Please verify my differential geometry results

In summary, the conversation discusses a method for defining a system of coordinates for a sphere, called stereographic projection, which involves mapping points on the sphere to points on the xy plane. It is shown that the inverse of this projection can be expressed in terms of the coordinates on the sphere. The conversation also explores the use of stereographic projection to create coordinate neighborhoods on the sphere. Finally, the conversation delves into the topic of parameterization of the unit sphere and how it can be achieved using logarithmic and trigonometric functions. The resulting new parameterization is shown to be conformal, preserving angles and maintaining the meridians and latitudes on the sphere.
  • #1
Cauchy1789
46
0

Homework Statement





Q1)

One way to define a system of coordinants for a Sphere [tex]S^2[/tex] given by [tex]x^2 + y^2 + (z-1)^2 = 1[/tex] is socalled stereographical projection [tex]\pi \thilde \{N} \rightarrow R^2[/tex] which carries a point p=(x,y,z) of the sphere minus the Northpole (0,0,2) onto the intersection of the xy plane with the straight line which connects N to p.

Let (u,v) = pi(x,y,z) where [tex]S^2 \thilde {N}[/tex] and [tex]u,v \in xy[/tex]

1) show that [tex]pi^-1 = R^2 \rightarrow S^2[/tex] is given by

[tex]x = \frac{4u}{u^2+v^2+4}, y = \frac{4v}{u^2+v^2+4}, z = \frac{2(u^2+v^2)}{u^2+v^2+4}[/tex]

The Attempt at a Solution


Solution (1)

The line passing through N and (u,v,0) can be given by l(t) = t(0,0,2) + (1-t)(u,v,0) = ((1-t)u, (1-t)v, 2t)) When l intersecs with the unit sphere, we have (1-t)u)^2+(1-t)v)^2+ (2t-1)^2 = 1 and thus t =1 or \frac{u^2+v^2}{u^2+v^2+4} = (\frac{4u}{u^2+v^2+4}, \frac{4v}{u^2+v^2+4}, \frac{2(u^2+v^2)}{u^2+v^2+4})

is then pi^-1(u,v)

2) Show if possible using stereographical projection the sphere with two coordinant neighbourhoods.

The Attempt at a Solution


solution 2)

Define F: S^2 -> S^2 by F(x,y,z) = (x,y,2-z). Then F is rigid motion of S^2 which the upper hemisphere to the lower hemisphere and vice versa. Now pi^-1: R^2 -> S^2 is a parameterization of S^2 which covers the whole sphere except N. Similary F o pi^-1: R^2 -> S^2is a parameterisation of S^2 missing the origin. So pi^-1, F o pi^-1 together covers the whole S^2, and these are both regular parameterization by definition.

Q2)

Prove that the absolute vale of the torsion tau at a point on an assymptotical curve, whose curvature is nowhere zero, is given by

|tau| = sqrt(-K)


The Attempt at a Solution



Let α(s) be an asymptotic curve on S parametrized with arc length, with curvature k ≠ 0,
and with α(0)=p.
First, note that N, the unit normal vector to S at p, is normal to the osculating plane of
α(s) at p: Clearly, N is normal to the tangent vector α’(0), and, since α(s) is asymptotic:
kn = 0 = k <n, N> (by definition of kn, p. 141).
Since k ≠ 0, this means that N is also normal to n (the normal vector of α(s)).
Thus, for some orientation of S, N coincides with the binormal b = α’ ^ n, then,
N’(s) = b’ = τ n, and:
|N’(s)|2 = τ2. (1)
On the other hand, N’(s) is normal to N, and therefore belongs to the tangent plane of S
at α(s), and can be expressed in terms of the unit vectors along the principal directions
of S at α(s), e1 and e2 (which form an orthonormal basis of the tangent plane):
N’(s)= < N’(s), e1> e1 + < N’(s), e2> e2
Now let’s compute these componets: if θ is the angle formed by α’(s) and e1, we have:
< N’(s), e1> = <dN(α’(s)), e1> = <dN(e1 cos θ + e2 sin θ), e1> =
= <cos θ dN(e1) + sin θ dN(e2), e1> = cosθ <dN(e1), e1> + sin θ <dN(e2), e1> =
= cosθ < – k1e1, e1> + sin θ < – k2e2, e1> = – k1 cosθ.
Similarly, we get: < N’(s), e2> = – k2 sinθ, so that
N’(s)= – k1 cosθ e1 – k2 sinθ e2
and
|N’(s)|2 = <N’(s), N’(s)> = k1
2 cos2 θ + k2
2 sin2 θ
On the other hand, according to the Euler formula, the normal curvature of the
asymptotic curve α(s) is:
kn= 0 = k1 cos2 θ + k2 sin2 θ = k1 cos2 θ + k2 (1- cos2 θ)
yielding:
cos2 θ = − k2 /( k1 – k2),
sin2 θ = k1 /( k1 – k2).
Summarizing,
|N’(s)|2 = – k1
2 k2 /( k1 – k2) + k2
2 k1 /( k1 – k2) = – k1 k2 = –K
and using (1):
|N’(s)| = |τ| = sqrt(–K).

