Paper: On the physical association of Fermi-LAT blazars with their low-energy counterparts

Associating γ-ray sources to their low-energy counterparts is one of the major challenges of modern γ-ray astronomy. In the context of the Fourth Fermi Large Area Telescope Source Catalog (4FGL), the associations rely mainly on parameters as apparent magnitude, integrated flux, and angular separation between the γ-ray source and its low-energy candidate counterpart. In this work we propose a new use of likelihood ratio (LR) and a complementary random forest (RF) technique to associate γ-ray blazars in 4FGL, based only on spectral parameters as γ-ray photon index, mid-infrared colors and radio-loudness.

The LR method

We modified the LR method used in 4FGL to estimate the association probability ρ of a counterpart for a given γ-ray  source. For each 4FGL source, we considered the γ-ray photon index interval ΔΓ= [Γ -σ, Γ + σ], where σ is the uncertainty on the photon index as listed in 4FGL. Then, given the known correlations between  Γ, the mid-infrared (MIR) colors c, and the MIR radio-loudness q for γ-ray blazars, we selected only those sources with parameters lying within the ΔΓ intervals, as in the figure below:


For each MIR counterpart i of a γ-ray blazar source j we computed the LR and then the association probability ρ as a function of the LR, as shown in the figure below.


The RF method

The RF is an ensemble classifier that uses decision trees as building blocks for classification. For classifying a new object, each tree in the forest chooses one class and, by aggregating the predictions of all decision trees, the RF makes a final prediction based on the choice made by the majority of the trees, thus improving the predictive capability and reducing the tendency of standard decision trees to overfit the training sample.

We trained the RF method with basically the same atributes used in the LR approach obtaining a modest final accuracy of ~80%. As this accuracy is not very high, we opted by using the RF approach as a complementary method to the LR.


Compared with 4FGL, the association probabilities found by our methods tend to be higher than those listed in 4FGL (see figure below).

LR_comparison_zero.pngIn this work we were able to associate ~1000 blazars with both LR and RF methods, where ~300 of them are good targets for future optical spectroscopic campaigns. This was also the first work where spectral properties of blazars were used to associate γ-ray sources to their low-energy counterparts. Previous methods rely basically on parameters like apparent magnitude and angular separation between the γ-ray source center and the position of its candidate counterpart.

The original discussion of this work can be found in de Menezes et al. (2020).

This work was supported by FAPESP (Fundação de Ampara à Pesquisa do Estado de São Paulo) under grants 2016/25484-9, 2018/24801-6 (R. de Menezes) and 2017/01461-2 (R. Nemmen); and by many other institutions.



Paper: “Jet efficiencies and black hole spins in jetted quasars”

Black holes are fundamentally simple objects: only their mass and spin are enough to properly describe them. However, direct measurements of these two properties are not simple. Often, we look for observables which, upon being inserted in a physical model, may give us information about one of these fundamental properties of black holes. One such observable is luminosity. Our paper “Jet efficiencies and black hole spins in jetted quasars”, recently accepted by MNRAS (see preprint here), aims to relate jet properties with black hole spins.

With a sample of 154 flat-spectrum radio quasars (FSRQs), a subclass of blazars, whose masses had been previously estimated, we set out to find their gamma-ray luminosities in the Fermi 4FGL catalog, which comprises 8 years of observations performed with the Fermi Large Area Telescope (Fermi-LAT). Our first result is a correlation between jet luminosity and black hole mass, suggesting that more luminous jets are powered by more massive black holes.


We also estimated the jet power of these blazars. For this, we used a relation found in Nemmen et al. 2012. This, along with the assumption that these black holes are accreting at around 10 per cent of the Eddington rate — such high accretion rates are necessary for the thin discs believed to feed FSRQs — allowed us to estimate the jet efficiencies in these objects.


Having estimated the jet efficiencies, we used a simulation-based model to estimate the black hole spins. We found that, overall, these black holes are rotating very fast: the average spin was 0.84. This result is consistent with scenarios for the cosmological evolution of SMBHs which support rapidly rotating black holes as their host galaxies — and the black holes themselves — merge.

The preprint of the accepted version, with a full discussion, can be found here.

Artur defends his MsC dissertation

The black hole group has a new master: Artur Vemado defended his MsC dissertation, entitled “radiative cooling and state transitions in stellar mass black holes”. The defense was very successful.

Here, Artur reported his numerical simulations of black hole accretion flows where he incorporated radiative cooling (with some approximations otherwise the problem is essentially intractable!). We observe the self-consistent emergence of a hot corona enveloping a cold thin accretion disk. Artur quantified the inner radius of the thin disk, the size of the corona, and how these properties respond to varying the mass accretion rate onto the black hole. The resulting simulated black holes are similar to observations of stellar mass black holes in binary systems.

We are looking forward to reporting these exciting results on the emergence of the corona (not the covid-19!) and truncated disk in an upcoming publication.

Many thanks to FAPESP funding through grant 2017/25710-1.

