Talk:Coriolis force/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

Talk:Coriolis effect/archive1

---

Some time ago I had written that the coriolis force only acts in tangential direction. I discovered that that statement of mine was wrong. The current article presents the physics better than before.

Currently, the article is huge. I'm not happy with that, but it is a very complicated subject. As I wrote earlier, I am very keen to present the physics of the coriolis effect in such a way that an understanding of the coriolis effect in atmospheric motion is prepared for as much as possible. --Cleon Teunissen | Talk 00:27, 10 Mar 2005 (UTC)

In-depth discussion of inertia belongs in Inertia article

William M Connolley commented: Remove inertia section. Without judging its quality, it doesn't belong here: it belongs in inertia.

Hi William,

I agree with the removal of the inertia stuff.
The only reason for writing the section on inertia at all was to justify that in the article the physics of the coriolis effect is explained in terms of laws of motion in an inertial frame of reference.

I think that very often the importance of the pervasive presence of inertia is under-appreciated.

Also, you removed the statement that 'the formula describes the no-drag situation'. I think you were correct in doing that. I wasn't sure at first but the formula can also be used to calculate the amount of force that is required to prevent any deflection due to coriolis effect, which is what is represented in the four diagrams. When deflection due to coriolis effect is forcefully prevented, an acccelerometer will give a reading. I propose to call that force 'anti-coriolis force'. Anti-coriolis force is to coriolis force what centripetal force is to centrifugal force; the same in magnitude, but in opposite direction. --Cleon Teunissen | Talk 23:35, 10 Mar 2005 (UTC)

Should the 'rotating mercury mirror' stuff be moved to an article of its own?

Its not uncommon to have article titles of the following form: field (physics)
Which can also be linked to by writing [[field (physics)|field]], showing up as field
I wonder, would it be better if an extra article is created, with the current article coriolis effect giving a simplified exposition of the coriolis effect, and a specialized article [[Coriolis effect (meteorology)]] explaining the full coriolis effect.

The rotating mercury mirror stuff is difficult physics, but the formula F=2mwv is valid only in that context, the formula F=2mwv is not valid on a flat rotating disk. --Cleon Teunissen | Talk 10:15, 14 Mar 2005 (UTC)

Moving the formula; proposal for change

(William M. Connolley 20:12, 24 Mar 2005 (UTC)) I moved the formula to the top, having experienced frustration myself in trying to find it. I also think the header needs to be modified... thusly...

The Coriolis "effect" is primarily a change-of-coordinates effect. In this, it is primarily a kinematic thing, which you derive from the mathematics, and observe that an acceleration term appears in the equations when viewed from different frames of reference. There is no requirement that either frame be inertial; indeed there is no physics at all in this initial POV: all we say is that if F1 and F2 are two (Euclidean) frames of reference, rotating at omega wrt each other, with accelerations (measured wrt frames F1 and F2 respectively) a1 and a2, then a1 = a2 + 2 * v cross omega (sign?). From this, it is easy enough to introduce the physics: that if the equations of motion in F1 are F = m * a1, then in F2 the equations are F = m * a2 + 2 * m * v cross omega.

Actually thats not a proposed intro; its a proposal from which to write an intro.

It might also clarify things to change the name of the page. Coriolis effect is odd; Coriolis force is possible; I would (from the above) prefer Coriolis acceleration, with redirects.

I disagree with this proposal. In physics a change of coordinate system does not have any effect on the actual physics. On the rotating earth air moves the way it moves, independent of the choice of reference frame of the meteorologists. The physics of the coriolis effect is that on a rotating planet air currents move different from how they would move on a non-rotating planet. The physics of the coriolis effect is not caused by rotation of the coordinate system, it can occur when physical object are actually rotating.

Another example is how a physicist calculates how much Earth will bulge at the equator. Some choices of coordinate system make the calculation more tedious than others, but all calculations that are performed properly will predict the same ellipsoidal shape. Physiscs is not about the choice of coordinate system, physics is about the underlying principles that are independent of the choice of coordinate system. --Cleon Teunissen | Talk 21:12, 24 Mar 2005 (UTC)

---

(William M. Connolley 20:12, 24 Mar 2005 (UTC)) The Coriolis "effect" is primarily a change-of-coordinates effect. In this, it is primarily a kinematic thing, which you derive from the mathematics, and observe that an acceleration term appears in the equations when viewed from different frames of reference.

