Twenty hours or so after the crack opened, the force pulling at the crack would have rotated and be pulling in a north-south direction. At this point, the crack would be traveling west in response to the changing stress.

Twenty hours after that, the force would be pulling in toward the southwest and northeast. The crack would have responded accordingly, and be moving southeast. A slowly-propagating crack would thus be curved by the changing direction of the tidal stress.

For about half the moon's orbit, the stress might not be sufficient to cause shearing, so the crack would stop.

The next day, though, when force built up again, the crack would likely continue from its end point. Europa would have returned to the spot in its orbit where the forces are acting in the same direction as they were the previous day when the crack began. The crack would resume, but be traveling northeast. Thus, a second arcuate segment would begin. This way, chains of repeating segments could form over several days.

"By varying the parameters a little bit, changing the speeds, or changing the initial starting conditions, or stopping conditions in the ice -- that's how you can make either larger arcuate segments or smaller ones," Hoppa explained. The segments vary in length from about 47 miles (75 kilometers) to more than 125 miles (200 kilometers) long.

This model "does work remarkably well in explaining the shapes of these segments," said Gregory Hoppa, a post-doctoral researcher at the University of Arizona's Lunar and planetary Laboratory. Hoppa developed the model for simulating the effects of tidal forces on Europa's frozen surface. "We are able to reproduce some of the actual features that we see on the surface remarkably well," he said.