Satellite galaxies are probes of dark matter and cosmology. Their fate after falling into the host halo is determined by tidal stripping, gas loss, and quenching of star formation, and the rate of these processes is set by the minimum distance to the center, the pericenter. In the literature, satellite orbits are almost always computed in a spherical potential because it is simpler: energy and angular momentum fully determine the orbit, and turning points are found analytically. But cosmological simulations show that dark matter halos are typically triaxial, with the minor-to-major axis ratio around 2/3. This work stands apart, with few close neighbors: nearby researchers study dark halos and black holes, but do not systematically compare halo shape and satellite orbits. Previous comparisons in cosmological simulations could not isolate the effect of shape because potential, dynamical friction, and substructure all varied simultaneously.
The authors built nine static halos with the same virial mass of $10^{12}$ solar masses and the same spherically averaged NFW profile, but different triaxiality T and oblateness s. Through each halo, they ran the same million satellite orbits drawn from the cosmological infall distribution of Li et al. (2020). Orbits were integrated for 50 billion years with a step of 1 million years, using the adaptive eighth-order Dormand–Prince method in the galpy package. For each satellite, they recorded successive pericenters, apocenters, and radial periods. Chaos was assessed via Lyapunov exponents: next to each orbit, a second, slightly shifted orbit was launched, and the rate of divergence was measured.
The median distributions of pericenters, apocenters, and periods barely change with halo shape: the shift does not exceed 2% for typical shapes. However, the scatter of successive pericenters for a single orbit is 5–23% depending on shape, and it does not depend on orbital energy. The minimum pericenter over a Hubble time decreases systematically: on average by 5–12%, and in the most flattened halos by up to 21%. In strongly flattened halos, the distribution of minimum pericenters forms a peak near 0.1 virial radius. Chaotic orbits make up to 57% of orbits in the most triaxial halos, but their Lyapunov times are 30–800 billion years, much longer than the Hubble time. Reconstructing the orbit in a spherical potential instead of the true triaxial one gives a median position error of 0.1 virial radius within 2–4 billion years. For the dwarf galaxy Triangulum II, the pericenter uncertainty due to unknown halo shape and orientation reaches ~50%, which is five times larger than the stated error and an order of magnitude larger than the shift caused by the Large Magellanic Cloud.
This work is the first to isolate the effect of halo shape while keeping the mass profile fixed, showing that the spherical approximation does not bias median values but introduces large scatter for individual orbits. Unlike neighboring studies, which either compare populations in full cosmological simulations or study streams and stars, here only the halo shape is systematically varied, and the consequences for individual satellites are traced. This allows separating the contributions of shape, orientation, and chaos.
One could construct a 'dephasing kernel', a compact correction that randomly perturbs the orbital angular momentum at each pericenter passage depending on halo shape. This would give semi-analytic models of subhalo evolution almost all the benefits of triaxial modeling at minimal computational cost. It is also important to include halo shape uncertainty in standard procedures for reconstructing the orbits of observed satellites.
The results will affect semi-analytic models of subhalo evolution (SatGen, Galacticus), estimates of tidal stripping, and the interpretation of Milky Way satellite orbits from Gaia data.
Test the predictions against cosmological simulations with triaxial halos and develop a practical scheme for accounting for halo shape in orbital reconstructions.
For galaxy dynamics, an open question remains: how much does the dark halo shape and its evolution over time affect the survival and observed properties of satellites? This work shows that ignoring shape leads to a systematic underestimation of minimum pericenters and, consequently, to errors in the rates of tidal disruption and star formation quenching.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The authors built nine static triaxial halos with the same mass profile and ran a million satellite orbits through them, integrating for 50 billion years. They showed that the median orbital parameters barely change, but the successive pericenters of each orbit scatter by 5–23%, the minimum distance to the center systematically decreases by up to 21%, and the error in spherical reconstruction reaches 0.1 virial radius within 2–4 billion years. For Triangulum II, the pericenter uncertainty due to halo shape is five times larger than the reported error.
Where its strength is
A systematic comparison of halo shape effects on orbits, with a fixed spherical profile, isolates the contribution of triaxiality, which previously got lost among other processes. This provides a controlled basis for estimating errors in orbit reconstruction for real satellites.
Where it could go next
Worth trying
Why this may matter
If halo shape systematically shifts pericenters, it directly affects estimates of satellite stripping and quenching rates, and thus their use as probes of dark matter and cosmology. Accounting for triaxiality could change conclusions about dark matter distribution in dwarf galaxies.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.57, against a usual 0.64. It looks like few have drilled nearby.
In the most flattened halos, the distribution of minimum pericenters forms a peak near 0.1 virial radius, although no such peak exists in the spherical case. This peak arises from chaotic orbits that dive deeper toward the center.