Periodic Reporting for period 3 - QUENCH (Star formation quenching and feedback in galaxies throughout the cosmic epochs)
Período documentado: 2019-10-01 hasta 2021-03-31
Galaxies are broadly divided in star forming and passive. The mechanism responsible for quenching star formation in galaxies and transforming them
into passive systems is still not clear. Understanding this transformational process is one of the main topics of modern astrophysics.
Two main categories of quenching mechanisms have been invoked: 1) removal of gas, primarily through outflows driven by accreting
supermassive black holes (AGN-quasars), which implies removing fuel for star formation, or 2) "starvation", which consists in preventing
the galaxy from accreting fresh supplies of gas, implying that the galaxy cannot form more stars once the gas available in the galaxy
has been used up.
Moreover, either of the two mechanisms can be associated either with the galaxy properties (e.g. its mass, its star formation rate, the
accretion on its supermassive black hole) or with its environment (e.g. the surrounding density of galaxies, whether
it is a central or a satellite galaxy).
The aim of this project is to use multi-band observations to identify what are the primary mechanisms responsible for quenching
star formation in galaxies, as a function of cosmic epoch and for different categories of galaxies, and to compare the results
with the expectations of theoretical models.
Some of the key facilities to be used in this project are the Atacama Large Millimetre Array (ALMA), the James Webb Space Telescope (JWST,
and in particular its main spectrograph NIRSpec), MOONS (which is the next generation optical/infrared multi-object spectrograph at the
Very Large Telescope, VLT), as well as other VLT instruments.
into passive systems is still not clear. Understanding this transformational process is one of the main topics of modern astrophysics.
Two main categories of quenching mechanisms have been invoked: 1) removal of gas, primarily through outflows driven by accreting
supermassive black holes (AGN-quasars), which implies removing fuel for star formation, or 2) "starvation", which consists in preventing
the galaxy from accreting fresh supplies of gas, implying that the galaxy cannot form more stars once the gas available in the galaxy
has been used up.
Moreover, either of the two mechanisms can be associated either with the galaxy properties (e.g. its mass, its star formation rate, the
accretion on its supermassive black hole) or with its environment (e.g. the surrounding density of galaxies, whether
it is a central or a satellite galaxy).
The aim of this project is to use multi-band observations to identify what are the primary mechanisms responsible for quenching
star formation in galaxies, as a function of cosmic epoch and for different categories of galaxies, and to compare the results
with the expectations of theoretical models.
Some of the key facilities to be used in this project are the Atacama Large Millimetre Array (ALMA), the James Webb Space Telescope (JWST,
and in particular its main spectrograph NIRSpec), MOONS (which is the next generation optical/infrared multi-object spectrograph at the
Very Large Telescope, VLT), as well as other VLT instruments.
During the first period the team has exploited multi-band data
to observationally investigate the mechanisms responsible for
quenching star formation in galaxies. This first period has primarily
focused on the use of the ALMA data, since JWST and MOONS data are
not available yet. However, some important constraints have already been
obtained by exploiting current optical/infrared instruments at the VLT as well as other facilities.
The following is a list of the main achievements obtained in the first period:
- By using archival and new ALMA data we have obtained the largest
sample of molecular outflows in the local Universe. By combining with
other new and archival data we have also explored the scaling relations
between molecular, ionized and neutral atomic outflows. We have determined
the scaling relations between outflow properties
and galactic properties such as AGN power, star formation rate and galaxy mass.
We have found that the AGN ejective mode is less effective than expected by
theoretical models. Moreover, only a small fraction of the outflow
reaches the escape velocity. Therefore, the ejective mode of AGN-driven
outflows is unlikely to be effective in quenching star formation in galaxies.
We have also discovered, for the first time, evidence for the presence of
"fossil" outflows, i.e. outflows driven and outlasting a
previous AGN phase, which has now faded; this is a phenomenon predicted
by models but never identified observationally so far.
Finally we have shown that radio jets generally do not play an important role
in driving massive outflows (Fluetsch et al. 2019).
- A similar investigation about cold outflows has been performed for high-z
quasars (much more powerful than the local population), by exploiting new and
archival ALMA data. We have detected evidence of outflows
in the combined spectrum of more than 40 quasars at z~6. However, we infer
that also in this case the quasar-driven outflows are significantly weaker
than expected by the ejective ("blast-wave") models, therefore questioning
the effectiveness of the so-called ejective "quasar mode" in quenching
star formation in galaxies (Bischetti et al. submitted, Carniani et al. 2017b)
- Molecular outflows have been investigated also in star forming galaxies
at high redshift. In particular, through the study of water absorption
in the most extreme starburst galaxy at high redshift we have revealed one
of the most powerful and massive outflows ever detected. However, this starburst
driven outflow seems to follow the same scaling relations as other outflows in less
extreme star forming local galaxies (Jones et al. submitted).
