# Invariant properties of metric tensor

• I
• olgerm
In summary, metric tensors are invariant under base vectors, but there are properties of metric tensor, that are basevector invariant and describe space itself.
olgerm
Gold Member
Which properties of metric tensor are invariant of basevectors transforms? I know that metric tensor depends of basevectors, but are there properties of metric tensor, that are basevector invariant and describe space itself?

olgerm said:
Which properties of metric tensor are invariant of basevectors transforms? I know that metric tensor depends of basevectors, but are there properties of metric tensor, that are basevector invariant and describe space itself?
A metric tensor ##g## above an affine point space ##A## with a real translation space ##V## is a map form ##A## into the space of scalar products on ##V ##, i.e. ##g(P)\, : \,V \times V \longrightarrow \mathbb{R}## is a symmetric, positive definite bilinear form on ##V## for every ##P\in A##.

No basis vectors anywhere around.

Or more generally, nothing about any tensor is basis dependent except its components given a particular basis.

##g_{i;j}=e_i\otimes e_j##.

in base 1:
##g_{i;j}=
\begin{bmatrix}
\vec{e_0}\cdot\vec{e_0}&\vec{e_0}\cdot\vec{e_1} \\
\vec{e_1}\cdot\vec{e_0}&\vec{e_1}\cdot\vec{e_1}
\end{bmatrix}=
\begin{bmatrix}
1&0 \\
0&1
\end{bmatrix}##

in base 2:
##\vec{e´_0}=2*\vec{e_0}##
##\vec{e´_1}=\vec{e_1}##

##g_{i;j}=
\begin{bmatrix}
\vec{e´_0}\cdot\vec{e´_0}&\vec{e´_0}\cdot\vec{e´_1} \\
\vec{e´_1}\cdot\vec{e´_0}&\vec{e´_1}\cdot\vec{e´_1}
\end{bmatrix}=
\begin{bmatrix}
(2*\vec{e_0})\cdot(2*\vec{e_0})&\vec{e_0}\cdot\vec{e_1} \\
\vec{e_1}\cdot\vec{e_0}&\vec{e_1}\cdot\vec{e_1}
\end{bmatrix}=
\begin{bmatrix}
4&0\\
0&1
\end{bmatrix}##

This is the old difficulty to distinguish vectors and their coordinates. It is meaningless to ask about a description of a vector (matrix, tensor) once you described them by coordinates. Coordinates are the tool, not the object. It is just difficult to describe the object without coordinates, but the definition in post #2 does it, namely as a map.

DEvens
fresh_42 said:
their coordinates
Or their components

But I agree.

for example minkowsky metric tensor is often given only by components
##
\begin{bmatrix}
-1 & 0 &0 &0 \\
0& 1 & 0 &0 \\
0& 0 & 1 &0 \\
0& 0 & 0 &1
\end{bmatrix}
##
without specifiyng base vectors. Do they assume some specific base vectors? Which ones?

Is there something invariant in the components?
How can spaces with different elemens be compared by their metric tensors if they have different elements and therefore we can't choose same basevectors there?

Last edited:
The standard assumption on Minkowski space is that you are using a set of standard affine Minkowski coordinates.

Orodruin said:
The standard assumption on Minkowski space is that you are using a set of standard affine Minkowski coordinates.
What are these?

There should be something invariant in components of metric tensor because it is probably impossible to choose base where minkowsky metric has components
##\begin{bmatrix}
1 & 0 &0 &0 \\
0& 1 & 0 &0 \\
0& 0 & 1 &0 \\
0& 0 & 0 &1
\end{bmatrix}##

olgerm said:
it is probably impossible to choose base where minkowsky metric has components
##\begin{bmatrix}
1 & 0 &0 &0 \\
0& 1 & 0 &0 \\
0& 0 & 1 &0 \\
0& 0 & 0 &1
\end{bmatrix}##
I was wrng it is possible if ##\vec{e_0'}=\sqrt{-1}*\vec{e_0}##

Is there any relation between metric tensor and transformation matrix? Can I derive lorentz tranformation matrix from minkowsky metric tensor?

olgerm said:
Is there any relation between metric tensor and transformation matrix? Can I derive lorentz tranformation matrix from minkowsky metric tensor?
The metric is a rank 2 tensor under general coordinate transformations, and hence transforms as such (with "two transformation matrices"). The Lorentz transformations are those transformations which keep the Minkowski metric form invariant. These special transformations, which are a subset of the general coordinate transformations, connect al those observers who would use the very same components for the Minkowski metric and we call them inertial observers. So yes, you can derive the Lorentz transformations from this property ("which transformations keep the form of the metric the same? "). It's covered in any basic book about GR, I guess.

Technically, one says that "the isometries of a metric break the general coordinate transformations down to a subgroup of them." This means that the metric transforms as a tensor under general coordinate transformations, but is kept invariant under a subgroup of this group. And those form the isometries (=symmetries) of the spacetime this metric describes.

Hope this helps ;)

haushofer said:
So yes, you can derive the Lorentz transformations from this property ("which transformations keep the form of the metric the same? "). It's covered in any basic book about GR, I guess.
Any basic book on SR should suffice too.

## 1. What are invariant properties of metric tensor?

The invariant properties of metric tensor refer to the characteristics or properties of the tensor that do not change under coordinate transformations. These properties are essential in understanding the geometric properties of a space.

## 2. Why are invariant properties important in science?

Invariant properties are important because they allow scientists to make accurate predictions and calculations regardless of the chosen coordinate system. This is especially useful in fields such as physics and engineering where precise measurements and calculations are crucial.

## 3. How do invariant properties relate to general relativity?

Invariant properties are fundamental in the theory of general relativity as they are used to describe the curvature of spacetime. The metric tensor is used to calculate the invariant properties of spacetime, which in turn, helps to understand the gravitational effects of massive objects.

## 4. Can you give an example of an invariant property of metric tensor?

One example of an invariant property of metric tensor is the length of a vector. Regardless of the coordinate system used, the length of a vector will remain the same. This is because the metric tensor is used to calculate the length of a vector, and it is an invariant property.

## 5. How are invariant properties of metric tensor used in practical applications?

Invariant properties of metric tensor are used in various practical applications, such as navigation systems, GPS technology, and satellite communications. These technologies rely on precise measurements and calculations, and the use of invariant properties ensures accurate results regardless of the coordinate system being used.

• Differential Geometry
Replies
2
Views
515
• Differential Geometry
Replies
14
Views
2K
• Differential Geometry
Replies
9
Views
2K
• Differential Geometry
Replies
37
Views
6K
• Special and General Relativity
Replies
22
Views
2K
• Special and General Relativity
Replies
26
Views
365
• Special and General Relativity
Replies
11
Views
1K
• Differential Geometry
Replies
2
Views
893
• Differential Geometry
Replies
3
Views
6K
• Differential Geometry
Replies
9
Views
416