On 20 Oct 2021 UT, I observed the star TYC 724-273-1 in the constellation Orion being covered up by the asteroid 444 Gyptis. The star disappeared at 5:31:53.856 UT and reappeared at 5:32:10.506, a duration of 16.65 seconds.

The published apparent visual magnitude of this star is 11.5 and the published apparent visual magnitude of 444 Gyptis at the time of the event is 12.5.

The combined magnitude (m_{c}) of star + asteroid just before (and after) the occultation event is given by

where m_{o} is the magnitude of the asteroid

and m_{*} is the magnitude of the star

This gives us a combined magnitude of 11.14 just before the occultation.

While the asteroid is covering up the star, you should only see the asteroid, so the magnitude should decrease from 11.14 to 12.5, a magnitude drop of 1.36 magnitudes.

Much to my surprise, I observed a magnitude drop of only 0.54.

Is it possible that 444 Gyptis only covered up one component of a previously undiscovered double star? That idea is bolstered by the fact that the event occurred 14.8 seconds earlier than predicted, a full 3.7σ early.

Entertaining the double-star idea, our task is to determine the magnitudes of the two blended stars and which one got covered up. Let us call the magnitudes of the two components m_{*1} and m_{*2}, with m_{*1} being the component that got covered. We already know that m_{*1} + m_{*2} must equal m_{*} = 11.5. We also know that the observed magnitude drop of the m_{*1} plus the unobserved magnitude drop that the m_{*2} star would have had must equal the expected magnitude drop of 1.36. This gives us enough information to calculate m_{*1} and m_{*2} individually.

where m_{o} is the magnitude of the asteroid

and m_{*} is the magnitude of the star

and m_{c} is the magnitude of the star + asteroid

and m_{*1} is the magnitude of the occulted star component

and m_{*2} is the magnitude of the unocculted star component

and Δm_{obs} is the observed magnitude drop

This gives us a magnitude of 12.36 for the occulted component and 12.15 for the unocculted component. Thus we can see that I observed the fainter component of the double star being occulted by asteroid 444 Gyptis.

Finally, we can do an extra check to make sure that the magnitudes of the two star components plus the asteroid equals the combined magnitude of 11.14 we expected right before the occultation occurred.

Here’s a little SAS program I wrote to do the calculations.

```
data magdrop;
format mstar mastr mcomb pdelm odelm mstr1 mstr2 mtot 5.2;
mstar = 11.5;
mastr = 12.5;
odelm = 0.54;
x = 0.4*(mastr - mstar);
mcomb = mastr - 2.5*log10(10**x + 1);
pdelm = mastr - mcomb;
mstr1 = log10(10**((mcomb+odelm)/-2.5) - 10**(-0.4*mastr))/-0.4;
mstr2 = log10(10**(mstar/-2.5) - 10**(-0.4*mstr1))/-0.4;
mtot = -2.5*log10(10**(-0.4*mstr1)+10**(-0.4*mstr2)+10**(-0.4*mastr));
file print;
put 'Published Magnitude of Occulted Star = ' mstar;
put 'Magnitude of Asteroid = ' mastr;
put 'Combined Magnitude Right Before Occultation = ' mcomb;
put 'Predicted Magnitude Drop = ' pdelm;
put 'Observed Magnitude Drop = ' odelm;
if (odelm/pdelm > 0.5 and mstr1 > mstr2) or
(odelm/pdelm < 0.5 and mstr1 < mstr2) then do;
put 'Magnitude of Star Component Occulted = ' mstr2;
put 'Magnitude of Star Component Not Occulted = ' mstr1;
end;
else do;
put 'Magnitude of Star Component Occulted = ' mstr1;
put 'Magnitude of Star Component Not Occulted = ' mstr2;
end;
put 'Total Magnitude of Both Star Components + Asteroid = ' mtot;
run;
Published Magnitude of Occulted Star = 11.50
Magnitude of Asteroid = 12.50
Combined Magnitude Right Before Occultation = 11.14
Predicted Magnitude Drop = 1.36
Observed Magnitude Drop = 0.54
Magnitude of Star Component Occulted = 12.36
Magnitude of Star Component Not Occulted = 12.15
Total Magnitude of Both Star Components + Asteroid = 11.14
```

That’s something I’ve always wanted to try. Just never in the exact place, time, and clear skies!

It’s an exciting time to be recording stellar occultations by minor planets. Gaia EDR3 is now giving us very accurate positions for both stars and asteroids (and thus improved orbital elements). This in turn provides us with much more accurate predictions of asteroid shadow paths on the Earth’s surface. That means a much greater probability of success (i.e. fewer misses).

Camera sensitivity and quantum efficiency keeps improving, thus allowing us to record shorter events (my limit right now is about 0.2 second).

Even one occultation chord across a small asteroid gives us a positional accuracy that is one to two orders of magnitude better than even Gaia can provide. Only a spacecraft flyby gives us better astrometry.

Interesting factino: the astrometric positions derived for minor planets during stellar occultation events is now becoming so precise that the gravitational deflection of starlight and sunlight reflected off the asteroid has to be taken into account. The gravitational deflection of a star’s light as it travels through the entire solar system is different than the gravitational deflection of sunlight reflecting off the asteroid and traveling the much shorter distance to Earth, even though they are in exactly the same direction at the time of the occultation.