Jet bundle

From Wikipedia, the free encyclopedia
Jump to: navigation, search

In differential geometry, the jet bundle is a certain construction that makes a new smooth fiber bundle out of a given smooth fiber bundle. It makes it possible to write differential equations on sections of a fiber bundle in an invariant form. Jets may also be seen as the coordinate free versions of Taylor expansions.

Historically, jet bundles are attributed to Ehresmann, and were an advance on the method (prolongation) of Élie Cartan, of dealing geometrically with higher derivatives, by imposing differential form conditions on newly introduced formal variables. Jet bundles are sometimes called sprays, although sprays usually refer more specifically to the associated vector field induced on the corresponding bundle (e.g., the geodesic spray on Finsler manifolds.)

More recently, jet bundles have appeared as a concise way to describe phenomena associated with the derivatives of maps, particularly those associated with the calculus of variations. Consequently, the jet bundle is now recognized as the correct domain for a geometrical covariant field theory and much work is done in general relativistic formulations of fields using this approach.

Jets[edit]

Main article: Jet (mathematics)

Suppose M is an m-dimensional manifold and that (E, π, M) is a fiber bundle. For pM, let Γ(π) denote the set of all local sections whose domain contains p. Let I = (I(1), I(2), ..., I(m)) be a multi-index (an ordered m-tuple of integers), then

|I| := \sum_{i=1}^{m} I(i)
\frac{\partial^{|I|}}{\partial x^{I}} := \prod_{i=1}^{m} \left( \frac{\partial}{\partial x^{i}} \right)^{I(i)}.

Define the local sections σ, η ∈ Γ(π) to have the same r-jet at p if

\left.\frac{\partial^{|I|} \sigma^{\alpha}}{\partial x^{I}}\right|_{p} = \left.\frac{\partial^{|I|} \eta^{\alpha}}{\partial x^{I}}\right|_{p}, \quad 0 \leq |I| \leq r.

The relation that two maps have the same r-jet is an equivalence relation. An r-jet is an equivalence class under this relation, and the r-jet with representative σ is denoted j^r_p\sigma . The integer r is also called the order of the jet, p is its source and σ(p) is its target.

Jet manifolds[edit]

The r-th jet manifold of π is the set

\{j^{r}_{p}\sigma:p \in M, \sigma \in \Gamma(\pi)\}

and is denoted Jr(π). We may define projections πr and πr,0 called the source and target projections respectively, by

\begin{cases} \pi_r: J^{r}(\pi) \to M \\ j^{r}_{p}\sigma \mapsto p \end{cases}
\begin{cases} \pi_{r, 0}: J^{r}(\pi) \to E \\  j^{r}_{p}\sigma \mapsto \sigma(p) \end{cases}

If 1 ≤ kr, then the k-jet projection is the function πr,k defined by

\begin{cases} \pi_{r, k}: J^{r}(\pi) \to  J^{k}(\pi)\\  j^{r}_{p}\sigma \mapsto j^{k}_{p}\sigma \end{cases}

From this definition, it is clear that πr = π o πr,0 and that if 0 ≤ mk, then πr,m = πk,m o πr,k. It is conventional to regard πr,r = idJr(π), the identity map on Jr(π) and to identify J0(π) with E.

The functions πr,k, πr,0 and πr are smooth surjective submersions.

Jet Bundle Image FbN.png

A coordinate system on E will generate a coordinate system on Jr(π). Let (U, u) be an adapted coordinate chart on E, where u = (xi, uα). The induced coordinate chart (Ur, ur) on Jr(π) is defined by

 U^{r} = \{ j^{r}_{p}\sigma: \sigma(p) \in U \} \,
 u^{r} = (x^{i}, u^{\alpha}, u^{\alpha}_{I})\,

where

x^{i}(j^{r}_{p}\sigma) = x^{i}(p)
u^{\alpha}(j^{r}_{p}\sigma) = u^{\alpha}(\sigma(p))

and the n \left( {}^{m+r}C_{r} -1\right)\, functions

u^{\alpha}_{I}:U^{k} \to \mathbf{R}\,

are specified by

u^{\alpha}_{I}(j^{r}_{p}\sigma) = \left.\frac{\partial^{|I|} \sigma^{\alpha}}{\partial x^{I}}\right|_{p}

and are known as the derivative coordinates.

