Extending the frequency reach of pulsar timing array based gravitational wave search without high cadence observations

Gravitational wave (GW) searches using pulsar timing arrays (PTAs) are assumed to be limited by the typical average observational cadence of $1/(2~{\rm weeks})$ for a single pulsar to GW frequencies $\lesssim 4\times 10^{-7}$ Hz. We show that this assumption is incorrect and that a PTA can detect signals with much higher frequencies, which are preserved in the data due to aliasing, by exploiting asynchronous observations from multiple pulsars. This allows an observation strategy that is scalable to future large-scale PTAs containing $O(10^3)$ pulsars, enabled by the Five-hundred meter Aperture Spherical Telescope and the Square Kilometer Array, without requiring a higher per-pulsar observation cadence. We show that higher frequency GW observations, reaching up to $4\times 10^{-4}$ Hz with an SKA-era PTA, have significant astrophysical implications, such as (i) a three orders of magnitude better constraint than current high-cadence observations on GW strain in the $[10,400]$ $\mu{\rm Hz}$ band, and (ii) sensitive tests of the no-hair theorem in the mass range of supermassive black hole binaries using their inspiral, merger, and ringdown signals.