Estimating the velocity of seismic waves through an idealized Earth

AI Thread Summary
The discussion focuses on estimating the velocity of seismic waves through an idealized Earth model using geometric relationships. A P wave takes 14 minutes to travel 110 degrees, leading to a calculated mantle velocity of approximately 12.43 kilometers per second and a core radius of about 3654 kilometers. The conversation highlights the need for additional data to accurately determine the velocity of seismic waves in the core, as the provided information is limited to the shadow zone and the Earth's radius. The assumption of constant velocity complicates the model, as real seismic waves increase in speed with depth. The participants agree that more precise data is necessary to refine the calculations and improve the model's accuracy.
Remixex
Messages
57
Reaction score
4
Homework Statement
Suppose that seismic observation shows that P wave arrives 14 min later at 110 degrees epicentral distance for shallow earthquakes. Between 110 and 140 degrees is the P wave shadow zone. Assuming the Earth can be simplified as two homogeneous parts, the mantle and the core, derive and plot the P travel-time curve T(∆). Estimate P velocities of the mantle and core and the size of the core from the observation. Assume the radius of the Earth is Rt = 6371 km.
Relevant Equations
Snell's law, trigonometry relationships, arc length of the Earth.
Hello, this is a repost from a much less-clear question I posted before (link to question: https://www.physicsforums.com/threa...s-inside-an-ideal-earth.1011998/#post-6596165).

It's kind of a loaded question, however it can be expressed as triangles within a circle or a sphere, as shown in the image. Given the available information (that the P wave takes 14 minutes to travel 110 degrees, or 12232 kilometers of arc length) I obtained the linear distance the P-wave travels through in the mantle, I named it ##a## and it is about 10438 kilometers. With this we can get the velocity through the idealized mantle, at about 12.43 kilometers per second. (##\frac{a}{14*60} km/s ##)

Then, using half of this distance ##a/2##, knowing it is tangent to the core and that it can form a right triangle with a distance ##d## that goes from the source to 55 degrees of epicentral distance and the distance from the surface of the core to the surface of the mantle, named ##R_m##. ##d=2*R_t*sin(55/2)## and then we can obtain ##R_m=\sqrt{d^2 - 0.25*a^2}## which is about 2717 kilometers, therefore the core radius is 3654 kilometers.

We can also do another right triangle, composed of the radius of the Earth, the radius of the core, and the distance ##a/2##. With this we can get the take-off angle ##\beta = Acos(((a/2)^2 + R_t^2 - R_c^2)/(a*R_t))##, which is 35 degrees.

Now it gets a little hairy, to get the velocity at the core we must apply Snell's law. For this I assumed there is a second ray that enters the core with a take off angle ##\alpha##, which is known, an incident angle ##\theta_1## and refracts with angle ##\theta_2## at the surface of the core.

The triangle composed by ##R_t##, ##R_c## and the distance traveled by this new ray ##c## can be completely obtained through law of cosines, given that we also know the take-off angle ##\alpha##. Using this triangle I obtained the incident angle ##\theta_1## (equations in the image 1, 2 and 3)

This is where I get stuck. Equations 4 through 6 are my attempts at trying to get the refracted angle ##\theta_2##, which is key to obtain the velocity through the core via Snell's law. But I could only reach a bigger angle ##\epsilon=\delta + \theta_2##.

I feel like I am missing something, a triangle I have not drawn to obtain ##\theta_2##.

Thank you for reading. Any direction is greatly appreciated.

Fm3XZSs - Imgur.jpg
 
Physics news on Phys.org
Remixex said:
Homework Statement:: Suppose that seismic observation shows that P wave arrives 14 min later at 110 degrees epicentral distance for shallow earthquakes. Between 110 and 140 degrees is the P wave shadow zone. Assuming the Earth can be simplified as two homogeneous parts, the mantle and the core, derive and plot the P travel-time curve T(∆). Estimate P velocities of the mantle and core and the size of the core from the observation. Assume the radius of the Earth is Rt = 6371 km.
Relevant Equations:: Snell's law, trigonometry relationships, arc length of the Earth.

Hello, this is a repost from a much less-clear question I posted before (link to question: https://www.physicsforums.com/threa...s-inside-an-ideal-earth.1011998/#post-6596165).

It's kind of a loaded question, however it can be expressed as triangles within a circle or a sphere, as shown in the image. Given the available information (that the P wave takes 14 minutes to travel 110 degrees, or 12232 kilometers of arc length) I obtained the linear distance the P-wave travels through in the mantle, I named it ##a## and it is about 10438 kilometers. With this we can get the velocity through the idealized mantle, at about 12.43 kilometers per second. (##\frac{a}{14*60} km/s ##)

Then, using half of this distance ##a/2##, knowing it is tangent to the core and that it can form a right triangle with a distance ##d## that goes from the source to 55 degrees of epicentral distance and the distance from the surface of the core to the surface of the mantle, named ##R_m##. ##d=2*R_t*sin(55/2)## and then we can obtain ##R_m=\sqrt{d^2 - 0.25*a^2}## which is about 2717 kilometers, therefore the core radius is 3654 kilometers.
I'm with you until here.

My question is: what other data do you have to work with now? Is it only the 110-140 degree shadow zone? Nothing else? I guess seismic waves don't make it directly though the centre of the Earth?
 
PS when I look up the radius of the Earth's core, I find that it's ##3,486 \ km##. That implies the lower shadow angle should be about ##114## degrees, rather than ##110##. Also, ##14## minutes seems quite a round number given the precision of the other data.
 
PPS okay, I understand this shadow angle now. It's just ##360 - (2 \times 110) = 140##.

That leaves us needing some addition data to make any further progress.
 
  • Like
Likes Remixex
PeroK said:
I'm with you until here.

My question is: what other data do you have to work with now? Is it only the 110-140 degree shadow zone? Nothing else? I guess seismic waves don't make it directly though the centre of the Earth?
In this case they can make it through, if the take-off angle were to be ##0## degrees it would make it right through and arrive at 180 degrees. However we do not have a travel time for that wave in this problem. The only data given is the radius of the Earth and the velocity of the P wave in the mantle (after some minor manipulation I outlined)
PeroK said:
PS when I look up the radius of the Earth's core, I find that it's ##3,486 \ km##. That implies the lower shadow angle should be about ##114## degrees, rather than ##110##. Also, ##14## minutes seems quite a round number given the precision of the other data.
Regarding this. In the real case, the velocity of waves increases with depth, so the waves actually draw an arc. In this problem this velocity is assumed constant, therefore waves travel linearly. The assumption of linearity exaggerates the core dimensions.
1200px-Earthquake_wave_shadow_zone.svg.png
 
Remixex said:
In this case they can make it through, if the take-off angle were to be ##0## degrees it would make it right through and arrive at 180 degrees. However we do not have a travel time for that wave in this problem. The only data given is the radius of the Earth and the velocity of the P wave in the mantle (after some minor manipulation I outlined)
Okay, you can get the core radius and the velocity through the mantle from that. But, you need more data to estimate the speed through the core.
 
  • Informative
Likes Remixex
Thank you very much for your advice. The wording of this problem is a little ambiguous, so I will take some liberties and assume that the wave travels through the Earth and reaches 180 degrees takes 35 minutes (similar to a model called PREM) and go on from there. Should be a straight shot to get the rest given that.
PeroK said:
Okay, you can get the core radius and the velocity through the mantle from that. But, you need more data to estimate the speed through the core.
 
Back
Top