectifiable Curves
- Binary Operations, Monoids, Groups and Subgroups
- Bounded Variation and the Riemann–Stieltjes Integral
- Compactness in Metric Spaces
- Completeness, Completion, and Uniform Continuity
- Complex Coordinate and Path Transport
- Complex Differentiability and the Cauchy–Riemann Equations
- Complex Power Series and Analytic Functions
- Connectedness
- Construction of the Natural Numbers
- Construction of the Real Numbers via Cauchy Sequences
- Construction of the Real Numbers via Dedekind Cuts
- Continuity, IVT, EVT, and Uniform Continuity
- Contour Integration
- Cosets, Index and Lagrange's Theorem
- Countability and Uncountability
- Determinants of Matrices over a Commutative Ring
- Divisibility, Euclidean Domains, Principal Ideal Domains and Unique Factorisation
- Euclidean Ordinary Differential Equations with Smooth Dependence
- Filters and Ultrafilters
- Finite Counting, Factorials and Binomial Coefficients
- Foundations of the Real Numbers for Analysis
- Fundamental Trigonometric Identities
- Goursat's Theorem and Cauchy's Theorem in a Convex Domain
- Group Actions, Orbits, Stabilisers and Cayley's Theorem
- Holomorphic Functions of Several Complex Variables
- Ideals, Quotient Rings and the Isomorphism Theorems for Rings
- Limits of Real Functions
- limsup, liminf, and Subsequential Limits
- Line Integrals and the Gradient Theorem
- Linear Independence, Bases and Dimension
- Linear Transformations, Rank-Nullity and Quotient Spaces
- Matrices, the Matrix of a Linear Map, and Change of Basis
- Metric Spaces
- Mixed Partials, Taylor Formulae, and Extrema
- Monotone Functions, Discontinuities, and Continuity Sets
- Monotone Sequences, Bolzano-Weierstrass, and Cauchy Completeness
- Normal Subgroups and Quotient Groups
- Order, Zorn's Lemma, and the Axiom of Choice
- Partial Differential Equations and Characteristics
- Partitions of Unity and Paracompactness
- pi: the Equivalent Characterizations
- Picard-Lindelöf and First-Order Ordinary Differential Equations
- Polynomial Rings, the Division Algorithm and Roots
- Power Series and Real-Analytic Functions
- Properties of the Integral and the Working FTC
- Relations, Functions, and Quotients
- Rings, Subrings, Integral Domains and Fields
- Rⁿ as a Normed Space; Vector-Valued Functions
- Roots, Rational Powers, and Classical Inequalities
- Separation Axioms: the Hierarchy
- Sequences and Limits
- Sequences and Series of Functions; Uniform Convergence
- Series: Convergence and the Nonnegative Tests
- Simple Field Extensions and the Construction of the Complex Numbers
- Sine, Cosine, and the Definition of Pi
- Subspaces, Products, and Quotients
- Suprema and Infima
- Symmetric Groups, Cycle Decomposition and the Sign Homomorphism
- The Cantor Set, Baire Category, and Measure Zero in ℝ
- The Complex Exponential and Euler's Formula
- The Derivative and the Mean Value Theorems
- The Exponential Function
- The Fundamental Theorems of Calculus
- The Holomorphic Inverse Function Theorem and Weierstrass Preparation
- The Identity Theorem, the Maximum Principle and the Open Mapping Theorem
- The Inverse and Implicit Function Theorems
- The Logarithm and General Powers
- The Riemann Integral: Definition and Integrability
- The Topology of Euclidean Space
- The Total Derivative in ℝᵐ → ℝⁿ
- The Winding Number and the Global Cauchy Theorem
- The ZFC Axioms and the Basic Set Constructions
- Topological Spaces and Continuity
- Topology of ℝ
- Vector Spaces, Linear Subspaces, Span and Direct Sums
2 · Summary
Starting with a quasilinear Cauchy problem, this page separates ODE solvability from the inverse-projection step that actually constructs a PDE graph. It then makes projection failure precise through Burgers caustics, develops both Charpit invariants for fully nonlinear equations. The closing scope remark links the locally proved Cauchy–Kovalevskaya on a noncharacteristic analytic hypersurface: analytic Cauchy data on a noncharacteristic analytic hypersurface yield a unique local analytic solution germ. For an implicit nonlinear equation the theorem fixes a compatible jet and its selected analytic branch.
