Stellar eclipses

One event we have not considered in the previous section is the possibility that in the course of the inner binary orbit, the two stars align with the view axis from the planet, and so there is an eclipse happening.

We're more familiar with a solar eclipse where a non-radiating object (the Moon) obscures a sun - but a star can equally well obscure another star. Dependent on the luminosities of the two stars, this can dim the total amount of radiation the planet receives quite a lot.

The most spectacular case occurs when a large bit dim star (a red giant for example) obscures a smaller but much more luminous star - in this case, the eclipse lasts long and reduces the overall luminosity quite dramatically.

Note that in a planar solver, the frequency of eclipses is rather large. For a non-planar problem, the orbital inclination usually implies that the eclipses have to occur close to ascending and descending node, and even a few degrees of inclination can so drastically reduce the number of occurrances. Keep this in mind when studying the effect of stellar eclipses - even in a more realistic moderately non-planar case in which nearly everything else would come out the same, the number of eclipses would be much reduced.

Observing an eclipse

Eclipses are simulated when the eclipse finder is switched on like this:

eclipse_finder true

The finder is only active when the evolve keyword is used, the orbit mode does not compute eclipses.

When an event is found, output like this is produced:

Eclipse of Janus-A by Janus-B in second 0
Eclipse ends in second 24200
Eclipse of Janus-B by Janus-A on day 17.5729
Eclipse ends on day 18.1134

so that the user can opt to specifically look at the temperature profile at the beginning of an eclipsis or at its end. The reduction of luminosity according to the degree of obscuration during the eclipse is done automatically when the finder is running. Note that the eclipse might be a short-duration event dependent on the radii of the stars and their distance - to properly resolve it might require to set a low evolution timestep.

Simulating 18 days to the eclipse situation and switching the eclipse finder on and off yields the following plot:

Temperatures for an eclipse of Janus-B by Janus-A compared with the situation without.

The partial obscuration of the hot and bright Janus-B by the cooler Janus-A which lasts for a bit more than half a day reduces daylight temperatures by nearly 20 K towards the end of the eclipse.

While the larger Janus-B obscures Janus-A completely and for a longer time, the effect of this situation (on day 35) is significantly less spectacular and only 2-3 K of daylight temperature.

Continue with Helliconia.

Back to main index     Back to science     Back to worldbuilder

Created by Thorsten Renk 2021 - see the disclaimer, privacy statement and contact information.