Could redshift drift effects be isolated if you compared post-merger oscillation ratios across multiple observed merger masses?

Yes, your intuition is sound. Post-merger oscillation ratios—particularly frequency ratios and quality factors of the post-merger signal—can indeed isolate redshift drift because they are dimensionless parameters that break the mass-redshift degeneracy [AG-2025.05-280].

Here's why this works:

The degeneracy problem: Gravitational waves redshift as they travel through the expanding universe, so what we observe is the product $\mathcal{M}(1+z)$—the true mass scaled by redshift. A massive distant source looks identical to a less massive nearby one [AG-2025.05-280]. This makes it hard to disentangle true source mass from distance.

Why dimensionless ratios help: The post-merger signal from neutron star mergers contains oscillations in the hypermassive or supramassive neutron star. The frequencies and decay times of these modes depend on the *true* mass and the neutron star equation of state. Crucially, frequency ratios (e.g., the ratio of the f-mode frequency to the fundamental quadrupole mode) are dimensionless and scale with mass in predictable ways that redshift does not simply erase. Since redshift multiplies all frequencies equally, comparing *ratios* across multiple observed mergers at different redshifts lets you infer the true mass spectrum and subtract out the redshift effect [AG-2025.05-280].

Practical sensitivity: Using only the post-merger signal (observable by next-generation detectors like the Einstein Telescope), fractional redshift uncertainties of ~30% are achievable for sources at $0.01 < z < 0.09$ [AG-2025.05-280].

A complementary note: Binary neutron stars are also advantageous cosmologically because their mass distribution evolves very little with redshift, unlike black hole mergers [AG-2024.11-058]—so you avoid biases from assuming a constant mass spectrum when it actually changes over cosmic time.