I am trying to track the definition of Lie differentiation. According to my notes:

Given $p\in\mathbb{R}^n$, the tangent space to $\mathbb{R}^n$ at $p$ is the set of pairs $$T_p\mathbb{R}^n=\{(p,v)\}; v\in\mathbb{R}^n.$$

Let $U$ be an open subset of $\mathbb{R}^n$. A vector field on $U$ is a function $v$ which assigns to each point $p\in U$ a vector $v(p)\in T_p\mathbb{R}^n$.

Given a fixed vector $v\in \mathbb{R}^n$, the function $p\in\mathbb{R}^n\rightarrow (p,v)$ is a vector field. Let $e_1,\ldots,e_n$ be the standard basis vectors of $\mathbb{R}^n$. If $v=e_i$, we will denote the vector field by $\dfrac{\partial}{\partial x_i}$.

If $f\in C^1(U)$, we define the Lie differentiation $L_vf$ to be the function on $U$ whose value at $p$ is given by $L_vf(p)=Df(p)v$.

So far so good. Now, the notes say:

If $v=\sum_{i=1}^n g_i\dfrac{\partial}{\partial x_i}$, where $g_i:U\rightarrow\mathbb{R}$ is the function $p\rightarrow g_i(p)$, then $$L_vf=\sum_{i=1}^n g_i\dfrac{\partial}{\partial x_i}f.$$

I'm absolutely confused here. Where does this formula come from? By the definition, $L_vf$ is a function from $U\subseteq \mathbb{R}^n$ to $\mathbb{R}^m$. Does the right-hand side also have the same domain and range?


Though they are not mentioned, I will assume that you know about differential forms, since you are using them when you write down $df(p)$. At this point, it ends up being a simple matter of computation: You know that \begin{align*} df(p) &= \left. \frac{\partial f}{\partial x^1}\right|_p dx^1(p) + \cdots + \left. \frac{\partial f}{\partial x^n}\right|_p dx^n(p) = \sum_{i=1}^n \left. \frac{\partial f}{\partial x^i}\right|_p dx^i(p) \\ v(p) &= \sum_{j=1}^n g_i(p) \left.\frac{\partial}{\partial x^i}\right|_p. \end{align*} From here on out, I'm dropping the $p$'s. Now by definition, we know that $$ dx^i\left(\frac{\partial}{\partial x^j}\right) = \delta_{i,j}:=\begin{cases} 1 & i =j \\ 0 & i \neq j \end{cases} \tag{*} .$$ To deduce the desired equation, we simply apply the definition; that is, we know that $L_v f = df.v$, which gives \begin{align*} L_v f&= df.v = \left( \sum_{i=1}^n \frac{\partial f}{\partial x^i}dx^i \right) \left( \sum_{j=1}^n g_j \frac{\partial}{\partial x^i} \right) \\ &= \sum_{i,j=1}^n g_j \frac{\partial f}{\partial x^i} dx^i\left(\frac{\partial}{\partial x^j}\right) \\ &= \sum_{i,j=1}^n g_j \frac{\partial f}{\partial x^i} \delta_{i,j} & \text{ by (*)} \\ &= \sum_{i=1}^n g_i \frac{\partial}{\partial x^i}f. \end{align*} This is precisely what you wanted to show.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.