Given an atlas of adapted charts (U, u) on E, the corresponding collection of charts (Ur, ur) is a finite-dimensional C atlas on Jr(π).

Jet bundles[edit]

Since the atlas on each Jr(π) defines a manifold, the triples (Jr(π), πr,k, Jk(π)), (Jr(π), πr,0, E) and (Jr(π), πr, M) all define fibered manifolds. In particular, if (E, π, M) is a fiber bundle, the triple (Jr(π), πr, M) defines the r-th jet bundle of π.

If WM is an open submanifold, then

 J^{r}\left(\pi|_{\pi^{-1}(W)}\right) \cong \pi^{-1}_{r}(W).\,

If pM, then the fiber \pi^{-1}_{r}(p)\, is denoted J^{r}_{p}(\pi).

Let σ be a local section of π with domain WM. The r-th jet prolongation of σ is the map jrσ: WJr(π) defined by

 (j^{r}\sigma)(p) = j^{r}_{p}\sigma. \,

Note that πr o jrσ = idW, so jrσ really is a section. In local coordinates, jrσ is given by

 \left(\sigma^{\alpha}, \frac{\partial^{|I|} \sigma^{\alpha}}{\partial x^{|I|}}\right) \qquad 1 \leq |I| \leq r. \,

We identify j0σ with σ.

Example[edit]

If π is the trivial bundle (M × R, pr1, M), then there is a canonical diffeomorphism between the first jet bundle J1(π) and T*M × R. To construct this diffeomorphism, for each σ in ΓM(π) write \bar{\sigma} = pr_{2} \circ \sigma \in C^{\infty}(M)\,.

Then, whenever pM

 j^{1}_{p}\sigma = \{ \psi : \psi \in \Gamma_{p}(\pi); \bar{\psi}(p) = \bar{\sigma}(p); d\bar{\psi}_{p} = d\bar{\sigma}_{p} \}. \,

Consequently, the mapping

\begin{cases} J^{1}(\pi) \to T^*M \times \mathbf{R} \\ j^{1}_{p}\sigma \mapsto (d\bar{\sigma}_{p},\bar{\sigma}(p)) \end{cases}

is well-defined and is clearly injective. Writing it out in coordinates shows that it is a diffeomorphism, because if (xi, u) are coordinates on M × R, where u = idR is the identity coordinate, then the derivative coordinates ui on J1(π) correspond to the coordinates ∂i on T*M.

Likewise, if π is the trivial bundle (R × M, pr1, R), then there exists a canonical diffeomorphism between J1(π) and R × TM.

Contact forms[edit]

A differential 1-form θ on the space Jr(π) is called a contact form (i.e. \theta \in \Lambda_{C}^{r}\pi\,) if it is pulled back to the zero form on M by all prolongations. In other words, if \theta \in \Lambda^{1}J^{r+1}\pi\,, then \theta \in \Lambda_{C}^{1}\pi_{r+1,r}\, if and only if, for every open submanifold WM and every σ in ΓM(π)

(j^{k+1}\sigma)^{*}\theta = 0.\,

The distribution on Jr(π) generated by the contact forms is called the Cartan distribution. It is the main geometrical structure on jet spaces and plays an important role in the geometric theory of partial differential equations. The Cartan distributions are not involutive and are of growing dimension when passing to higher order jet spaces. Surprisingly though, when passing to the space of infinite order jets J this distribution is involutive and finite dimensional. Its dimension coinciding with the dimension of the base manifold M.

