Why do hydrodynamic errors dominate over spacetime errors in neutron star waveforms?

# Hydrodynamic Errors Dominate in Neutron Star Waveforms

The short answer: when two neutron stars collide, the fluid dynamics inside each star is harder to simulate accurately than the spacetime curvature around them—and the waveforms are more sensitive to errors in fluid behavior [AG-2025.09-564].

Here's why this happens in practice.

The Two Sources of Error

When numerical relativity codes simulate merging neutron stars, errors creep in from two directions:

1. Spacetime errors: mistakes in solving Einstein's equations for the gravitational field geometry.

2. Hydrodynamic errors: mistakes in evolving the neutron star fluid (density, pressure, velocity) using relativistic fluid equations.

You might expect these to matter equally. They don't. [AG-2025.09-564] compared two independent codes (FIL and SpEC) that use completely different numerical methods and discretization schemes—yet despite these differences, "the error in the waveforms is primarily dominated by the hydrodynamic evolution." The spacetime solver turns out to be the easier problem.

Why Hydrodynamics Is Harder

Neutron star mergers involve extreme fluid dynamics: shock waves, turbulence, and matter compressed to nuclear density. These are notoriously difficult to resolve numerically. [AG-2024.09-336] identified a concrete example: artificial heating at the neutron star surface during merger simulations, where numerical shock treatment incorrectly elevates temperatures and distorts the tidal deformability—a key observable for constraining the equation of state.

Spacetime evolution, by contrast, is a smooth geometric problem (at least away from singularities). Modern spectral codes handle it with high accuracy.

The Consequence for Gravitational Wave Measurements

This matters because future detectors like Einstein Telescope will try to infer the nuclear equation of state—the relationship between density and pressure in neutron star cores—by comparing observed waveforms to predictions. But if hydrodynamic errors dominate [AG-2025.09-564], then waveform models must be calibrated against expensive high-resolution neutron star merger simulations, not just spacetime codes. This drives up computational cost and sets a floor on achievable accuracy.

The provided preprints do directly address your question.