A summary of my research interests
Quasars are extremely luminous types of Active Galactic Nuclei (AGN). The majority of the quasar population have very blue optical colours, typically due to an unobscured view of the accreting supermassive black hole. However, there is an important fraction that are much redder, called "red quasars". In previous work we have found fundamental differences in the radio properties of red quasars compared to blue (Klindt et al. 2019; Fawcett et al. 2020; Rosario et al. 2020) that rules out a simple orientation model alone, favoring the evolutionary scenario (see figure below). My PhD focused on exploring the radio and spectroscopic properties of red quasars. My current work expands the parameter space into more obscured systems using both DESI and, in the future, VLT/MOONS.
A potential evolutionary sequence, in which a red quasar represents an obscured phase with compact radio jets and powerful winds. Eventually, the gas and dust is blown away, revealing an unobscured typical blue quasar. Credit: L. Kindt and S. Munro.
Analysing an SDSS-selected sample of red and blue quasars, observed by X-shooter, we find that dust is the most likely cause of the red colours for the majority of the red quasars, with moderate extinctions ranging from Av~0.06-0.7 mag (Fawcett et al. 2022).
Fitting a simple thin accretion disc, we do not find any significant differences in the accretion properties of red quasars; this suggests the differences we find in the radio properties are due to other AGN mechanisms, such as outflows. Calistro-Rivera et al. (2021) find evidence for larger FWHM in the [OIII] component of red quasars, indicative of outflows, which supports this model.
Gif: A movie demonstrating our basic extinction curve fitting approach. The cQSO composite is displayed by the cyan line and the rQSO spectra are displayed in red. The cQSO composite with increasing amounts of dust extinction is displayed in grey. This shows that the red colours are consistent with dust. Both the composite and animation are available through GitHub .
Composites of DESI quasars in bins of dust reddening, from the bluest bin (blue) to the reddest bin (yellow). The red and blue quasar X-shooter composites from Fawcett+22 (F22) are displayed by the blue and black dot-dashed lines, respectively; DESI can push to quasars with more extinction than previously observed with SDSS.
DESI will provide spectroscopic data for 35 million galaxies and 2.4 million quasars over a 5 year survey, concluding in ~2026. I am currently a continuing participant of DESI and a Diversity Equality and Inclusion (DEI) committee member. I was previously on the Early Career Scientist (ECS) committee.
Following a successful proposal, I have a secondary target program (implemented from February 2021), to observe dust reddened quasars that would otherwise be missed by the nominal DESI QSO selection. We find that red quasars observed by DESI are more highly extinguished than the SDSS red quasars used in our previous study (see above figure). This will test whether the fundamental differences in the radio properties of red quasars extend to more extreme systems.
Comparison of wavelength coverage of the exisiting and upcoming spectroscopic surveys, demonstrating the unique capabilities of MOONS to push into the near-infrared. Combined with DESI spectra, we can obtain X-shooter like spectra for thousands of AGN.
Find more details here.
MOONS is the next generation spectrograph for the Very Large Telescope (VLT). It will have a wavelength coverage of 0.65-1.8 microns, pushing much further in the IR than both current and upcoming multi-object spectrographs (e.g., DESI, 4MOST). One of the Guaranteed Time Observation (GTO) surveys is the MOONS Redshift-Intensive Survey Experiment (MOONRISE) which will target COSMOS and two VIDEO fields. MOONRISE is expected to observe a few thousand Type 1 AGN and a few tens thousands Type 2 AGN, measuring [OIII] outflows out to z~2.6.
I am part of the technical working group, testing spectral fitting pipelines, and the AGN science working group, exploring target selection and preparing scientific projects for eventual data collection.
In Fawcett et al. (2020), I used high resolution, deep radio data of a sample of red and blue quasars to explore the fundamental differences in the radio properties of red quasars. Building on Klindt et al. (2019), we use deeper radio data and find an enhancement in the radio detection rate of red quasars compared to blue quasars down to micro-Janksy levels.
Exploring radio morphologies with high resolution VLA data (Hodge et al. 2012), we find that red quasars show an enhancement of compact radio emission (<43 kpc; see figure). Utilizing e-MERLIN data of 20 red and 20 blue quasars we find that red quasars show an enhancement in ~kpc scale radio emission (Rosario et al. 2021).