Example[edit]

Let us consider the case (E, π, M), where ER2 and MR. Then, (J1(π), π, M) defines the first jet bundle, and may be coordinated by (x, u, u1), where

x(j^{1}_{p}\sigma) = x(p) = x\,
u(j^{1}_{p}\sigma) = u(\sigma(p)) = u(\sigma(x)) = \sigma(x) \,
u_{1}(j^{1}_{p}\sigma) = \left.\frac{\partial \sigma}{\partial x}\right|_{p} = \sigma'(x)

for all pM and σ in Γp(π). A general 1-form on J1(π) takes the form

\theta = a(x, u, u_{1})dx + b(x, u, u_{1})du + c(x, u,u_{1})du_{1}\,

A section σ in Γp(π) has first prolongation

 j^{1}\sigma = (u,u_{1}) = \left(\sigma(p), \left. \frac{\partial \sigma}{\partial x} \right|_{p} \right).

Hence, (j1σ)*θ can be calculated as

(j^{1}_{p}\sigma)^{*} \theta \, = \theta \circ j^{1}_{p}\sigma \,
= a(x, \sigma(x), \sigma'(x))dx + b(x, \sigma(x), \sigma'(x))d(\sigma(x)) + c(x, \sigma(x),\sigma'(x))d(\sigma'(x)) \,
= a(x, \sigma(x), \sigma'(x))dx + b(x, \sigma(x), \sigma'(x))\sigma'(x)dx + c(x, \sigma(x),\sigma'(x))\sigma''(x)dx \,
= [\, a(x, \sigma(x), \sigma'(x)) + b(x, \sigma(x), \sigma'(x))\sigma'(x) + c(x, \sigma(x),\sigma'(x))\sigma''(x)\, ]dx \,

This will vanish for all sections σ if and only if c = 0 and a = −bσ′(x). Hence, θ = b(x, u, u10 must necessarily be a multiple of the basic contact form θ0 = duu1dx. Proceeding to the second jet space J2(π) with additional coordinate u2, such that

u_{2}(j^{2}_{p}\sigma)=\left.\frac{\partial^{2} \sigma}{\partial x^{2}}\right|_{p} = \sigma''(x)\,

a general 1-form has the construction

 \theta = a(x, u, u_{1},u_{2})dx + b(x, u, u_{1},u_{2})du + c(x, u, u_{1},u_{2})du_{1} + e(x, u, u_{1},u_{2})du_{2}\,

This is a contact form if and only if

 (j^{2}_{p}\sigma)^{*} \theta \, = \theta \circ j^{2}_{p}\sigma \,
= a(x, \sigma(x), \sigma'(x),\sigma''(x))dx + b(x, \sigma(x),\sigma'(x),\sigma''(x))d(\sigma(x))+ \,
+ c(x, \sigma(x),\sigma'(x),\sigma'(x))d(\sigma'(x)) +  e(x, \sigma(x), \sigma'(x),\sigma''(x))d(\sigma''(x)) \,
= adx + b\sigma'(x)dx + c\sigma''(x)dx + e\sigma'''(x)dx\,
= [\, a + b\sigma'(x) + c\sigma''(x) + e\sigma'''(x)\,]dx\,
= 0\,

which implies that e = 0 and a = −bσ′(x)cσ′′(x). Therefore, θ is a contact form if and only if

\theta = b(x, \sigma(x), \sigma'(x))\theta_{0} + c(x, \sigma(x), \sigma'(x))\theta_{1}\,

where θ1 = du1 − u2dx is the next basic contact form (Note that here we are identifying the form θ0 with its pull-back (\pi_{2,1})^{*}\theta_{0}\, to J2(π)).

In general, providing x, uR, a contact form on Jr+1(π) can be written as a linear combination of the basic contact forms

\theta_{k} = du_{k} - u_{k+1}dx \qquad k=0, \ldots, r-1\,

where  u_{k}(j^{k}\sigma)= \left.\frac{\partial^{k} \sigma}{\partial x^{k}}\right|_{p}\,.

Similar arguments lead to a complete characterization of all contact forms.

In local coordinates, every contact one-form on Jr+1(π) can be written as a linear combination

\theta = \sum_{|I|=0}^{r} P_{\alpha}^{I}\theta_{I}^{\alpha}\,

with smooth coefficients P^{\alpha}_{I}(x^{i},u^{\alpha})\, of the basic contact forms

\theta_{I}^{\alpha} = du^{\alpha}_{I} - u^{\alpha}_{I,i}dx^{i}\,

|I| is known as the order of the contact form \theta_{I}^{\alpha}. Note that contact forms on Jr+1(π) have orders at most r. Contact forms provide a characterization of those local sections of πr+1 which are prolongations of sections of π.