Which terms appear when is a matter of mathematical detail. William, you are a climate modeler. Meteorologists use the formula for the coriolis force all the time. Whenever they apply the formula for the coriolis force, they use the sidereal day as the rotation rate of Earth. The rotation rate of earth is not a difference between two arbitrarily chosen frames, it is the rotation of Earth with respect to the local (Earth enveloping) inertial frame. When rotation is involved you only get correct calculation outcomes if you apply the correct rotation rate, the rotation rate with respect to the local inertial frame.

In your opinion, should meteorologists stop using the period of 23 hours 56 minutes 4 seconds as the rotation rate, and start using arbitrary values? If not, why? --Cleon Teunissen | Talk 21:33, 24 Mar 2005 (UTC)

(William M. Connolley 20:05, 28 Mar 2005 (UTC)) I really don't see the point in what you're saying here, so I suspect you've misunderstood mine. You write about local inertial frames: this is not relevant. Given one arbitrary frame, and equations valid in that frame, then the maths tells you what the equations are wrt a frame rotating wrt the first frame. Thats all. No mystery. Of course, if you want the maths to match reality, you have to use the correct rotation rate.

Yeah, the misunderstanding is still there. I'll try something else that I think/hope will be helpful in getting at the root of the misunderstanding. It is about how demanding to be when wanting the math to match reality

The derivations of coriolis acceleration that I've encountered so far are of the following type. A stationary system is taken. Then a rotating coordinate system is taken. Seen from the perspective of the rotating coordinate system everything that is stationary seems to be in circular motion. A hypothetical force can be retrofitted to that apparent motion.
Then something is taken that is moving in a straight line as seen from the perspective of the stationary system. Seen from the perspective of the rotating coordinate system, this looks like curvilinear motion, a spiral that winds around the center of rotation, and the corresponding acceleration can be retrofitted with a force to match that apparent acceleration.
To me that looks like just a mathematical exercise in projective geometry, no physics there.

(William M. Connolley 21:04, 29 Mar 2005 (UTC)) Its not projective geometry of course - thats a different subject. But yes, its just geometry/kinematics (providing you stick to accelerations not forces). And of course the first frame doesn't have to be "stationary" - the term doesn't mean anything by itself anyway - all that matters is the relative motion of the two frames.

What meteorologists are faced with is the actually rotating Earth, and they use a co-rotating coordinate system, and to my surprise, the hypothetical, retrofitted force law works too in describing the actual physics of mass that moves with respect to the rotating physical system.

I am not satisfied with the 'projective geometry' derivation, because the context in which that derivation is performed is different from the context in which the resulting accceleration rule is subsequently applied.

(William M. Connolley 21:04, 29 Mar 2005 (UTC)) I don't understand what you mean here. Nor why you find the rotating earth in some way different.

In the current coriolis force article I attempt to bridge that gap. I model the physics by way of the rotating mercury mirror, which has a period of 12 seconds. The curvature of this paraboloid surface acts effectively as a centripetal force field. Stable orbits in that centripetal force field are elliptical in shape, and thay have a period of 12 seconds. Seen from the perspective of a co-rotating coordinate system, the elliptical orbit transforms to a small circular motion, with a period of 6 seconds. That small circular motion, with 2 times the angular velocity of the system's angular velocity, that is how I attempt to construct the coriolis acceleration. I want to derive the coriolis acceleration in the same physical context as in which it is to be applied. --Cleon Teunissen | Talk 15:01, 29 Mar 2005 (UTC)

---

(William M. Connolley 21:04, 29 Mar 2005 (UTC)) Its not projective geometry of course - thats a different subject. But yes, its just geometry/kinematics (providing you stick to accelerations not forces). And of course the first frame doesn't have to be "stationary" - the term doesn't mean anything by itself anyway - all that matters is the relative motion of the two frames.
(William M. Connolley 21:04, 29 Mar 2005 (UTC)) I don't understand what you mean here. Nor why you find the rotating earth in some way different.

