The Moon is not just a satellite; it is a potential gravitational bell. When massive bodies collide, spacetime shudders, and that shudder, gravitational waves, makes everything in their path vibrate. On Earth, such vibrations drown in seismic noise below 10 Hz, and in space, the LISA antenna only hears millihertz. In between lies a quiet decihertz range, where intermediate-mass black hole mergers are thought to sound. The LILA project (Lunar Interferometer for Low-frequency Astrophysics) proposes placing a laser interferometer on the Moon, where the seismic background is orders of magnitude lower than on Earth, and the vacuum eliminates acoustic disturbances. The lunar surface becomes an ideal stage for such an instrument.

The key to sensitivity lies in the Moon's resonant modes. When a gravitational wave passes through the satellite, it excites its natural oscillations, like striking a bell. These modes can amplify the interferometer's response by up to $10^4$ times at low frequencies. The authors calculated the fundamental noises for a suspension-free interferometer: thermal noise in the optics, arising from random atomic vibrations at 300 K, dominates below 0.1 Hz, while laser shot noise, due to quantum uncertainty in photon number, dominates above. For LILA Pioneer with a 5 km baseline, the characteristic sensitivity reaches ~$10^{-21}$ at 1 mHz, comparable to LISA but with an extension into the decihertz range. LILA Horizon with a 40 km baseline and a triangular configuration improves sensitivity by about an order of magnitude.

This approach is fundamentally different: ground-based detectors suspend mirrors to isolate them from seismic noise, while space antennas fly freely. LILA uses the Moon itself as a resonant element, opening the decihertz window. Until now, there were no quantitative sensitivity estimates for a lunar laser interferometer without suspensions, accounting for the Moon's actual resonant modes. Now it is shown that LILA Pioneer achieves a sensitivity of ~$10^{-21}$ in the millihertz–decihertz range, filling the gap between LISA and ground-based detectors. This became possible by combining analytical models of thermal and shot noise with numerical data on lunar normal modes and their quality factors.

The practical benefit of this work lies in informing the design of future lunar missions. If LILA is built, it could detect intermediate-mass black hole mergers, double white dwarfs, and possibly the primordial gravitational-wave background. This would expand our understanding of stellar evolution and the early Universe. An open question remains the actual level of lunar seismic background in the millihertz range: there are no direct measurements, and sensitivity could be limited by unknown local noises. The next step is detailed modeling of thermal noise for specific optical components and developing a stabilized laser for lunar conditions. The Moon, silent for centuries, may sound for those who know how to listen.