Measuring and predicting galaxy assembly bias across galaxy samples

Abstract: One of the most important effects shaping small-scale galaxy clustering is galaxy assembly bias, which refers to the dependence of galaxy clustering on halo properties. We investigate this effect using galaxy samples selected according to stellar mass, r-band magnitude, and broad-band colors from the largest hydrodynamical simulation of the IllustrisTNG suite. We find that galaxy assembly bias depends strongly upon the selection criteria, number density, and redshift of the sample, increasing or decreasing the clustering by as much as 25%. Interestingly, no single secondary halo property fully captures the strength of this effect for any galaxy population. Therefore, empirical approaches modeling galaxy assembly bias as a function of a single halo property cannot reproduce predictions from hydrodynamical simulations. We then study how galaxy assembly bias emerges from the interplay of halo assembly bias -- the dependence of halo clustering on properties other than mass -- and occupancy variation -- the correlation between galaxy occupation and secondary halo properties -- and provide an analytical expression that predicts the amount of galaxy assembly bias caused by any halo property. This expression facilitates understanding the dependence of galaxy assembly bias on halo properties and enables the straightforward incorporation of this effect into halo model approaches.