Cleon Teunissen | Talk 16:50, 31 Mar 2005 (UTC) What I meant is that the usual derivation of a coriolis acceleration is done by mapping one space onto another, so I put it in the mathematical category of 'projection'. I wasn't thinking of linear projection, of course.
In the physics of the coriolis effect only two reference frames are practical: the frame that is stationary with respect to the local inertial frame of reference, and the frame that is co-rotating with Earth.
What's different about the rotating Earth is that manifestation of inertia of the air of the atmosphere is rather like the manifestation of inertia of objects in orbit around a planet, and very different from the inertial effects in the atmospher of a non-rotating planet. Because of that there is for example inertial wind, wich would be highly unstable on a non-rotating planet. --Cleon Teunissen | Talk 16:50, 31 Mar 2005 (UTC)

Naming the coriolis physics

(William M. Connolley 20:12, 24 Mar 2005 (UTC)) It might also clarify things to change the name of the page. Coriolis effect is odd; Coriolis force is possible; I would (from the above) prefer Coriolis acceleration, with redirects.

The name 'Coriolis effect' is not neccessarily odd. In physics, the word 'effect' is used in a very flexible way. In the case of the Coriolis effect, it says that coriolis dynamics is a 'factor' in the dynamics of wind.

A sailboat provides an example of a deflection effect. A sailboat can maintain speed against water resistance with wind blowing at a right angle to the direction of motion. The sail and the keel combined act to deflect the force of the wind through an angle of 90 degrees. When the wind picks up, the velocity of the sailboat picks up, so the force of the wind is doing work, even though the direction of the wind is perpendicular to the direction of motion of the boat. It is uncommon to call that 'the sail effect' but as a name it fits. The sail effects a "ricochet" of the wind.

The action of the sail is that wind that is initially moving perpendicular to the direction of the boat is accelerated in a direction pointing to the rear of the boat. Newtons third law descibes the dynamics of that situation: the force that is exerted by the sail on the wind, accelerating the wind towards the rear of the boat, is mirrored by the force exerted by the wind on the sail, pushing towards the direction of motion of the boat. So the sail is doing work in two opposite directions simultaneously. That's the sail effect. --Cleon Teunissen | Talk 03:14, 26 Mar 2005 (UTC)

the local inertial frame of reference

(William M. Connolley 21:04, 29 Mar 2005 (UTC)) "stationary" - the term doesn't mean anything by itself anyway

Yet in everyday language people often use the word 'statonairy' and they have no problem understanding each other. But yeah, I could have used the full expression: 'A system that is stationary with respect to the local inertial frame of reference'. For example, astronomical calculations of the orbits of the planets of the solar system are performed in the local inertial frame of the solar system. In that frame the orbits of all the planets can be described by an inverse square law of gravitation, etc etc. It would be very cumbersome to use a frame of reference that is rotating with respect to the local inertial frame of reference, because the math would be much more complicated then. The local inertial frame of reference is the only frame in which all the planets' motions can be described by a single law of gravity.
In everyday language (and in physicists' language) the local inertial frame of reference is the default frame. If somebody says: 'stationary' without going to the length of specifying with respect to what frame, then the default frame is meant.

The local inertial frame of reference of the solar system is co-moving with the common center of mass of the solar system. The local inertial frame of reference of The Earth-Moon system is co-orbiting the Sun with the common center of mass of the Earth-Moon-system. The local inertial frame of reference of satellites orbiting Earth is co-moving with the center of mass of the Earth. The local inertial frame of reference of a space station orbiting Earth is co-orbiting with the center of mass of the space station. The local inertial frame of reference of a person on the surface of Earth is by and large co-moving with that person.

This gives an idea of what in physics is meant by the expression: 'local inertial frame of reference'. All of the above frames of reference are different from each other in that they are moving in very complicated ways with respect to each other, but they have in common that they do not rotate with respect to each other. All the motions of the planets, their rotations, the motions of moons, artificial satellites, they are "in tune" with each other as far as rotation of the frame of reference is concerned, they never "drift". In physics, that which they are all in tune with is very important, it correlates many aspects of physics: space-time geometry correlates these things. --Cleon Teunissen | Talk 16:55, 31 Mar 2005 (UTC)

(William M. Connolley 20:02, 31 Mar 2005 (UTC)) You're missing the point, which is that there is absolutely no need for a local inertial frame: its completely irrelevant. The coriolis acceleration/force is a change-of-coordinates force, which is manifested when you view a given system in a different (rotating wrt the first) frame.

