Rigorous quantification of biases from unmodeled chromatic effects in PTA data

Quantify rigorously, using realistic multi-frequency pulsar timing datasets and models, the biases that unmodeled chromatic effects (such as scattering delays with radio-frequency index χ > 2) introduce into estimates of achromatic red noise and dispersion measure, extending beyond the simplified two-frequency analytical case and determining under what observational conditions these effects arise in real data.

Background

In the appendix, the authors present a simplified analytic demonstration for two radio frequencies showing that unmodeled chromatic processes (e.g., scattering with χ > 2) can bias dispersion measure and induce spurious achromatic delays of opposite sign, potentially affecting red-noise inferences.

They note that this derivation is illustrative and may not capture the full complexity of real, multi-frequency PTA observations. A rigorous quantification under realistic conditions is needed to assess the prevalence and magnitude of such biases and to guide improved noise modeling in PTA analyses.