0
$\begingroup$

I’m trying to back-calculate a value in astronomy using spherical trigonometry. Using the following formula, where $δ$ = the Declination of the Galactic North Pole in degrees (i.e., the angle between the North Celestial Pole and the Galactic Equator).

Given that

$$\arccos(\sin(δ)\cos(23.43928°)-\sin(192.8583°)\cos(δ)\sin(23.439289°)) = 60.18894°$$

where

  • $23.43928° =$ obliquity of the Ecliptic (tilt of Earth's axis relative to Ecliptic North Pole)

  • $192.8583° =$ right ascension of Galactic North Pole in degrees

  • $60.18894° =$ angle between the Galactic and Ecliptic North Poles (or the angle between the Galactic and Ecliptic Planes)

  • $δ =$ declination of Galactic North Pole using Equatorial Coordinates

Can you solve for $δ$?

I've tried doing the calculation for δ in Excel, Wolfram Alpha, and "3.2. Galactic and equatorial coordinates" at https://aas.aanda.org/articles/aas/full/1998/01/ds1449/node3.html, but $δ$ is always given as a known value, i.e., 27.41 degrees, but I've had no luck in finding out how $δ$ is calculated in the first place.

You may ask, why do I want to calculate $δ$ if I already know its value? Call it intellectual curiosity, or the satisfaction of knowing the answer to something I've been struggling with for the past 5 years.

$\endgroup$
2
  • $\begingroup$ Don't spend an excessive amount of time on a single problem, that is how you fail. If the problem is reasonably complicated, then spend no more than a week trying to figure it out, then move on to the next problem. You have to learn how to manage your time and not spend that time on problems that will bring little to no value to your learning, and in your case, intellectual curiosity. $\endgroup$
    – Tachyon
    Commented Nov 10, 2021 at 1:16
  • $\begingroup$ I didn't spend an excessive amount of time on it in one go. I just kept coming back to it over time. I asked here because I thought I might get an actual answer to the question itself. $\endgroup$
    – user52025
    Commented Nov 10, 2021 at 2:51

2 Answers 2

3
$\begingroup$

Here is how to solve for $\delta$ the following trig expression involving $\sin \delta$ and $\cos \delta$.

$$ A \sin \delta+B \cos \delta = C \tag{1}$$

Presenting below are two different methods for the same problem. In your case, use one of the two to find $\delta$.

  1. Trig expansion

    Consider $\delta$ having two parts $\delta= \varphi + \psi$ and expand the trig function to

    $$ \cos \psi \left( B \cos \varphi+A \sin \varphi \right) + \sin \psi \left( A \cos \varphi - B \sin \varphi \right) = C \tag{2} $$

    Now you force $A \cos\varphi - B \sin\varphi =0$ since the choice of how to split $\delta$ is ours.

    $$ \varphi = \mathrm{atan}\left(\frac{A}{B}\right) $$

    And the equation (2) is now

    $$ \cos\psi \left( \sqrt{A^2+B^2} \right) + \sin \psi \left( 0 \right) = C $$

    which is solved for

    $$ \psi = \pm\, \mathrm{acos}\left( \frac{C}{\sqrt{A^2+B^2}} \right) $$

    Putting $\delta$ back together we have the solution

    $$ \delta = \mathrm{atan}\left(\frac{A}{B}\right) \pm\, \mathrm{acos}\left( \frac{C}{\sqrt{A^2+B^2}} \right) \tag{3}$$

  2. Trig substitution

    Consider the tan half-angle substitution, $t = \tan \left( \frac{\delta}{2} \right)$ which yields the following expressions $$ \cos \delta = \frac{1-t^2}{1+t^2} \tag{4}$$ $$ \sin \delta = \frac{2 t}{1+t^2} \tag{5}$$ $$ \delta = 2 \,\mathrm{atan}(t) \tag{6}$$

    And use these in (1) to get a polynomial in terms of $t$

    $$ A \frac{2 t}{1+t^2} + B \frac{1-t^2}{1+t^2} = C $$

    $$ (B+C) t^2 -2 A t + (C-B) = 0 \tag{7}$$

    Solve the quadratic for $t$ and use it in (6)

    $$\delta= 2 \mathrm{atan}\left( \frac{A \pm \sqrt{A^2 + B^2 - C^2 }}{B+C} \right) \tag{8} $$

Even though (3) and (8) are different expressions, the numeric results are the same. But note the usual quadrant restrictions that go with trig functions.


Consider the simple example of $2 \sin\delta+ 3 \cos \delta = 1$

If you apply any of the methods above you will find $\delta \approx -40.207818721 °$ and $\delta \approx 107.58795377 °$ as the two solutions.


Based on the comments by @DavidHammen there is a special case when $B+C=0$. In that case the solution to (7) is $t = \frac{C}{A}$ or

$$\delta = 2\, \mathrm{atan}\left( \frac{C}{A} \right)$$

$\endgroup$
11
  • 1
    $\begingroup$ You beat me to it. My partially written answer goes down the drain! Personally I like the tangent half angle expression (your equation 8) because that form makes the quadrant restrictions go away. The 360° (or $2\pi$) issue doesn't go away, but who cares? $\endgroup$ Commented Nov 10, 2021 at 9:02
  • $\begingroup$ @DavidHammen - You should submit your answer anyway, as there are always things to be learned from the nuances of different approaches. $\endgroup$ Commented Nov 10, 2021 at 9:05
  • $\begingroup$ My answer was your essentially equations (4) to (8). There's no need to write that up. There is one interesting case you didn't cover: What if $B+C=0$? $\endgroup$ Commented Nov 10, 2021 at 9:18
  • $\begingroup$ @DavidHammen, great suggestion. I have added this to the answer. $\endgroup$ Commented Nov 10, 2021 at 13:24
  • $\begingroup$ @user52025 in (3) $\theta = \varphi + \psi$ and $\varphi = {\rm atan}(A/B)$ $\endgroup$ Commented Nov 11, 2021 at 0:12
0
$\begingroup$

$$ \delta = \arcsin \left( \frac{\cos \left( \theta \right)}{\sqrt{a^{2} + b^{2}}} \right) + \arctan \left( \frac{b}{a} \right) = 27.13^{o} $$

where

$23.43928^{o} =$ obliquity of the Ecliptic (tilt of Earth's axis relative to Ecliptic North Pole)

$192.8583^{o} =$ right ascension of Galactic North Pole in degrees

$60.18894^{o} =$ angle between the Galactic and Ecliptic North Poles (or the angle between the Galactic and Ecliptic Planes)

With the substitutions:

$$ a = \cos \left( 23.43928^{o} \right) \approx 0.9174821 $$ $$ b = \sin \left( 192.8583^{o} \right) \cdot \sin \left( 23.439289^{o} \right) \approx −0.0885216 $$ $$ \theta = 60.18894^{o} $$

Calculations for deriving $\delta$ are at: https://astronomy.stackexchange.com/questions/53312/calculate-declination-of-galactic-n-pole-given-obliquity-of-ecliptic-rt-ascen/53313?noredirect=1#comment119196_53313

$\endgroup$

Your Answer

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