I was wondering how can one change from Cartesian coordinate system to some other like polar coordinates or spherical coordinates, in the context of special relativity. For example, with the four-velocity, $$V^{\mu}=\frac{\mathrm{d} x^{\mu}}{\mathrm{d} \tau},\tag{1}$$ where $\mu=0,1,2,3$, $x^0=ct$, $x^1=x$, $x^2=y$ and $x^3=z$, if I want to change to cylindrical coordinates, can I simply use the transformations $x=r\cos (\theta)$, $y=r\sin(\theta)$, $z=z$, $t=t$ and then use the chain rule with partial derivatives to work out the four velocity in spherical coordinate, or I am missing something?

Another example, in the relativistic Lagrangian, $$\mathcal{L}=-mc^2\sqrt{1-\frac{|\mathbf{v}|^2}{c^2}}-U(\mathbf{r}),\tag{2}$$ it is correct to choose $r$ and $\theta$ as generalised coordinates, substitute $|\mathbf{v}|^2=\dot{r}^2+r^2\dot{\theta}^2$, to have $$\mathcal{L}=-mc^2\sqrt{1-\frac{\dot{r}^2+r^2\dot{\theta}^2}{c^2}}-U(r),\tag{3}$$ to create a "relativistic central force problem"?

Another question that I have is how this will affect the metric tensor? And the line element?

  • $\begingroup$ "or am I missing something?" Please explain why you think you're missing something here. It seems you're doing it right. $\endgroup$
    – Ari
    Commented Mar 28, 2020 at 2:53
  • $\begingroup$ @Ari It is because I don't know how it will affect the Minkowski metric, or even if it still counts as Special and not general relativity. $\endgroup$
    – Don Al
    Commented Mar 28, 2020 at 2:57

1 Answer 1


You are doing it correctly.

In reference to your comment, to find the minkowski metric in cylindrical coordinates, it is easiest to transform the line element so: $$ ds^2=g_{\mu\nu}dx^\alpha dx^\beta \tag{1}, $$ Where usually the minkowski metric is denoted by $\eta_{\mu\nu}$

Which is $$ ds^2 = -c^2dt^2+dx^2+dy^2+dz^2 \tag{2} $$ In Cartesian.

If you use the transformation equations and chain rule as you said, you get: $$ ds^2=-c^2dt^2 + dr^2 + r^2d\theta^2+dz^2 \tag{3} $$

It's easy to see that $\eta_{\mu\nu}\neq g_{\mu\nu}$ from looking at the first equation.

To summarize:

$$\eta_{\mu\nu} = Diag(-1, 1, 1, 1) \tag{4}$$ $$g_{\mu\nu}= Diag(-1, 1, r^2, 1) \tag{5}$$

With $g_{\mu\nu}$ being the minkowski metric in cylindrical coordinates.

And this usually falls under General relativity rather than special as it's concerned with the curvature of space time.

  • 2
    $\begingroup$ Yes, technically this is special relativity as spacetime is flat, but the coordinates are curvilinear, requiring maths normally associated with general relativity. $\endgroup$ Commented Mar 28, 2020 at 6:59
  • $\begingroup$ The curved surface is embedded in Minkowski space - just like the curved surface of the Earth is embedded in Euclidean space. In both cases the the Riemann tensor $R=0$, i.e., both spaces are flat. $\endgroup$ Commented Mar 28, 2020 at 8:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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