(Left) comparison of radio images for different resolution data, (top) morphology fractions using VLA Stripe 82 data.
Graph showing the radio enhancement in red quasars that peaks around the radio-quiet threshold, but falls away at both the extreme radio-quiet and radio-loud end (Fawcett et al. 2021). Using the wide multi-wavelength data available in the COSMOS field, we split the sources by whether their radio emission is AGN or SF-dominated (green and purple stars, respectively); we find the drop in enhancement is due to SF diluting the radio emission
(Fawcett et al. 2020).
Moving to lower frequency LOFAR data, we still find this enhancement in red quasars and confirm the decrease towards the radio-quiet end, with more source statistics
(Rosario et al. 2020).
Using multi-wavelength radio data, we find that the radio detection enhancement of red quasars is consistent across 0.144-3 GHz frequencies, peaking around the radio-quiet/radio-intermediate regime (see figure). We also find that star-formation dilutes the radio emission at the radio-quiet end, causing the decrease in enhancement.
In Fawcett et al. (2023) we explored the radio and dust extinction properties of ~35,000 DESI quasars. We find a striking positive relationship between the LOFAR radio detection fraction and the amount of dust extinction in a quasar. This demonstrates an intrinsic link between the amount of dust in a quasar and the production of radio emission, suggesting the radio emission is due to winds and/or jets causing shocks in a dusty environment. Therefore, red quasars are likely in a dusty "blow-out" phase before transitioning to a blue quasar.
In future work utilising an extended DESI sample and MOONS infrared spectra, we will populate the highest Av bin to explore whether the radio detection fraction continues to rise or plateaus with increasing dust extinction.
(Top) the radio detection fraction versus dust extinction for ~35,000 DESI quasars. (Bottom) composites of DESI spectra in bins of increasing dust extinction.
Radio detection fraction versus line-of-sight dust extinction for DESI QSOs at z<0.9 (purple). The sources with high velocity (w80>1000 km/s) outflows and low/no outflows (w80<600 km/s) are displayed as the light and dark blue stars, respectively. We find the radio-dust connection is stronger in the powerful outflow sample and is not significant in the weak/no outflow sample.
In Fawcett et al. (2026) we explore this result further, analysing the [OIII] ionised outflow properties of the same DESI QSO sample, restricted to redshifts z<0.9.
We find that the radio-dust connection is dominated by sources with high velocity ionised outflows, and found no connection for sources with weak/no outflows.
However, we find no direct connection between the amount of line-of-sight dust extinction and outflow velocity, which suggests dust extinction alone does not determine whether a quasar hosts a powerful outflow. On the other hand, there is a connection between outflow velocity and optical-MIR colour, consistent with what is found for the Extremely Red Quasar (ERQ) population.
Finally, we find a connection between MIR excess (traced by the 6 micron to 5100 Angstrom ratio) and outflow velocity.
These results suggest that powerful outflows intereact and shock the surrounding dust/gas, heating the dust and producing synchrotron emission.
These shocks would either destroy or blow-out the surrounding dust/gas, eventually reducing the obscuration level, consistent with dusty quasars residing in a blow-out phase in the evolution of galaxies.
Visit to the uGMRT in Pune
We have uGMRT data of 19 red and 19 blue quasars, also observed with e-MERLIN in Rosario et al. (2021).
In combination with LoTSS, FIRST, and VLASS radio data, we have constructed sensitive radio SEDs in order to understand the origin of the radio emission (Fawcett et al. 2025).
Our results suggest that 1) a higher fraction of red quasars have steep radio spectral slopes, 2) there is a tentative connection between amount of dust and the steepness of the spectral slope, and 3) blue quasars are more likely to have extended low frequency radio emission.
This work was later confirmed and extended in Sargent et al. (2026).
One example of a radio SED (left) and radio imaging (right) for one of our red quasars
Galspec conference 2021:
'An X-shooter insight into dust-reddened quasars'
Hypatia Colloquium 2022:
'Why is colour special? Fundamental differences between red and blue quasars'
CosKASI seminar 2022:
'Red and Extremely Red Quasars in DESI: strong link between dust and radio emission'
Mexborough & Swinton Astronomical Society 2022:
'Red quasars: the missing link in galaxy evolution?'