Constructing a chart with coord. basis equal to given basis at one pt.

  • Thread starter Thread starter center o bass
  • Start date Start date
  • Tags Tags
    Basis
center o bass
Messages
545
Reaction score
2
Suppose we have a manifold ##M## and at ##p \in M## we have a basis for the tangent space of vectors ##X_i##. Since ##M## is a manifold, there exists a local chart ##(U,\phi)## about ##p##. Now the question is, given such a chart , how can we construct a new chart in a such that ##X_i = \left. \frac{\partial}{\partial x^\mu} \right|_{p}##.

I know there is a theorem that says given commuting vector fields we can find a chart such that these vector fields locally is the coordinate basis of that chart.

However, I want to prove in a transparent manner the less general statement above; that we can construct a coordinate system such that the vector ##X_i## _in the tangent space at p_ is the coordinate basis ##\left. \tfrac{\partial}{\partial x^\mu} \right|_{p}## just at p.
 
Physics news on Phys.org
Use the standard construction for Riemann normal coordinates, but forget the fact that the basis is orthonormal.
 
center o bass said:
Suppose we have a manifold ##M## and at ##p \in M## we have a basis for the tangent space of vectors ##X_i##. Since ##M## is a manifold, there exists a local chart ##(U,\phi)## about ##p##. Now the question is, given such a chart , how can we construct a new chart in a such that ##X_i = \left. \frac{\partial}{\partial x^\mu} \right|_{p}##.

I know there is a theorem that says given commuting vector fields we can find a chart such that these vector fields locally is the coordinate basis of that chart.

However, I want to prove in a transparent manner the less general statement above; that we can construct a coordinate system such that the vector ##X_i## _in the tangent space at p_ is the coordinate basis ##\left. \tfrac{\partial}{\partial x^\mu} \right|_{p}## just at p.

Let (U,\phi) be a chart on M whose domain contains p, and consider the chart (U, A \circ \phi) where A : \mathbb{R}^n \to \mathbb{R}^n is linear and invertible. This chart is smoothly compatible with the original chart (U,\phi).

The transition function A then pushes forward to a linear map A_{*} : T_pM \to T_pM, giving <br /> \left.\frac{\partial}{\partial x^i}\right|_p = \frac{\partial \tilde x^j}{\partial x^i} <br /> \left.\frac{\partial}{\partial \tilde x^j}\right|_p = <br /> (A_{*})_i{}^j \left.\frac{\partial}{\partial \tilde x^j}\right|_p where (x^i) are the coordinate functions of the chart (U,\phi) and (\tilde x^j) are those of the chart (U, A \circ \phi). Since \tilde x^j = A^j{}_i x^i we have that \frac{\partial \tilde x^j}{\partial x^i} = (A_{*})_i{}^j = A^j{}_i. If {X_j} is a basis for T_pM we may then set <br /> \left.\frac{\partial}{\partial \tilde x^j}\right|_p = X_j in the above to obtain <br /> \left.\frac{\partial}{\partial x^i}\right|_p = A^j{}_i X_j.
 
Last edited:

Similar threads

Back
Top