q3)

Let x: U \subset R^2 -> R^3 where

U = (omega, phi) \in R^2, 0 < \omega < pi, 0 <\phi <2pi

x(omega, phi) = sin(omega)cos(phi), sin(phi), cos(omega)

be a parameterisation of the unit sphere S^2 let

log tan 1/2 omega = u, phi = v

and show that the new parameterization of the coordinant neighbourhood x(U) = V

is given by

y(u,v) = (sech u cos v, sech u sin v tanh u)

Solution Q3)

We can solve log tan(omega/2) = u, phi = v, for omega, phi,

omega = 2arctan(e^u), pi, = v

note that

0<omega<pi corresponds to -infinity < u < inftiny
0<omega<pi corresponds to 0 < v < 2pi

If w set Y(u,v) = X(2arctan e^u, v) where X(omega, phi) = (sin(omega) cos(phi), sin(omega)sin(ph), cos(omega)))

then Y:U' -> \subseteq S^2 is a chart on the sphere S^2 where U' = {(u,v) : 0 < v < 2pi}; V is the imega of X, name S^2 with a half great circle from the pole to South Pole removed.

under the changed variable we find

sin(omega)= sin(2 arctan(e^u) = 2sin(arctan(e^u)cos(arctan(e^u) = 2eu/1+e^2u

cos(omega) = -tanh u

So Y(u,v) = (sech u cos(v), sech u sin v, -tanh(u)) (Mistake in do Carmo?)

From this we find E=G=sech^2(u), F=0, for Y, so Y is conformal and therefore Y^-1: V->U' is conformal as well. The meridians on the sphere corespondes to the constant value of omega, and we realize by constant values of U in U'(vetical lines) while the latittudes on the sphere corresponds to the constant value of phi, giving horisontal lines in U'.
 
Last edited:
Physics news on Phys.org
  • #2
Since Y is conformal, it preserves angles and hence the meridians and latitudes are maitained under the new parameterization.
 

FAQ: Please verify my differential geometry results

1. What is differential geometry?

Differential geometry is a branch of mathematics that deals with the study of curves and surfaces in multidimensional spaces. It combines the principles of calculus and linear algebra to understand the geometry and topology of these objects.

2. Why is it important to verify differential geometry results?

Verifying differential geometry results is crucial because it ensures the accuracy and validity of mathematical theories and models. It allows for the detection of errors and provides a level of confidence in the results obtained.

3. How do you verify differential geometry results?

Differential geometry results can be verified through rigorous mathematical proofs, simulations, or experiments. It is also important to compare and contrast the results with existing theories and known results in the field.

4. What are some common applications of differential geometry?

Differential geometry has various applications in fields such as physics, engineering, computer graphics, and robotics. It is used to model and understand the behavior of physical systems, design and optimize structures, and create realistic computer-generated images.

5. What skills are necessary to work with differential geometry?

To work with differential geometry, one needs a strong foundation in mathematics, particularly in calculus, linear algebra, and differential equations. Knowledge of computer programming and visualization tools can also be helpful in applying differential geometry to real-world problems.

Back
Top