1: Through stacking of far infrared and radio data of a large number of quiescent galaxy candidates as a function of both their mass and redshift, we have shown that genuinely quiescent galaxies exist all the way back to z=3, when the Universe was just 2 Gyr old. We also showed that quiescent galaxies have higher radio emission than expected from their low star formation rates, suggesting that they harbour radio loud active galactic nuclei that may be involved in quenching them.
This is an important result because it is inherently difficult to confirm the quiescent nature of galaxies from their UV-NIR emission alone, due to the dust/age degeneracy.
In a second study, published in Nature Astronomy, we show that the average dust mass and gas mass in high redshift quiescent galaxies is surprisingly high, given that they are not forming stars.
2: We have played in a key role in a large ALMA study of galaxies from z=0-3, probing their dust and gas. This study for the first time derived the accretion rates of gas onto galaxies from observations, and showed that these are very large. This further strengthens the mystery of early quenched galaxies. Not only is it strange that they finish their starformation so early in the history of the universe, but also that they remained quenched, despite the large amount of gas (fuel of star formation) that is pouring onto them from the cosmic filaments.
3. We have performed a high resolution study of the stellar populations and dusty starformation regions in a sample of the highest redshift starburst galaxies known. We find that the properties of these starburst galaxies are consistent with being progenitors of the highest redshift quiescent galaxies, once they have completed their nuclear starbursts and merged with minor companions.
4: A major result was derived from a detailed study of a gravitationally lensed quiescent galaxy at z=2. The lensing makes it possible for the first time to resolve its inner structure and stellar populations, both photometrically and spectroscopically. The galaxy turns out to be a rotationally supported disk galaxy, rather than a the dispersion dominated proto-bulge that was expected from theory. This has major consequences for how it formed. A major merger would not lead to a compact disk. We were able to measure gradients in its stellar populations, showing for the first time that the galaxy quenched inside out over a timescale of 300 Myr. We also found evidence for outflows from a central AGN which could explain the quenching.
This study is published in Nature.
5. We have studied a complete sample of spectroscopically confirmed ultra massive quiescent galaxies at z=2. This allowed us to link to their direct dependents in the local Universe, and show for the first time directly that the grow by by a factor of two in mass and four in size through minor mergong.
6. We spectroscopically confirmed a sample of the most distant known quiescent galaxies at z=4 and in some cases even constrained their kinematics, showing that already 1.5 Gyr after the big bang, fully evolved quiescent galaxies exist. We also showed that quiescent galaxies at z=4 are even more compact than similar mass galaxies at z=2. Compared to z=0, they are up to 10 times smaller. We showed that their progenitors are dusty starforming galaxies at z>5, and that state of the art cosmological simulations are currently not able to account for massive quiescent galaxies this early.
7. We have lead the creation of the new COSMOS2020 photometric catalog, which is currently under review.