I am trying to figure out how your statement might relate to the physics of air movements in the atmosphere of the Earth. What you describe, can that also work the other way round? As you understand coriolis effect, does that mean that a particular change of view in a different frame can make the coriolis effect vanish? For example, the coriolis effect is involved in the trade winds. Do you believe that it is possible to stop the trade winds by a particular change-of-coordinates? --Cleon Teunissen | Talk 22:03, 31 Mar 2005 (UTC)

I'm not pulling your leg. What I am thinking about is that it must work both ways. If a change-of-coordinates causes the coriolis effect, then the inverse change of coordinates should make it disappear. I'm pretty sure the coriolis effect cannot be undone that way. Of course I do see the parallels between the geometrical mapping of two spaces and the physics, it's no coincidence that the geometrically derived formula works fine in the real world. But the two worlds are not identical. --Cleon Teunissen | Talk 22:16, 31 Mar 2005 (UTC)

(William M. Connolley 08:51, 1 Apr 2005 (UTC)) Well of course you can make the CE vanish. All you need to do is transform back into the original frame. But this has no effect on the *physics* of course. This is just like gravity: you can make it vanish, locally, by an appropriate transform. But that doesn't stop you "falling".

Cleon Teunissen | Talk 10:24, 1 Apr 2005 (UTC)I wasn't sure at first, but you are talking about the transforms as they are available in general relativity.

Different aspects of general relativity can be applied in different ways. For example, back in 1918, Herman Weyl attempted to calculate the following: If a hollow sphere of mass, say about as large as a Galaxy, is rotating with respect to the universe as a whole at a rate of one revolution per day, what will the space-time geometry inside it be like? (Physically impossible of course, such a rotation, but Weyl wanted to explore to what extent general relativity is symmetrical). It was hoped that the calculation would show that on the inside the space-time geometry would literally be a centrifugal force field (including coriolis force field effects). A real, physical centrifugal force. Inside, space-time geometry would be rotating with respect to the whole of the universe outside. To the disappointment of many, it didn't work out that way. General relativity does predict frame dragging but it is not strong enough to make rotation really reciprocal. Anyway, that is how an actual coriolis force could be generated, if it would be possible for a Galaxy to rotate that fast.

Another aspect of general relativity is the mathematical form it is in, the formulation has build-in geometric transforms. When a geometric mapping is performed then corresponding inertial terms automagically appear in the equations. If the higher order terms are ignored, this relativistic coriolis formula is the same as the coriolis formula of classical mechanics. These inertial effects appear in the equations because the equations have been carefully designed to comply with what happens in nature.

I prefer not to say that the coriolis effect occurs Because the laws of general relativity say they should occur. I think inertial effects occur because mass has inertia. So yeah, geometric mapping is a very useful tool in handling this type of calculations, but the coordinate transform is far from the essence of an actually occurring coriolis effect --Cleon Teunissen | Talk 10:24, 1 Apr 2005 (UTC)

Remove reference to 'bad coriolis' site

I have removed the reference to the bad coriolis site, because it contains bad physics. On that site it is stated:

It is true that the Coriolis force does no work. Bad coriolis FAQ

There is more to the story than that. The coriolis force is involved in sustaining cyclonic flow. Contraction of cyclonic flow has the effect of sustaining the velocity of the cyclonic flow, and this is mediated by the coriolis force. At one end there is the pressure gradient force, and at the other end there is the kinetic energy of the velocity of the air in cyclonic flow.

It is somewhat analogous to the sail of a sailing ship. The sail is doing work on the wind blowing into the sail, changing its direction, and the sail is doing work on the boat, pushing it through the water. When directions of work being done are added up the resultant amount of work done by the sail adds up to zero. Overall the work done on the boat, pushing it through the water is correctly assigned to the wind blowing in a direction perpendicular to the direction of the boat. But without the mediation of the sail the boat would not move.