Let ψ ∈ ΓW(πr+1), then ψ = jr+1σ where σ ∈ ΓW(π) if and only if \psi^{*}(\theta|_{W})=0, \forall \theta \in \Lambda_{C}^{1}\pi_{r+1,r}.\,

Vector fields[edit]

A general vector field on the total space E, coordinated by (x,u) \ \stackrel{\mathrm{def}}{=}\  (x^{i},u^{\alpha})\,, is

V \ \stackrel{\mathrm{def}}{=}\  \rho^{i}(x,u)\frac{\partial}{\partial x^{i}} + \phi^{\alpha}(x,u)\frac{\partial}{\partial u^{\alpha}}.\,

A vector field is called horizontal, meaning that all the vertical coefficients vanish, i.e. φα = 0.

A vector field is called vertical, meaning that all the horizontal coefficients vanish, i.e. ρi = 0.

For fixed (x, u), we identify

 V_{(xu)} \ \stackrel{\mathrm{def}}{=}\  \rho^{i}(x,u) \frac{\partial}{\partial x^{i}} + \phi^{\alpha}(x,u) \frac{\partial}{\partial u^{\alpha}}\,

having coordinates (x, u, ρi, φα), with an element in the fiber TxuE of TE over (x,u) in E, called a tangent vector in TE. A section

 \begin{cases} \psi : E \to TE \\ (x,u) \mapsto \psi(x,u) = V \end{cases}

is called a vector field on E' with

 V = \rho^{i}(x,u) \frac{\partial}{\partial x^{i}} + \phi^{\alpha}(x,u) \frac{\partial}{\partial u^{\alpha}}\,

and ψ in Γ(TE).

The jet bundle Jr(π) is coordinated by (x,u,w) \ \stackrel{\mathrm{def}}{=}\  (x^{i},u^{\alpha},w_{i}^{\alpha})\,. For fixed (x,u,w), identify

V_{(xuw)} \ \stackrel{\mathrm{def}}{=}\  \, V^{i}(x,u,w) \frac{\partial}{\partial x^{i}} + V^{\alpha}(x,u,w) \frac{\partial}{\partial u^{\alpha}} \ + \ V^{\alpha}_{i}(x,u,w) \frac{\partial}{\partial w^{\alpha}_{i}} +\,
\qquad + \ V^{\alpha}_{i_{1}i_{2}}(x,u,w) \frac{\partial}{\partial w^{\alpha}_{i_{1}i_{2}}} + \cdots \ + \ \cdots + V^{\alpha}_{i_{1}i_{2} \cdots i_{r}}(x,u,w) \frac{\partial}{\partial w^{\alpha}_{i_{1}i_{2} \cdots i_{r}}}\,

having coordinates (x,u,w,v^{\alpha}_{i}, v^{\alpha}_{i_{1} i_{2}},\ldots,v^{ \alpha}_{i_{1}i_{2} \cdots i_{r}})\,, with an element in the fiber T_{xuw}(J^{r}\pi)\, of TJr(π) over (x, u, w)Jr(π), called a tangent vector in TJr(π). Here,

v^{\alpha}_{i}, v^{\alpha}_{i_{1}i_{2}},\ldots,v^{\alpha}_{i_{1}i_{2} \cdots i_{r}}\,

are real-valued functions on Jr(π). A section

 \begin{cases} \Psi : J^{r}(\pi) \to  TJ^{r}(\pi) \\ (x,u,w) \mapsto \Psi(u,w) = V \end{cases}

is a vector field on Jr(π), and we say \Psi \in \Gamma(T(J^{r}\pi))\,.

Partial differential equations[edit]

Let (E, π, M) be a fiber bundle. An r-th order partial differential equation on π is a closed embedded submanifold S of the jet manifold Jr(π). A solution is a local section σ ∈ ΓW(π) satisfying j^{r}_{p}\sigma \in S, forall p in M.

Let us consider an example of a first order partial differential equation.