- The effectiveness of feedback in star forming galaxies has also been
investigated in high-z (z~5-7)
galaxies by comparing the morphology observed in the cold
ISM and the distribution of (unabsorbed) star forming regions (traced by the UV light).
We have found that the distribution of cold gas is generally offset with respect
to the UV light. Through comparison with models, this result is partly interpreted
as resulting from dust extinction but also as the effect of star-formation driven
outflows expelling gas out of the galaxy (Carniani et al. 2017b, 2018a, 2018b).
- We have fully exploited the data on stellar metallicities for
hundred thousands local galaxies from the SDSS survey to investigate the
quenching mechanisms. We have found that the large stellar metallicity difference
between passive galaxies and their star forming progenitors can only be explained
as a consequence of an extensive phase of starvation, although some contribution
from outflow or gas heating must have contributed too.
By comparing with models we found that a typical quenching timescale of 2 billion years
can explain the observed metallicity difference.
It is also found that galaxy environment plays a role only in satellite galaxies and only in very dense
(cluster-like) environments (Trussler et al. submitted).
It is suggested that the starvation is resulting from the AGN heating of the halo, which
results in the suppression of gas accretion, hence a delayed quenching from starvation.
- By investigating the fraction of passive galaxies as a function of galaxy parameters (by using machine learning techniques),
we have found that the black hole mass (not necessarily the presence of an active AGN)
is the parameter that is most strongly correlated with galaxy quenching. This is in support
of the scenario in which black hole accretion is responsible for galaxy quenching
through a delayed quenching by heating the galaxy halo (Bluck et al., in prep.)
- We have obtained direct observational evidence for the AGN heating of a massive galaxy
halo by exploiting a very deep ALMA exposure of a powerful quasar at z=1.7 to detect the Sunyaev-Zeldovich (S-Z)
effect associated with the hot halo of the galaxy. We find that the S-Z signal is in agreement
with the expectations of recent cosmological simulations and supporting the delayed quenching scenario.
This is the first time that the halo
heating scenario is directly tested through observations at high redshift (Brownson et al., submitted).
- By using new VLT data, as well as data from the MaNGA-SDSS survey,
we have also obtained, for the very first time, evidence that star formation can occur
inside galactic outflows, as predicted by some recent models. This result has far reaching
implications, as it may imply that this new mode of star formation can contribute
significantly to the formation of the spheroidal component of galaxies (Maiolino et al. 2017, Gallagher et
al. 2019).
- During the first period we have also worked on preparatory activities for the JWST and MOONS observing
programmes. More specifically, by using simulations and through an extensive analysis of the performances
of these two facilities we have identified an optimal observing strategy plan for the Guarantee Observing
Time as well as optimized the selection of targets. A detailed observing plan has been submitted for the JWST GTO programme.
- As part of the ALPINE ALMA Large Programme (which has targeted a sample of 120 galaxies
at z~4-6), we have contributed to the data processing and data analysis. We are using stacking
techniques as well as direct kinematical modelling to infer evidence of outflows in these primeval galaxies.
to observationally investigate the mechanisms responsible for
quenching star formation in galaxies. This first period has primarily
focused on the use of the ALMA data, since JWST and MOONS data are
not available yet. However, some important constraints have already been
obtained by exploiting current optical/infrared instruments at the VLT as well as other facilities.
The following is a list of the main achievements obtained in the first period:
- By using archival and new ALMA data we have obtained the largest
sample of molecular outflows in the local Universe. By combining with
other new and archival data we have also explored the scaling relations
between molecular, ionized and neutral atomic outflows. We have determined
the scaling relations between outflow properties
and galactic properties such as AGN power, star formation rate and galaxy mass.
We have found that the AGN ejective mode is less effective than expected by
theoretical models. Moreover, only a small fraction of the outflow
reaches the escape velocity. Therefore, the ejective mode of AGN-driven
outflows is unlikely to be effective in quenching star formation in galaxies.
We have also discovered, for the first time, evidence for the presence of
"fossil" outflows, i.e. outflows driven and outlasting a
previous AGN phase, which has now faded; this is a phenomenon predicted
by models but never identified observationally so far.
Finally we have shown that radio jets generally do not play an important role
in driving massive outflows (Fluetsch et al. 2019).
- A similar investigation about cold outflows has been performed for high-z
quasars (much more powerful than the local population), by exploiting new and
archival ALMA data. We have detected evidence of outflows
in the combined spectrum of more than 40 quasars at z~6. However, we infer
that also in this case the quasar-driven outflows are significantly weaker
than expected by the ejective ("blast-wave") models, therefore questioning
the effectiveness of the so-called ejective "quasar mode" in quenching
star formation in galaxies (Bischetti et al. submitted, Carniani et al. 2017b)
- Molecular outflows have been investigated also in star forming galaxies
at high redshift. In particular, through the study of water absorption
in the most extreme starburst galaxy at high redshift we have revealed one
of the most powerful and massive outflows ever detected. However, this starburst
driven outflow seems to follow the same scaling relations as other outflows in less
extreme star forming local galaxies (Jones et al. submitted).
