View Single Post
  #7  
Old March 24th 04, 04:35 PM
Victor
external usenet poster
 
Posts: n/a
Default

**** Post for FREE via your newsreader at post.usenet.com ****

Thank you Mike, but I just found the answer. It happens that the tropical
year length that I was using is actually the AVERAGE tropical year length.
This number changes depending on where on the Earth's orbit you measure the
tropical year from. So, we have that the vernal Equinox tropical year length
is about 15 seconds longer than the average. These 15 seconds multiplied by
a 2000 year period gives a more than 8 hours error, which is what I get.
Also, the 3 hours gap due to the shortening of the year's length doesn't
apply in this case since the vernal equinox tropical year length has
remained and will remain quite stable for a few millenia due to several
factors like the slowdown of the rotation, the acceleration of the mean
orbital motion, and the effect at the vernal point of shape changes in the
Earth's orbit's that happen to almost cancel out.
If you want more information go to:
http://en.wikipedia.org/wiki/Tropical_year

As I said before, thanks Mike.

"Mike Dworetsky" escribió en el mensaje
...


"Victor" no@spam wrote in message ...
**** Post for FREE via your newsreader at post.usenet.com ****

Mike,
I have measured how long does Redshift consider a tropical year to be

and
it
uses 365,242190 days (which is the current ptropical year length)

always.
So
I went and added up the difference between the current tropical year

length
and each year's tropical year length for all years from now to 4004AD.

The
total error is only about 3 hours.The displacement I find in the program

is
about 10 hours so there are about seven hours which remain unaccounted

for.
Another "discovery":
Set the clock to current year's vernal equinox. Set the time step to one
tropical year.Press the step forward button repeatedly. The sun will

move
back and forth from the from the vernal equinox apparently randomly with

a
tendency to move away from it on the long run. Can you or anyone

explain
please it?
Thanks

"Mike Dworetsky" escribió en el

mensaje
...


"Victor" no@spam wrote in message ...
**** Post for FREE via your newsreader at post.usenet.com ****

Ok, this is what I do:

I set Redshift's time to 03/20/2004 06h:48m which is the time for

the
vernal
equinox this year
Set the sun's image to "icon". Look at the sun, it is over the

vernal
equinox.
Now enter 2000 and select Tropical years in time step. Press time

step
forward once. The time now is 03/19/4004 15h:55m.
Look at the sun: It has displaced from the vernal equinox and you

need
to
move the time forward about 10 hours in order to put them together

again.
Anyone has an explanation?



I'm not sure, but it may be that the "tropical year" timestep is in

current
tropical year lengths, while the Earth's ephemeris calculation is in
terrestrial dynamic time with corrections to UT calculated by

extrapolations
of the UT-DT difference formula.

--
Mike Dworetsky


I would guess that the random shifts come from the actual tropical year
varying slightly due to planetary perturbations from year to year, while

the
mean value changes very slowly and smoothly. For example, there are
probably periodic terms with the Venus and Jupiter synodic periods.

As for why the shift is 10 hours rather than 3 hours, I have no idea, but

it
possibly has something to do with the way RedShift4 extrapolates the

future
rotation of the Earth based on past behaviour. If the formula has

quadratic
terms, errors can build up very fast. These would probably be questions
that should be addressed to the company that produced it.

I understand a new version is just out.

--
Mike Dworetsky

(Remove "pants" spamblock to send e-mail)







-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
*** Usenet.com - The #1 Usenet Newsgroup Service on The Planet! ***
http://www.usenet.com
Unlimited Download - 19 Seperate Servers - 90,000 groups - Uncensored
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=