3 · Logical flowchart
4 · Definitions, theorems and proofs
Semilinear and quasilinear first-order Cauchy problems on a parametrised hypersurface
Definition
Let and be open, let and be , with for every , and let and be smooth on an open set . Assume that for every . The Cauchy problem for the quasilinear equation is
It is semilinear when is independent of . At , a classical local solution is a function on an open neighbourhood of such that and the PDE holds for every , and such that there is a neighbourhood of with and for every . This specializes the first-order classification in Linear, semilinear, quasilinear, and fully nonlinear partial differential equations; the word noncharacteristic will be tested by the rank calculation below, rather than by importing the space-time transport convention of Noncharacteristic Cauchy surfaces for first-order transport.
The augmented characteristic system for a quasilinear first-order PDE
Definition
For the problem in Semilinear and quasilinear first-order Cauchy problems on a parametrised hypersurface, its augmented characteristic strip is a map satisfying
Here is the characteristic parameter and labels the initial point. The system concerns only ; no differential equation for a putative gradient is included in this definition.
Local solvability and C1 parameter dependence for the augmented characteristic ODE
Statement
At every , the system in The augmented characteristic system for a quasilinear first-order PDE has, after shrinking to and a neighbourhood of , a unique common solution for . It is in .
Facts & Assumptions
Given: The smooth coefficients and initial strip in the definition, and .
Proof
Put and . The coefficients in the defining Cauchy problem are smooth, so is smooth.
Apply Smooth dependence of ODE solutions on parameters to with initial value . It gives a common local interval, uniqueness, and dependence on ; including the ODE variable gives dependence on .
A quasilinear solution lifts to augmented characteristics
Statement
Let solve . If solves while it stays in , then satisfies . Thus solves the augmented characteristic ODE system.
Facts & Assumptions
Given: A solution , a differentiable curve with the stated ODE, and .
Proof
The chain rule gives .
Substitute and the PDE: .
Compatibility of a characteristic strip with Cauchy data
Statement
If a solution has and , then
The first condition is tangential compatibility. The rank condition for the projected strip is a separate hypothesis; it is not derived from this identity.
Facts & Assumptions
Given: The Cauchy data and a solution attaining them.
Proof
Differentiate with respect to to obtain .
Evaluate at and use to obtain the second identity.
Jacobian of a characteristic strip at its initial surface
Statement
For the local strip , the derivative of at is
Consequently is invertible exactly when this matrix has rank .
Facts & Assumptions
Given: The local strip supplied by the preceding lemma.
Proof
Its initial condition gives , and its ODE gives .
These are precisely the first and remaining columns of , proving the formula; an derivative is invertible exactly when it has rank .
Local quasilinear characteristic graph construction
Statement
At , suppose has rank . Then, after shrinking the characteristic strip, is a diffeomorphism onto an open set , and
is the unique function obtained by this inverse-projection construction. It attains ; the next lemma verifies its PDE.
Facts & Assumptions
Given: The smooth coefficients, data, and the stated full-rank condition at .
Proof
The local ODE lemma supplies a strip, and the Jacobian lemma makes invertible.
By The Euclidean inverse function theorem, shrink to a neighbourhood on which has a inverse. Define there.
At , and , hence . Any inverse-projected function from this strip has the same formula and is therefore identical to .
The inverse-projected characteristic graph satisfies the quasilinear PDE
Statement
The function constructed in Local quasilinear characteristic graph construction satisfies on its local domain.
Facts & Assumptions
Given: The local inverse-projected graph and its characteristic identities and .
Proof
The identity differentiates in to .
Substitute the two characteristic identities to get .
Since ranges over the local domain and , this is for every there.
Characteristic crossing and caustic for a first-order PDE
Definition
For a characteristic strip, a crossing or caustic is a point or time at which the projected map loses local rank or local one-to-one graphing. It is a failure of the projection needed to define , not a claim that the lifted ODE has reached its maximal lifespan. For a forward-time family, a first crossing time is an infimum in , with value when the relevant set is empty.
The Burgers slope obeys a Riccati law along characteristics
Statement
Let solve on an open . Let be an open interval containing zero and satisfy and . Then satisfies
for every ; in particular cannot vanish in .
Facts & Assumptions
Given: A classical Burgers solution and one of its projected characteristics.
Proof
Put . Differentiate in to obtain ; equality of mixed partials holds since is .
Along , the total-derivative chain rule gives , so .
Set and . Then and , so differentiation shows is constant and hence zero. Thus . If , this gives ; if , the identity rules out a zero denominator in and yields the displayed formula.
Inviscid Burgers characteristic formula and first crossing time
Statement
For with datum , every classical characteristic has
Thus and its first forward crossing time is
where the infimum of the empty set is .
Facts & Assumptions
Given: A classical Burgers solution with the stated initial datum.
Proof
Along , the chain rule and the PDE give .
The initial condition makes , hence and integration from gives .
Differentiating in gives , so the definition of caustic gives exactly the displayed set and its stated infimum convention.