Example[edit]

Let π be the trivial bundle (R2 × R, pr1, R2) with global coordinates (x1, x2, u1). Then the map F : J1(π) R defined by

F = u^{1}_{1}u^{1}_{2} - 2x^{2}u^{1}\,

gives rise to the differential equation

S = \{ j^{1}_{p}\sigma \in J^{1}\pi : (u^{1}_{1}u^{1}_{2} - 2x^{2}u^{1})(j^{1}_{p}\sigma)=0 \} \,

which can be written

\frac{\partial \sigma}{\partial x^{1}}\frac{\partial \sigma}{\partial x^{2}} - 2x^{2}\sigma = 0. \,

The particular section σ: R2R2 × R defined by

\sigma(p_{1},p_{2}) = (p^{1},p^{2},p^{1}(p^{2})^{2}) \,

has first prolongation given by

 j^{1}\sigma(p_{1},p_{2}) = (p^{1},p^{2},p^{1}(p^{2})^{2},(p^{2})^{2},2p^{1}p^{2}) \,

and is a solution of this differential equation, because

(u^{1}_{1}u^{1}_{2} - 2x^{2}u^{1})(j^{1}_{p}\sigma) \, = u^{1}_{1}(j^{1}_{p}\sigma)u^{1}_{2}(j^{1}_{p}\sigma) - 2x^{2}(j^{1}_{p}\sigma)u^{1}(j^{1}_{p}\sigma) \,
= (p^{2})^{2} \cdot 2p^{1}p^{2} - 2 \cdot p^{2} \cdot p^{1}(p^{2})^{2} \,
= 2p^{1}(p^{2})^3 - 2p^{1}(p^{2})^3 \,
= 0 \,

and so j^{1}_{p}\sigma \in S for every pR2.

Jet Prolongation[edit]

A local diffeomorphism ψ: Jr(π)Jr(π) defines a contact transformation of order r if it preserves the contact ideal, meaning that if θ is any contact form on Jr(π), then ψ*θ is also a contact form.

The flow generated by a vector field Vr on the jet space Jr(π) forms a one-parameter group of contact transformations if and only if the Lie derivative \mathcal{L}_{V^{r}}(\theta) of any contact form θ preserves the contact ideal.

Let us begin with the first order case. Consider a general vector field V1 on J1(π), given by

V^1 \ \stackrel{\mathrm{def}}{=}\  \rho^{i}(u^{1})\frac{\partial}{\partial x^{i}} + \phi^{\alpha}(u^{1})\frac{\partial}{\partial u^{\alpha}} + \chi^{\alpha}_{i}(u^{1})\frac{\partial}{\partial u^{\alpha}_{i}}.

We now apply \mathcal{L}_{V^{1}} to the basic contact forms \theta^{\alpha} = du^{\alpha} - u_{i}^{\alpha}dx^{i}\,, and obtain

\mathcal{L}_{V^{1}}(\theta^{\alpha}) = \mathcal{L}_{V^{1}}(du^{\alpha} - u_{i}^{\alpha}dx^{i})
= \mathcal{L}_{V^{1}}du^{\alpha} - (\mathcal{L}_{V^{1}}u_{i}^{\alpha})dx^{i} - u_{i}^{\alpha}(\mathcal{L}_{V^{1}}dx^{i}) \,
= d(V^{1}u^{\alpha}) - V^{1}u_{i}^{\alpha}dx^{i} - u_{i}^{\alpha}d(V^{1}x^{i}) \,
= d\phi^{\alpha} - \chi^{\alpha}_{i}dx^{i} - u_{i}^{\alpha}d\rho^{i} \,
= \frac{\partial \phi^{\alpha}}{\partial x^{i}}\, dx^{i} + \frac{\partial \phi^{\alpha}}{\partial u^{k}}\, du^{k} + \frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}}\, du^{k}_{i} - \chi^{\alpha}_{i}dx^{i} - u_{i}^{\alpha}\left[ \frac{\partial \rho^{i}}{\partial x^{m}}\, dx^{m} + \frac{\partial \rho^{i}}{\partial u^{k}}\, du^{k} + \frac{\partial \rho^{i}}{\partial u^{k}_{m}}\, du^{k}_{m} \right ] \,

where we have expanded the exterior derivative of the functions in terms of their coordinates. Next, we note that

