0

I'm writing a physics essay in high-school and have become stuck on a math problem related to the calculations that need to be done to interpret my measurements.

The measurement equipment I am using can detect particles from a square pyramid with the top in the measurement equipment and the base facing into space.

I need to calculate the surface area that is projected from the solid angle onto the earths atmosphere. Note that it is not the same as the surface area of the pyramid, but it is the surface area of the atmosphere it crosses. I know angle in the pyramid's top.

Thank you for any advice, I can't seem to find a good, free resource that covers this.

  • 2
    Are you asking to imagine a cone from the center of the earth ending at the base of your equipment and compute the area of the sides of the cone from the ground up to your equipment? What is the approximate altitude of your experiment? – Ross Millikan Jan 31 '23 at 15:37
  • @RossMillikan Rather imagine a pyramid from my equipment with the top facing down towards earth but ending at my equipment which is on surface level, and the base facing towards space. The altitude, or rather, length from my equipment to the atmosphere is 20 thousand kilometers. (Pfotzer Maximum). But this height is larger than the height to the base of the pyramid, since the pyramid is not spherical and I need the sphere's surface area. If you still have trouble understanding my description I could draw a picture for you. – albjohansson Jan 31 '23 at 15:44
  • Unless your base is very large the difference in area between the base and the part of a sphere cut off by the base is tiny. The radius of curvature is about 28,000 km. – Ross Millikan Jan 31 '23 at 15:48
  • @RossMillikan I would guess that the approximate surface area of the base is about 50 km^2. – albjohansson Jan 31 '23 at 15:50
  • If it is about 8 km in linear dimension the angle at the center of the earth is $\frac 8{28000}$ The length of the arc is then $28000 \arcsin \frac 8{28000}$ from Wikipedia. This differs from $8$ by $10^{-7}$ – Ross Millikan Jan 31 '23 at 16:03
  • It sounds like you want the area of a kind of spherical sector but made with a square pyramid instead of a cone? – user170231 Jan 31 '23 at 16:17
  • @user170231 That is exactly what I am looking for! – albjohansson Jan 31 '23 at 16:51

2 Answers2

0

Given that you want to find the measure of a kind of spherical sector formed by a square pyramid, equivalently you can measure the area of a spherical quadrilateral whose boundaries occur where the sides of the pyramid intersect a unit-radius sphere.

Assuming the apex of the pyramid is at the center of the sphere, each lateral face of the pyramid is a plane through the sphere's center and therefore intersects the sphere along a great circle. The angle $\theta$ where two of these great circles (made by two adjacent faces) meet is the dihedral angle between those two faces.

The sum of the interior angles of the quadrilateral is then $4\theta$ (all dihedral angles are equal since the base is square) and the area of the spherical quadrilateral is therefore

$$ S = 4\theta - 2\pi. $$

This is due to a theorem that the area of a spherical triangle with vertex angles $\alpha,$ $\beta,$ and $\gamma$ is equal to $\alpha+\beta+\gamma - \pi.$

Then $S$ is the measure of your solid angle in steradians.

David K
  • 108,155
  • Thank you very much for your answer! This seems to be exactly what I am looking for. As I'd like a bit more insight about why =4−2, and something to cite for my project, I was wondering if you could point me to a resource describing this theorem and how it leads to your formula (or explain to me)? – albjohansson Feb 01 '23 at 12:59
  • The basic formula is the one for the triangle; you can derive the formula for the square by drawing a diagonal, which divides the square into two triangles, and applying the triangle formula twice. The triangle formula is called the spherical excess if you want to look it up. – David K Feb 01 '23 at 13:12
  • Sorry for bothering once again, I thought I managed to calculate it yesterday but seems I am missing something. How would I calculate the dihedral angle? Just to make sure I've done things correctly. It is like angle B on page 4 in this pdf right: https://arxiv.org/ftp/arxiv/papers/2108/2108.05226.pdf – albjohansson Feb 01 '23 at 18:19
  • Yes, if I read the diagram correctly it is angle $B$. The dihedral angle between lateral sides of a square pyramid is also discussed in answers to Dihedral angle calculation in pyramid with square base ABCD and Compute the dihedral angle of a regular pyramid. Exactly how you do the calculation depends on exactly what data you start with. – David K Feb 01 '23 at 22:19
0

This partial answer relies on calculus. There are probably simpler solutions using more elementary methods. Slight adjustments need to be made to align with the pyramid you're actually using, but hopefully this serves as a decent starting point.

Let $S$ be the part of the sphere of radius $r$ bounded by the planes $z=k|x|$ and $z=k|y|$. By symmetry, the area of the spherical face is $4$ times the area of the part $S$ bounded by the planes $y=-x$ and $y=x$. The figure shows the phantom solid whose spherical face is $S$, and the quarter of it contained by $y=\pm x$ to either "side" and the plane $z=ky$ below. (Figure rendered with $r=k=2$; imagine the jagged edges are actually smooth)

enter image description here

The factor $k$ indicates the ratio of the pyramid's height to half its base length. Parameterize $S$ in spherical coordinates by

$$\vec s(u,v) = r\cos(u)\sin(v)\,\vec\imath+r\sin(u)\sin(v)\,\vec\jmath+r\cos(v)\,\vec k$$

where $u\in\left[\frac\pi4,\frac{3\pi}4\right]$ and $v\in\left[0,f(u)\right]$. The upper bound on $v$ is obtained by finding the intersection of $x^2+y^2+z^2=r^2$ and $z=ky$ :

$$\begin{align*} x^2 + \left(\frac zk\right)^2 + z^2 &= r^2 \\ r^2 \cos^2(u) \sin^2(v) + r^2 \left(1+\frac1{k^2}\right) \cos^2(v) &= r^2 \\ \sin^2(v) &= \frac1{k^2\sin^2(u)+1} \\[1ex] \implies f(u) &= \arcsin\left(\frac1{\sqrt{k^2\sin^2(u)+1}}\right) \end{align*}$$

Take the normal vector to $\vec s$ to be

$$\vec n = \frac{\partial \vec s}{\partial v} \times \frac{\partial \vec s}{\partial u} = r^2\cos(u)\sin^2(v)\,\vec\imath + r^2\sin(u)\sin^2(v)\,\vec\jmath + r^2\cos(v)\sin(v)\,\vec k$$

Then the surface element is

$$dA = \|\vec n\| \, du \, dv = r^2 \sin(v) \, du \, dv$$

and so the area of $S$ is given by the surface integral,

$$\begin{align*} \iint_S dA &= 4r^2 \int_{\frac\pi4}^{\frac{3\pi}4} \int_0^{\operatorname{arccsc}\left(\sqrt{k^2\sin^2(u)+1}\right)} \sin(v) \, dv \, du \\ &= 8r^2 \int_{\frac\pi4}^{\frac\pi2} \int_0^{\operatorname{arccsc}\left(\sqrt{k^2\sin^2(u)+1}\right)} \sin(v) \, dv \, du \end{align*}$$

(again, by symmetry). Adjust $k$ and the intercepts of the bounding planes as needed. Your pyramid does not have its vertex at the origin, rather on another sphere with radius slightly smaller than $r$.

user170231
  • 25,320