When directions of work are added up the work done by the coriolis force drops out of the calculation, and the physics taking place can be seen as pressure gradient force sustaining the velocity of the cyclonic flow. If the coriolis force would not do any work at all, it would not affect the flow of air masses. --Cleon Teunissen | Talk 12:18, 12 May 2005 (UTC)

comment by 203.173.174.94

203.173.174.94 wrote:

(unclear! You need to define 'the inertial frame'! Also why should the hovercraft move if it is not interacting with the mercury pool, and has no propulsion on?)

In this particular case the inertial frame is self-evident. The mercury mirror is rotating with respect to the inertial frame, hence the shape of the mercury becomes concave. The faster the rotation of the mirror with respect to the inertial frame, the stronger the concavity. In the absence of any other measuring equipment the inertial frame can still be unambiguously be identified by measuring the concavity of the mercury mirror.

In the first example the hovercraft moves because it is released from the rim of the concave surface. Gravity pulls the hovercraft down, so the hovercraft will aquire a velocity parallel to the surface, and (initially) in the direction of lowest point. --Cleon Teunissen | Talk 13:13, 4 Jun 2005 (UTC)

Comment by 68.42.138.135

68.42.138.135 wrote:

Be sure you understand that the Coriolis "force" is a fictitious force, not unlike centrifugal force. It only appears to be a necessary force from within the rotating frame of reference.

It should be understood that the Coriolis Effect is real physics. The magnitude of the coriolis effect is independent of the choice of frame of reference of the observer.

I don't believe this. The motion of the frame of reference clearly appears in the defn of the acceleration. William M. Connolley 2005-06-28 13:12:56 (UTC).

The suggestion that the coriolis effect is some sort of peek-a-boo physics, appearing and disappearing as a result of observer action, is incorrect. The only thing that changes if an inertial frame of reference is swapped for a rotating frame of reference is the mathematical representation of the physics taking place, but the physics itself is not changed by a change in point of view of the observer.

The Coriolis Effect involves inertia. Inertia is very much not fictitious.

The parallel with centrifugal force is as follows: in order to sustain circular motion a centripetal force must be exerted. Whenever an object is being accelerated, the acceleration is opposed by inertia. Inertia always opposes acceleration, but it never prevents the acceleration. The inward acceleration due to the centripetal force is opposed (but not prevented) by inertia pointing in the direction opposite to the force being exerted. This opposition from inertia is in the context of circular motion usually called 'centrifugal force'.

From a physics point of view the expression 'centrifugal force' is awkward. By general agreed definition of the concept, 'a force' has the property that it can cause acceleration with respect to the local inertial frame of reference. Unlike a force, inertia cannot cause acceleration with respect to the local inertial frame of reference, so it is inconvenient to call inertia a force. All in all, inertia is best called 'inertia'. The expressions 'centrifugal force' and 'coriolis force' are physics shorthand, they should be understood as referring to effects of inertia.

Every rotating frame of reference refers to the (local) inertial frame of reference.
In order to perform any calculation at all in the context of a rotating frame of reference, it must be established what the angular velocity of that rotating frame is with respect to the (local) inertial frame of reference, otherwise no meaningfull calculation can be set up. A physically meaningfull calculation in the context of a rotating frame of reference derives its meaningfullness from referring to the (local) inertial frame of reference.

Of course, if a fictitious angular velocity is used in the calculation, then the calculation does not represent reality. A calculation that uses the actually measured angular velocity will represent reality, and it will describe the coriolis effects that are actually going on. If a fictitious angular velocity is used in the calculation, then the calculation will yield a fictitious description of a coriolis effect.

In physics we have the situation that newtonian dynamics has been superseeded by relativistic dynamics. Many aspects of physics have been deeply reconceptualized in that change, but inertia is one of the concepts that is seen the same in newtonian dynamics and relativistic dynamics.