 \theta^{k} = du^{k} - u_{i}^{k}dx^{i} \quad \Longrightarrow \quad du^{k} = \theta^{k} + u_{i}^{k}dx^{i} \,

and so we may write

\mathcal{L}_{V^{1}}(\theta^{\alpha}) \, = \frac{\partial \phi^{\alpha}}{\partial x^{i}}\, dx^{i} + \frac{\partial \phi^{\alpha}}{\partial u^{k}}\, (\theta^{k} + u_{i}^{k}dx^{i}) + \frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}}\, du^{k}_{i} - \chi^{\alpha}_{i}dx^{i} - \,
-u_{l}^{\alpha} \left[ \frac{\partial \rho^{l}}{\partial x^{i}}\, dx^{i} + \frac{\partial \rho^{l}}{\partial u^{k}}\, (\theta^{k} + u_{i}^{k}dx^{i}) + \frac{\partial \rho^{l}}{\partial u^{k}_{i}}\, du^{k}_{i} \right ] \,
= \left[ \frac{\partial \phi^{\alpha}}{\partial x^{i}} + \frac{\partial \phi^{\alpha}}{\partial u^{k}}u_{i}^{k} - u_{l}^{\alpha}\left(\frac{\partial \rho^{l}}{\partial x^{i}} + \frac{\partial \rho^{l}}{\partial u^{k}}u_{i}^{k}\right)- \chi^{\alpha}_{i}\right]\, dx^{i} + \left[ \frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}} - u_{l}^{\alpha}\frac{\partial \rho^{l}}{\partial u^{k}_{i}}\right]\, du^{k}_{i} + \,
+ \left( \frac{\partial \phi^{\alpha}}{\partial u^{k}} - u_{l}^{\alpha}\frac{\partial \rho^{l}}{\partial u^{k}} \right)\theta^{k}.\,

Therefore, V1 determines a contact transformation if and only if the coefficients of dxi and du^{k}_{i}\, in the formula vanish. The latter requirements imply the contact conditions

\frac{\partial \phi^{\alpha}}{\partial u^{k}_{i}} - u^{\alpha}_{l} \frac{\partial \rho^{l}}{\partial u^{k}_{i}} = 0\,

The former requirements provide explicit formulae for the coefficients of the first derivative terms in V1:

\chi^{\alpha}_{i} = \widehat{D}_{i} \phi^{\alpha} - u^{\alpha}_{l}(\widehat{D}_{i}\rho^{l})

where

\widehat{D}_{i} = \frac{\partial}{\partial x^{i}} + u^{k}_{i}\frac{\partial}{\partial u^{k}}

denotes the zeroth order truncation of the total derivative Di.

Thus, the contact conditions uniquely prescribe the prolongation of any point or contact vector field. That is, if \mathcal{L}_{V^{r}}\, satisfies these equations, Vr is called the r-th prolongation of V to a vector field on Jr(π).

These results are best understood when applied to a particular example. Hence, let us examine the following.

Example[edit]

Let us consider the case (E, π, M), where ER2 and MR. Then, (J1(π), π, E) defines the first jet bundle, and may be coordinated by (x, u, u1), where

x(j^{1}_{p}\sigma) \, = x(p) = x \,
u(j^{1}_{p}\sigma) \, = u(\sigma(p)) = u(\sigma(x)) = \sigma(x) \,
u_{1}(j^{1}_{p}\sigma) \, = \left.\frac{\partial \sigma}{\partial x}\right|_{p} = \dot{\sigma}(x) \,

for all pM and σ in Γp(π). A contact form on J1(π) has the form

\theta = du - u_{1}dx \,

Let us consider a vector V on E, having the form

V = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} \,

Then, the first prolongation of this vector field to J1(π) is

V^{1} \, = V + Z \,
= x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + Z \,
= x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + \rho(x,u,u_{1})\frac{\partial}{\partial u_{1}} \,

