Back

Nocturnal Navigation via a Time-Compensated Lunar Compass in Bull Ants

Freas, C. A.; Cheng, K.

2025-09-01 ecology
10.1101/2025.08.28.672594 bioRxiv
Show abstract

Diurnal navigators are well known to orient via the sun, compensating for its movement over time by linking solar position and speed with terrestrial reference points (Beetz & el Jundi, 2018; Dingle, 2014; Dyer, 1987; Mouritsen & Frost, 2002; Stalleicken et al., 2005; Wehner & Lanfranconi, 1981; Wehner & Muller, 1993). In contrast to the sun, the moon is not only dimmer, but its presence is also highly variable, both temporally and in its illuminance, making it a more complex cue to predict. Yet nocturnal animals employ a celestial compass to orient at night (i.e. the moon and stars; (Dacke et al., 2003, 2013; Dreyer et al., 2025), but navigation via a time-compensated lunar compass has yet to be demonstrated. Nocturnal bull ants (Myrmecia midas) navigate home overnight under the dimmest of visual conditions, relying in part on a nocturnal path-integration system informed by lunar cues (Freas et al., 2024). Here, we show the first evidence of a time-compensated lunar compass in a homing animal, with path-integrating ants adjusting path headings by predicting the lunar azimuthal position throughout the night. Ants predict the moons future position based upon a generalised internal estimate of the lunar ephemeris. Rather than detailed knowledge of the moons nightly schedule, these ants form a lunar path representation via a speed-step function, reflecting the moons accelerating then decelerating arc. However, these ants also predictably under- or overestimate the shape of this arc based on individual experience, suggesting a level of temporal uncertainty regarding when the moon rapidly shifts from the eastern to western sky. Additionally, updating this lunar compass relies on the presence of a directionally informative skyline, which allows occasional cross-referencing of lunar movement with external fixed compass cues near the horizon.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.