Bursting and quenching in massive galaxies without major mergers or AGNs

We simulate the build-up of galaxies by spherical gas accretion through dark matter haloes, subject to the development of virial shocks. We find that a uniform cosmological accretion rate turns into a rapidly varying disc build-up rate. The generic sequence of events (Shocked-Accretion Massive Burst and Shutdown, SAMBA) consists of four distinct phases: (i) continuous cold accretion while the halo is below a threshold mass M-sh similar to 10(12) M-circle dot, (ii) tentative quenching of gas supply for similar to 2 Gyr, starting abruptly once the halo is similar to M-sh and growing a rapidly expanding shock, (iii) a massive burst due to the collapse of similar to 10(11) M-circle dot gas in similar to 0.5 Gyr, when the accumulated heated gas cools and joins new infalling gas and (iv) a long-term shutdown, enhanced by a temporary shock instability in late SAMBAs, those that quench at z similar to 2, burst at z similar to 1 and end up quenched in 10(12-13) M-circle dot haloes today. The quenching and bursting occur at all redshifts in galaxies of baryonic mass similar to 10(11) M-circle dot and involve a substantial fraction of this mass. They arise from rather smooth accretion, or minor mergers, which, unlike major mergers, may leave the disc intact while being built in a rapid pace. The early bursts match observed maximum starbursting discs at z greater than or similar to 2, predicted to reside in less than or similar to 10(13) M circle dot haloes. The late bursts resemble discy luminous infrared galaxies (LIRGs) at z less than or similar to 1. On the other hand, the tentative quenching gives rise to a substantial population of similar to 10(11) M-circle dot galaxies with a strongly suppressed star formation rate at z similar to 1-3. The predicted long-term shutdown leads to red and dead galaxies in groups. A complete shutdown in more massive clusters requires an additional quenching mechanism, as may be provided by clumpy accretion. Alternatively, the SAMBA bursts may trigger the active galactic nucleus (AGN) activity that couples to the hot gas above M-sh and helps the required quenching. The SAMBA phenomenon is predicted based on a spherical model that does not simulate star formation and feedback - it is yet to be investigated using detailed cosmological simulations.