If we now take the Lie derivative of the contact form with respect to this prolonged vector field, \mathcal{L}_{V^{1}}(\theta)\,, we obtain

\mathcal{L}_{V^{1}}(\theta) \, = \mathcal{L}_{V^{1}}(du - u_{1}dx) \,
= \mathcal{L}_{V^{1}}du - (\mathcal{L}_{V^{1}}u_{1})dx - u_{1}(\mathcal{L}_{V^{1}}dx) \,
= d(V^{1}u) - V^{1}u_{1}dx - u_{1}d(V^{1}x) \,
= dx - \rho(x,u,u_{1})dx + u_{1}du \,
= (1 - \rho(x,u,u_{1}) )dx + u_{1}du \,

But, we may identify du = θ + u1dx. Thus, we get

\mathcal{L}_{V^{1}}(\theta) \, = [\,1 - \rho(x,u,u_{1})\,]dx + u_{1}(\theta + u_{1}dx) \,
= [\,1 + u_{1}u_{1} - \rho(x,u,u_{1})\,]dx + u_{1}\theta  \,

Hence, for \mathcal{L}_{V^{1}}(\theta)\, to preserve the contact ideal, we require

1 + u_{1}u_{1} - \rho(x,u,u_{1}) = 0 \,
\Longrightarrow \quad \, \rho(x,u,u_{1}) = 1 + u_{1}u_{1}\,

And so the first prolongation of V to a vector field on J1(π) is

 V^{1} = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + (1 + u_{1}u_{1})\frac{\partial}{\partial u_{1}} \,

Let us also calculate the second prolongation of V to a vector field on J2(π). We have \{x,u,u_{1}, y_{2}\}\, as coordinates on J2(π). Hence, the prolonged vector has the form

 V^{2} = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + \rho(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{1}} + \phi(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{2}} \,

The contacts forms are

\theta \, = du - u_{1}dx \,
\theta_{1}  \, = du_{1} - u_{2}dx \,

To preserve the contact ideal, we require

\mathcal{L}_{V^{2}}(\theta) \, = 0\,
\mathcal{L}_{V^{2}}(\theta_{1}) \, = 0 \,

Now, θ has no u2 dependency. Hence, from this equation we will pick up the formula for ρ, which will necessarily be the same result as we found for V1. Therefore, the problem is analogous to prolonging the vector field V1 to J2(π). That is to say, we may generate the r-th prolongation of a vector field by recursively applying the Lie derivative of the contact forms with respect to the prolonged vector fields, r times. So, we have

 \rho(x,u,u_{1}) = 1 + u_{1}u_{1} \,

and so

V^{2} \, = V^{1} + \phi(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{2}} \,
= x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + (1 + u_{1}u_{1})\frac{\partial}{\partial u_{1}} + \phi(x,u,u_{1},u_{2})\frac{\partial}{\partial u_{2}} \,

Therefore, the Lie derivative of the second contact form with respect to V2 is

\mathcal{L}_{V^{2}}(\theta_{1}) \, = \mathcal{L}_{V^{2}}(du_{1} - u_{2}dx) \,
= \mathcal{L}_{V^{2}}du_{1} - (\mathcal{L}_{V^{2}}u_{2})dx - u_{2}(\mathcal{L}_{V^{2}}dx) \,
= d(V^{2}u_{1}) - V^{2}u_{2}dx - u_{2}d(V^{2}x) \,
= d(1-u_{1}u_{1}) - \phi(x,u,u_{1},u_{2})dx + u_{2}du \,
= 2u_{1}du_{1} - \phi(x,u,u_{1},u_{2})dx + u_{2}du \,

Again, let us identify du = θ + u1dx and du1 = θ1 + u2dx. Then we have

\mathcal{L}_{V^{2}}(\theta_{1}) \, = 2u_{1}(\theta_{1} + u_{2}dx) - \phi(x,u,u_{1},u_{2})dx + u_{2}(\theta + u_{1}dx) \,
= [\, 3u_{1}u_{2} - \phi(x,u,u_{1},u_{2})\,]dx + u_{2}\theta + 2u_{1}\theta_{1} \,