Relativistic physics is highly counterintuitive. Velocity is relative; there is no such thing as measuring your own velocity with respect to space, so there is no such thing as a strictly local measument of one's velocity. On the other hand, any accelerometer measures its own acceleration with respect to the local inertial frame of reference (I follow the standard definition: the local inertial frame of reference is the frame that is co-moving with free-falling test-masses) That is, an accelerometer measures its interaction with space-time itself. In relativistic physics, space-time is very much not a neutral background. Matter/energy interacts with space-time itself, space-time opposes acceleration of objects moving in it. --Cleon Teunissen | Talk 28 June 2005 07:55 (UTC)

Balancing the books

It should be understood that the Coriolis Effect is real physics. The magnitude of the coriolis effect is independent of the choice of frame of reference of the observer. Cleon Teunissen | Talk 28 June 2005 07:55 (UTC)

I don't believe this. The motion of the frame of reference clearly appears in the defn of the acceleration. William M. Connolley 2005-06-28 13:12:56 (UTC).

I will rephrase the statement.

It should be understood that the Coriolis Effect is real physics. The magnitude of the coriolis effect as it physically occurs in the atmosphere of a planet is independent of the choice of frame of reference of the observer.

Still wrong. If you sit in a ref frame of the fixed stars and do your equ's for the motion of the earths atmos, *there is no coriolis effect*. But we've been all over this before, why say it all again? William M. Connolley 2005-06-28 21:31:35 (UTC).

Of course the motion of a rotating frame of reference with respect to the local inertial frame of reference appears in the defn of the acceleration. If it wouldn't, calculations would come out wrong.

To describe acceleration correctly, the acceleration with respect to the local inertial frame of reference must be measured. To describe rotation correctly, the rotation with respect to the local inertial frame of reference must be measured.
(I use the standard definition: the (local) inertial frame of reference is the frame that is co-moving with free-falling test masses.)

In the end the books must be balanced, otherwise the description of the physics will be incorrect. Suppose you have a planet that rotates with an angular velocity with respect to the local inertial frame of one revolution per 24 hours, and suppose you use a rotating frame of reference with an angular velocity with respect to the local inertial frame of one revolution per 12 hours (both counterclockwise as seen from the north). Then, to get a correct result from the calculation, the angular velocity of the planet with respect to the rotating frame comes out as one revolution per 12 hours, clockwise as seen from the north.

So it is bookkeeping. No matter how the calculation is set up, no matter how the numbers are arranged (provided they are arranged selfconsistently), if the planet has an angular velocity of 1 rev per 24 hours, then the calculation must in some form or another make sure that that angular velocity of 1 rev per 24 hours is represented in the calculation. Of course, the correct angular velocity is usually inserted in the calculation by defining a rotating frame, co-rotating with the planet.
--Cleon Teunissen | Talk 28 June 2005 21:02 (UTC)

No need to bother with the fixed stars

If you sit in a ref frame of the fixed stars and do your equ's for the motion of the earths atmos, *there is no coriolis effect*. But we've been all over this before, why say it all again? William M. Connolley 2005-06-28 21:31:35 (UTC).

Yeah, there is no point in saying it all again.

I am curious though, whether you acknowledge the physics concept of measuring rotation with respect to the local inertial frame of reference. I am curious what according to you can by found out by means of strictly local measurement by a memoryless instrument. In the past you have expressed disbelief.

The following device measures its own rotation with respect to space itself, its reference is space itself, not the fixed stars.

External link Gyrocompass for military field use From the site:
The SIAGL is a portable gyrocompass used to determine a true north direction at any selected station, [...] Accuracy of the SIAGL is comparable to that of astro observation. [...] The SIAGL can determine true north with high accuracy without the help of celestial or landmark sightings. [...] The instrument can determine and indicate true north within about 20 minutes after power is applied. [end quote]

When the soldiers arrive on the field, they take the gyrocompass out of the transport case, it is mounted on its tripod, power is applied, the gyroscope inside spins up, and after 20 minutes of data-gathering the device has obtained enough to accurately indicate geographic North. Unlike a GPS device, or a magnetic compass, the gyrocompass does not use any outside source of information.

