I Showing Determinant of Metric Tensor is a Tensor Density

AndersF
Messages
27
Reaction score
4
TL;DR Summary
I'm trying to show that the determinant ##g \equiv \det(g_{ij})## of the metric tensor is a tensor density.
I'm trying to show that the determinant ##g \equiv \det(g_{ij})## of the metric tensor is a tensor density. Therefore, in order to do that, I need to show that the determinant of the metric tensor in the new basis, ##g'##, would be given by

##g'=\operatorname{sgn}\bigg(\big(\det(C)\big)\bigg)\big(\det(C)\big)^wg \quad \quad \quad (1)##

With ##C=(C^a_b)_{n \times n}## the change-of-basis matrix.

I know that the metric tensor transforms under a change of basis in this way

##\tilde{g}_{i j}=C_{i}^{\alpha} C_{j}^{\beta} g_{\alpha \beta} \quad \quad \quad (2)##

I see that if I could identify in this last equation (2) a matrix multiplication, then I could use the properties of the determinants to get something similar to equation (1). But I'm stuck here, since these terms don't have the form of "classical" matrix multiplications, ##P^i_j=M^i_k N^k_j##.

Could somebody give me a hint on how to accomplish this demonstration?
 
Last edited:
Physics news on Phys.org
In matrix notation you can write ##\mathrm{det} \tilde{g}=\mathrm{det}(C^{\text{T}} g C)=(\mathrm{det} C)^2 \mathrm{det} g##. So ##g_{\mu \nu}## is a tensor density of weight 2.
 
To add: in linear algebra one has the cyclic identity det(ABC)=det(BCA)=det(CAB), and det(A^T)=det(A) ;)

Edit: typo corrected, thnx robphy!
 
Last edited:
  • Like
Likes vanhees71 and AndersF
haushofer said:
To add: in linear algebra one has the cyclic identity det(ABC)=det(BCA)=det(CBA), and det(A^T)=det(A) ;)
Typo: The third expression in this cyclic identity should be det(CAB).
 
  • Like
Likes vanhees71, haushofer and AndersF
vanhees71 said:
In matrix notation you can write ##\mathrm{det} \tilde{g}=\mathrm{det}(C^{\text{T}} g C)=(\mathrm{det} C)^2 \mathrm{det} g##. So ##g_{\mu \nu}## is a tensor density of weight 2.
haushofer said:
To add: in linear algebra one has the cyclic identity det(ABC)=det(BCA)=det(CAB), and det(A^T)=det(A) ;)

Edit: typo corrected, thnx robphy!

Ok, these were just the "tricks" I was looking for, thank you very much!
 
From $$0 = \delta(g^{\alpha\mu}g_{\mu\nu}) = g^{\alpha\mu} \delta g_{\mu\nu} + g_{\mu\nu} \delta g^{\alpha\mu}$$ we have $$g^{\alpha\mu} \delta g_{\mu\nu} = -g_{\mu\nu} \delta g^{\alpha\mu} \,\, . $$ Multiply both sides by ##g_{\alpha\beta}## to get $$\delta g_{\beta\nu} = -g_{\alpha\beta} g_{\mu\nu} \delta g^{\alpha\mu} \qquad(*)$$ (This is Dirac's eq. (26.9) in "GTR".) On the other hand, the variation ##\delta g^{\alpha\mu} = \bar{g}^{\alpha\mu} - g^{\alpha\mu}## should be a tensor...
OK, so this has bugged me for a while about the equivalence principle and the black hole information paradox. If black holes "evaporate" via Hawking radiation, then they cannot exist forever. So, from my external perspective, watching the person fall in, they slow down, freeze, and redshift to "nothing," but never cross the event horizon. Does the equivalence principle say my perspective is valid? If it does, is it possible that that person really never crossed the event horizon? The...
Back
Top