Hence, for \mathcal{L}_{V^{2}}(\theta_{1})\, to preserve the contact ideal, we require

3u_{1}u_{2} - \phi(x,u,u_{1},u_{2}) = 0 \,
\Longrightarrow \quad \, \phi(x,u,u_{1},u_{2}) = 3u_{1}u_{2} \,

And so the second prolongation of V to a vector field on J2(π) is

 V^{2} = x \frac{\partial}{\partial u} - u \frac{\partial}{\partial x} + (1 + u_{1}u_{1})\frac{\partial}{\partial u_{1}} + 3u_{1}u_{2}\frac{\partial}{\partial u_{2}} \,

Note that the first prolongation of V can be recovered by omitting the second derivative terms in V2, or by projecting back to J1(π).

Infinite Jet Spaces[edit]

The inverse limit of the sequence of projections \pi_{k+1,k}:J^{k+1}(\pi)\to J^k(\pi) gives rise to the infinite jet space J(π). A point j_p^\infty(\sigma) is the equivalence class of sections of π that have the same k-jet in p as σ for all values of k. The natural projection π maps j_p^\infty(\sigma) into p.

Just by thinking in terms of coordinates, J(π) appears to be an infinite-dimensional geometric object. In fact, the simplest way of introducing a differentiable structure on J(π), not relying on differentiable charts, is given by the differential calculus over commutative algebras. Dual to the sequence of projections \pi_{k+1,k}:J^{k+1}(\pi)\to J^k(\pi) of manifolds is the sequence of injections \pi_{k+1,k}^*: C^\infty(J^{k}(\pi))\to C^\infty(J^{k+1}(\pi)) of commutative algebras. Let's denote C^\infty(J^{k}(\pi)) simply by \mathcal{F}_k(\pi). Take now the direct limit \mathcal{F}(\pi) of the \mathcal{F}_k(\pi)'s. It will be a commutative algebra, which can be assumed to be the smooth functions algebra over the geometric object J(π). Observe that \mathcal{F}(\pi), being born as a direct limit, carries an additional structure: it is a filtered commutative algebra.

Roughly speaking, a concrete element \varphi\in\mathcal{F}(\pi) will always belong to some \mathcal{F}_k(\pi), so it is a smooth function on the finite-dimensional manifold Jk(π) in the usual sense.

Infinitely prolonged PDEs[edit]

Given a k-th order system of PDEs EJk(π), the collection I(E) of vanishing on E smooth functions on J(π) is an ideal[disambiguation needed] in the algebra \mathcal{F}_k(\pi), and hence in the direct limit \mathcal{F}(\pi) too.

Enhance I(E) by adding all the possible compositions of total derivatives applied to all its elements. This way we get a new ideal I of \mathcal{F}(\pi) which is now closed under the operation of taking total derivative. The submanifold E(∞) of J(π) cut out by I is called the infinite prolongation of E.

Geometrically, E(∞) is the manifold of formal solutions of E. A point j_p^\infty(\sigma) of E(∞) can be easily seen to be represented by a section σ whose k-jet's graph is tangent to E at the point j_p^k(\sigma) with arbitrarily high order of tangency.

Analytically, if E is given by φ = 0, a formal solution can be understood as the set of Taylor coefficients of a section σ in a point p that make vanish the Taylor series of \varphi\circ j^k(\sigma) at the point p.

Most importantly, the closure properties of I imply that E(∞) is tangent to the infinite-order contact structure \mathcal{C} on J(π), so that by restricting \mathcal{C} to E(∞) one gets the diffiety (E_{(\infty)}, \mathcal{C}|_{E_{(\infty)}}), and can study the associated C-spectral sequence.

Remark[edit]

This article has defined jets of local sections of a bundle, but it is possible to define jets of functions f: MN, where M and N are manifolds; the jet of f then just corresponds to the jet of the section

grf: MM × N
grf(p) = (p, f(p))

(grf is known as the graph of the function f) of the trivial bundle (M × N, π1, M). However, this restriction does not simplify the theory, as the global triviality of π does not imply the global triviality of π1.

See also[edit]

References[edit]