# Flight QuestionWhat is changing Phobos' orbit?

#### Kolodez

##### New member
Hey, do you know why the orbit of Phobos seems to change so fast? When it approaches its apoapsis (on 2021-12-01), in Orbit MFD, the "ApT" value of Phobos decreases at 1.14 seconds per second, meaning that the apoapsis location is moving about 0.14 seconds (295 meters) backwards every second. If its orbit was nearly circular, I would understand, but it is very elliptic.
As I am doing my own gravitational calculations, my calculated orbit and the "real" orbit in Orbiter do not match anymore.
Thank you

#### Ajaja

##### Active member
How did you calculate the value 1.14?

#### Ajaja

##### Active member
Apoapsis and periapsis of Phobos are not stable and wobbling around some point.
There is a graph from GMAT (orbit of Phobos here is based on MAR097 ephemeris file):

#### Kolodez

##### New member
1.14 seconds per second because, for instance, Orbit MFD shows ApT=64.52 and one second later ApT=63.38 (the difference is 1.14).

#### Kolodez

##### New member
Oh, thanks, what are the x- and y-axis in the diagram?

#### Ajaja

##### Active member
x - GMAT’s Modified Julian Date (it's one day - 2021-12-01)
y - AOP (argument of periapsis in degrees)

But I still can't get 295 m/s. Looks like AOP of Phobos is wobbling with speed 0-70 m/s.

#### Ajaja

##### Active member
I've compared the default Phobos ephemeris module with SPICE module in Orbiter.
Yes, the default Phobos module has bigger amplitude and speed of apoapsis fluctuations.
With SPICE and mar097.bsp kernel it's 1.03 s (~63 m/s), not 1.14 s.

#### Kolodez

##### New member
But none of these fluctuations is created by Deimos, is it? So what is then the reason and how can we calculate the same Orbit externally?

#### Kolodez

##### New member
Is it maybe the fact that Mars it not a perfect ball?

#### Ajaja

##### Active member
But none of these fluctuations is created by Deimos, is it?
The bigger fluctuations 295 m/s is an artefact of numercial approximation of the orbit.
I'm mot sure about 63 m/s, but NASA uses this ephemerides (like MAR097) in real missions, so it's definitely closer to the real orbit. But it's hard to say how accurate it is in this case.

Is it maybe the fact that Mars it not a perfect ball?
Yes, apparently this is the reason. Or it's an artefact of numerical approximation too.

According to Doc\Credit.pdf Orbiter uses:
Phobos and Deimos ephemeris modules
Carl Romanik (“Chode”)
Code based on: Sinclair, Astron. Astrophys. 220, 321 (1989) “Testing against Horizons shows agreement within 20km for Phobos, 50km for Deimos for 2000-2024.

how can we calculate the same Orbit externally?

I wasn't able to find source code of the module. So it will not be easy to calculate precisely the same orbit externally.

#### Ajaja

##### Active member
I did other simulation in GMAT not using now MAR097 ephemerides for Phobos but integrating with Mars50c gravity field model.
It matches MAR097 trajectory almost precisely.
This wobbling definety happens because of the non-spherical perturbation of the gravity field. Deimos and Sun add almost nothing here.