Monotone Burgers data have no forward characteristic crossing
Statement
If for all , then for . Hence the Burgers projection has no forward crossing. This says nothing about any unrelated global lifespan condition.
Facts & Assumptions
Given: The characteristic formula and .
Proof
For , multiplication gives .
Therefore , so the zero set defining is empty and the projection has no forward crossing.
Uniqueness of a classical quasilinear solution before characteristic crossing
Statement
Let solve the same quasilinear Cauchy problem. On a connected region where their common characteristic projection from the data is a diffeomorphism, .
Facts & Assumptions
Given: Two solutions with identical data, and a connected region on which the common projection is a diffeomorphism.
Proof
The lifting lemma sends the restrictions of both solutions along each data-labelled characteristic to the same augmented initial-value problem.
Uniqueness of that ODE gives equal lifted values on the strip.
The diffeomorphic projection represents every point of the region as one , so step 2.1 gives there.
Fully nonlinear first-order PDEs and complete integrals
Definition
Let . A general first-order equation has the form ; it is fully nonlinear when its dependence on the highest-order variable is not affine after is fixed. A local complete integral on open sets is a family such that for every . Thus the parameters enter essentially rather than merely labelling repeated copies of one solution. A stationary envelope is a value selected by . This definition asserts neither global representation nor differentiability of an envelope.
A nondegenerate stationary envelope solves the Hamilton–Jacobi equation
Statement
If and is invertible, then locally defines a parameter . The envelope has and satisfies .
Facts & Assumptions
Given: A complete integral, a stationary point, and an invertible parameter Hessian there.
Proof
Apply the implicit function theorem to ; its derivative in is the invertible .
The chain rule gives on the stationary branch.
The complete-integral identity , evaluated at and using step 2.1, is .
The Lagrange–Charpit characteristic system
Definition
For on , the Lagrange–Charpit system is
with all derivatives evaluated at . The alternative contact normalization agrees with this one only along ; it is not silently substituted off that constraint.
The Charpit flow preserves the PDE constraint
Statement
Along a Charpit curve,
Consequently an initial point in remains in while the curve exists.
Facts & Assumptions
Given: A Charpit curve for a function .
Proof
The chain rule gives .
Substitute , , and ; the three terms cancel pairwise.
Hence is constant, so an initial zero stays zero.
Charpit contact compatibility is preserved
Statement
Let be a Charpit strip lying in . If , then
while the strip exists.
Facts & Assumptions
Given: A Charpit strip, its initial contact identity, and the constraint .
Proof
Set . Differentiate the Charpit equations in and apply the product and chain rules to obtain .
Differentiating in gives , so and .
Multiplication by the integrating factor makes constant; its initial value is zero, hence .
The Charpit momentum equation from differentiating Hamilton–Jacobi
Statement
If solves , put and let . Then
so the momentum equation is forced by differentiating the PDE.
Facts & Assumptions
Given: A classical solution and the stated projected characteristic.
Proof
Differentiate in to get .
Along , the chain rule gives .
Combining the two identities and gives .
Local fully nonlinear Charpit graph construction
Statement
Let be open and . Let be open, let , and let , , and be maps such that , , and for every . Suppose also that Then the Charpit strip through projects locally to a classical graph satisfying . It is unique while that projection is locally invertible among graphs obtained by inverse-projecting this fixed Charpit strip.
Facts & Assumptions
Given: The stated equation, compatible strip data, and full-rank condition at .
Proof
The Charpit vector field is because , hence locally Lipschitz. Continuous dependence gives a unique common local strip through the initial data .
The -dependence theorem makes this strip in . Its projected derivative at has columns , hence is invertible by the rank hypothesis.
The inverse function theorem supplies a local inverse of ; define .
Constraint preservation gives , while contact preservation gives ; the identity is . Since is invertible, these identities imply .
Substitution in the preserved constraint gives . The fixed strip and its local inverse determine this inverse-projected graph uniquely while the projection remains locally invertible.
Characteristics do not select a post-crossing weak solution
A crossing makes the characteristic projection fail to define a single-valued classical graph. It does not by itself choose a continuation: entropy conditions for conservation laws and viscosity inequalities for Hamilton–Jacobi equations are extra selection principles. No such weak-solution theorem is claimed here.
The proof boundary of Cauchy–Kovalevskaya
Cauchy–Kovalevskaya on a noncharacteristic analytic hypersurface is proved in this library by analytic flattening, normal-form reduction and coefficient-majorant arguments. Its conclusion is local existence and uniqueness of an analytic solution germ for analytic Cauchy data on a noncharacteristic analytic hypersurface. For an implicit nonlinear equation it fixes a compatible jet and its selected analytic branch. These hypotheses do not assert an extension to arbitrary smooth coefficients or data.
5 · Examples, counterexamples and false statements
None yet.