The casing of the gimbal-mounted gyroscopewheel is rotating with respect to local space, for it is co-rotating with Earth. That information is sufficient for the device to construct and display the direction of the geographic north. No operator-calibration is required; the device self-calibrates by looking inside, by referencing to space itself.
--Cleon Teunissen | Talk 29 June 2005 08:47 (UTC)

I'm not sure what the point of all that is. WRT the fixed stars, there is a coriolis accn. I don't see any reason why that shouldn't be measureable. William M. Connolley 2005-06-29 19:35:42 (UTC).

It is unclear why you keep invoking the fixed stars. I have repeatedly stressed that the reference of the measurements taken by a gyrocompass is not the fixed stars. A gyrocompass performs its task without the help of celestial or landmark sightings. The reference of a gyrocompass is local space itself. It appears that you do not acknowledge this aspect of the operating principle of gyrocompasses.

That is my point. It appears that there is something that measuring devices can measure that you do not acknowledge. A gyrocompass by itself cannot be used to say something about rotation with respect to the fixed stars, since a gyrocompass only looks inside, not outside. (Of course, it is general knowledge that the space of the solar system does not rotate significantly with respect to the fixed stars, but I'm focusssing here on what a gyrocompass by itself can and cannot do.)
--Cleon Teunissen | Talk 30 June 2005 06:13 (UTC)

More good info on coriolis effect by Anders Persson

There is the article by Anders Persson, that was published in the Bulletin of the American Meteorological Society, that is a good source of information about the coriolis effect, and I found a more recent publication by Anders Persson on the internet site of the Norwegian Meteorologial institute. This article is titled The Coriolis Effect – a conflict between common sense and mathematics.

Anders Persson confirms that the proper way to create an understanding of the coriolis effect is to get a feel for the dynamics of the physics taking place.

Persson writes on page 9 of the PDF-document:
A common way to illustrate the Coriolis deflection is to throw or roll a ball over a turntable rotating anticlockwise.[...] However, most of what we see here is not the Coriolis force, but the centrifugal force. If only the Coriolis force had been present the ball would return after having performed a circular orbit. [...] However, we can cancel out the centrifugal force by deforming the turntable into a slightly concave parabola.

Such a parabolic dish was in use by professor Norman A. Phillips in meteorological education at the Massachusetts Institute of Technology in the 1950’s. [...] Instead of a ball, Phillips used a puck, which consisted of a hollow plastic cylinder, about 1 inch in height and 1 inch in diameter. The puck was made to float on the cement disc by filling it with liquid nitrogen and capping it. The evaporating gas escapes out of the bottom, acting as a near perfect lubricant for the puck.
[end quote]

A recommendation of Anders Persson in the Climate and Weather Systems course given at the University of Saint Andrews A quote from Lecture 5
The best discussion of the Coriolis effect is in three recent articles by Anders Persson in the journal Weather. Understanding these still requires focused thought, but at least the explanations make sense (so it is a lot easier than trying to understand something that doesn't make sense!). [end quote]
--Cleon Teunissen | Talk 29 June 2005 18:58 (UTC)

In the magazine Weather published by the Royal Meteorological Society 18 articles about meteorology by Anders Persson have been published, most of them about coriolis physics.

It is clear to me that Anders Persson is to be considered the leading authority on the subject of coriolis physics in meteorology. His treatment of the subject is better than any other available on internet. The wikipedia article about the coriolis effect should be based on the insights provided by Anders Persson. --Cleon Teunissen | Talk 30 June 2005 09:24 (UTC)

A parabolic rotating turntable

Schematic representation of a puck moving frictionless over a parabolic surface.

The lecturer John Marshall, lecturing at MIT, lets undergaduate students work with a parabolic rotating turntable, so they can get a feel for the coriolis dynamics by watching it happening.

I will do an extensive rewrite of the coriolis effect article, following as much as possible the procedure of the physics demonstrations of that undergraduate course. This requires two animated GIF's, one of the motion of a puck as seen from the inertial frame, and on as seen by a co-rotating camera.
--Cleon Teunissen | Talk 1 July 2005 01:57 (UTC)

A co-rotating videocamera

Schematic representation of a puck moving frictionless over a parabolic surface, as seen by a rotating videocamera.

As seen from the point of view of a co-rotating videocamera, the eccentricity of the elliptical orbit looks like a small circular motion. --Cleon Teunissen | Talk 2 July 2005 10:48 (UTC)