- The effectiveness of feedback in star forming galaxies has also been
investigated in high-z (z~5-7)
galaxies by comparing the morphology observed in the cold
ISM and the distribution of (unabsorbed) star forming regions (traced by the UV light).
We have found that the distribution of cold gas is generally offset with respect
to the UV light. Through comparison with models, this result is partly interpreted
as resulting from dust extinction but also as the effect of star-formation driven
outflows expelling gas out of the galaxy (Carniani et al. 2017b, 2018a, 2018b).
- We have fully exploited the data on stellar metallicities for
hundred thousands local galaxies from the SDSS survey to investigate the
quenching mechanisms. We have found that the large stellar metallicity difference
between passive galaxies and their star forming progenitors can only be explained
as a consequence of an extensive phase of starvation, although some contribution
from outflow or gas heating must have contributed too.
By comparing with models we found that a typical quenching timescale of 2 billion years
can explain the observed metallicity difference.
It is also found that galaxy environment plays a role only in satellite galaxies and only in very dense
(cluster-like) environments (Trussler et al. submitted).
It is suggested that the starvation is resulting from the AGN heating of the halo, which
results in the suppression of gas accretion, hence a delayed quenching from starvation.
- By investigating the fraction of passive galaxies as a function of galaxy parameters (by using machine learning techniques),
we have found that the black hole mass (not necessarily the presence of an active AGN)
is the parameter that is most strongly correlated with galaxy quenching. This is in support
of the scenario in which black hole accretion is responsible for galaxy quenching
through a delayed quenching by heating the galaxy halo (Bluck et al., in prep.)
- We have obtained direct observational evidence for the AGN heating of a massive galaxy
halo by exploiting a very deep ALMA exposure of a powerful quasar at z=1.7 to detect the Sunyaev-Zeldovich (S-Z)
effect associated with the hot halo of the galaxy. We find that the S-Z signal is in agreement
with the expectations of recent cosmological simulations and supporting the delayed quenching scenario.
This is the first time that the halo
heating scenario is directly tested through observations at high redshift (Brownson et al., submitted).
- By using new VLT data, as well as data from the MaNGA-SDSS survey,
we have also obtained, for the very first time, evidence that star formation can occur
inside galactic outflows, as predicted by some recent models. This result has far reaching
implications, as it may imply that this new mode of star formation can contribute
significantly to the formation of the spheroidal component of galaxies (Maiolino et al. 2017, Gallagher et
al. 2019).
- During the first period we have also worked on preparatory activities for the JWST and MOONS observing
programmes. More specifically, by using simulations and through an extensive analysis of the performances
of these two facilities we have identified an optimal observing strategy plan for the Guarantee Observing
Time as well as optimized the selection of targets. A detailed observing plan has been submitted for the JWST GTO programme.
- As part of the ALPINE ALMA Large Programme (which has targeted a sample of 120 galaxies
at z~4-6), we have contributed to the data processing and data analysis. We are using stacking
techniques as well as direct kinematical modelling to infer evidence of outflows in these primeval galaxies.
The characterisation of cold outflows in local and distant galaxies, the identification of the main quenching mechanism through
the metallicity difference of passive and star forming progenitors, the discovery of the S-Z signal tracing the hot halo of a distant galaxy,
the detection of star formation inside galactic outflows, as discussed above, are all results that have gone well beyond
the state of the art in this field.
In the rest of the project the focus is expected to gradually shift more towards higher redshifts, primarily by using
new ALMA data on distant galaxies (especially by exploiting the ALPINE Large Programme), new data from the JWST GTO
and new data from the MOONS GTO. These data will enable us to characterize in detail the outflows of distant galaxies
and to constrain the quenching mechanism in the early Universe both by investigating the fraction of quenched galaxies
and by using indirect tracers of the quenching mechanism (such as the gas and stellar metallicities).
the metallicity difference of passive and star forming progenitors, the discovery of the S-Z signal tracing the hot halo of a distant galaxy,
the detection of star formation inside galactic outflows, as discussed above, are all results that have gone well beyond
the state of the art in this field.
In the rest of the project the focus is expected to gradually shift more towards higher redshifts, primarily by using
new ALMA data on distant galaxies (especially by exploiting the ALPINE Large Programme), new data from the JWST GTO
and new data from the MOONS GTO. These data will enable us to characterize in detail the outflows of distant galaxies
and to constrain the quenching mechanism in the early Universe both by investigating the fraction of quenched galaxies
and by using indirect tracers of the quenching mechanism (such as the gas and stellar metallicities).