Clustering redshifts with the 21cm-galaxy cross-bispectrum

The cross-correlation between 21-cm intensity mapping (IM) experiments and photometric surveys of galaxies (or any other cosmological tracer with a broad radial kernel) is severely degraded by the loss of long-wavelength radial modes due to Galactic foreground contamination. Higher-order correlators are able to restore some of these modes due to the non-linear coupling between them and the local small-scale clustering induced by gravitational collapse. We explore the possibility of recovering information from the bispectrum between a photometric galaxy sample and an IM experiment, in the context of the clustering-redshifts technique. We demonstrate that the bispectrum is able to calibrate the redshift distribution of the photometric sample to the required accuracy of future experiments such as the Rubin Observatory, using future single-dish and interferometric 21-cm observations, in situations where the two-point function is not able to do so due to foreground contamination. We also show how this calibration is affected by the photometric redshift width sigma(z,0) and maximum scale k(max). We find that it is important to reach scales k greater than or similar to 0.3 h Mpc(-1), with the constraints saturating at around k similar to 1 h Mpc(-1) for next-generation experiments.