Dr. Gustavo R. R. Soares, PhD in Astrophysics

Congratulations to the now Dr. Gustavo R. R. Soares, for a successful PhD thesis defense! 🎉🍾

The thesis is entitled “Accretion discs, jets, and black hole spins: a study of blazars” and was done under my supervision. The whole defense was entirely online, following the social distancing recommendations of the World Health Organization and the São Paulo State government, in order to ensure the safety of all involved with respect to COVID-19.

The defense lasted for almost five hours (!!), with the thesis committee members in two countries—Brasil and US—and in three states in Brasil: Pernambuco, Rio de Janeiro and São Paulo.

Gustavo presenting his thesis work

Thanks to Dr. Soares’s work, we now we know a bit more about the role of black holes in the universe, and how the supermassive ones power relativistic jets.

Gustavo will begin a postdoc at Oregon State University in the Fall. We are all wishing Gustavo a huge success for his future career!

Our thanks to the Brazilian science funding agencies CAPES, CNPq and FAPESP. Without them, this work would not have been possible.

Group at IAU GALFEED Symposium

On the week of March 2nd to 6th, the Black Hole Group attended the IAU Symposium 359: Galaxy Evolution and Feedback Across Different Environments, or simply the GALFEED symposium. The event occurred at the beautiful setting of Bento Gonçalves in the south of Brasil. This is the wine region of the country, which specializes in the merlot variety.

This was an incredible event which led to many fruitful interactions between the group members and the galaxy evolution community. Beginning on March 2nd, Fabio and Raniere gave poster flash talks where they had the daunting challenge of summarizing their posters in only one minute. Not so easy, but they did a great job!

On Tuesday, it was Ivan and Gustavo’s turn to face the one-minute-present-all-your-research challenge. And again, it went fantastic!

On Wednesday, we surprised Fabio with a surprise birthday party: it was his 40th birthday! There was cake and presents!

Fabio’s birthday. From left to right: Raniere, Gustavo, Fabio, Ivan, Rodrigo and Roderik

On Thursday, it was Rodrigo’s turn. He surprised the audience by beginning hist talk about spin estimates for M87*…

Rodrigo begins his talk on the spin of M87*. Certainly not the best of pictures…

… and then switching gears to talk about the first AI simulation of a black hole—the work of graduate student Roberta Pereira.

The first quarter of 2020 was very bright and productive for the group, despite the worrying news with COVID-19 all over the world.

Thanks to the IAU and FAPESP for funding our group’s attendance to this meeting.

A nonthermal “bomb” producing a strong flare in Sgr A*

Our paper titled “A nonthermal bomb explains the near-infrared superflare of Sgr A*” has been accepted for publication in The Astrophysical Journal Letters, and it is freely available in arXiv. The work was developed by PhD student Eduardo Gutiérrez, from the GARRA group at the Argentine Institute of Radio Astronomy, and Prof. Rodrigo Nemmen and PhD student Fabio Cafardo from the Black Hole Group at USP. In the Letter, we propose a physical mechanism to explain an unprecedented powerful flare that took place in Sgr A* last year.

On 13 May 2019, the supermassive black hole at our Galactic center experienced the strongest near-infrared flare detected so far. The observations made with the Keck Telescope led Prof. Tuan Do, from the University of California, Los Ángeles, and collaborators to publish a Letter reporting the results. The following animation is a superpositions of the images taken by the Keck telescope, and shows the strong variable emission detected that night:

Interestingly, the flare was catched when the brightness was already diminishing, which suggests that its peak luminosity might have been even higher!

Sgr A* is known to experience regular flares in several bands of the spectrum, extending from radio up to X-rays; however, the NIR flare on May 2019 was much brighter than any other ever before. In our work, we propose a possible physical mechanism which might have been the responsible for the flaring emission. We frame the event under the term nonthermal bomb. But what do we mean by this term?

Screenshot from 2020-01-27 15-10-32

The accretion flow onto our galactic center is extremely thin; very little amount of matter feeds the black hole. These low-density flows are usually called Radiatively Inefficient Accretion Flows (RIAFs) because most of the gravitational energy released by the matter is swallowed by the hole and not radiated. Given the very low densities, the plasma in these flows is extremely collisionless and particles have difficulties to exchange energy between themselves efficiently. As a consequence, they may not reach thermal equilibrium, and a fraction of them can be nonthermal. In fact, it is thought that a small population of nonthermal electrons steadily present in the accretion flow is the responsible for the quiescent radio emission of Sgr A*. To represent the ambient conditions in the flow we have followed the standard modelling of a RIAF around Sgr A* in the steady state (see Yuan et al. 2003). The following image shows the Spectral Energy  Distribution (SED) predicted by this model, which is in very good agreement with the multiwavelength data. The different colors in the plot represent the different emission processes that take place in the flow.

Over the background ambient responsible for the quiescent emission, an additional transient process must occur in order to produce a flare like the one we are dealing with. Do et al. (2019) suggested that a large increase in the accretion rate (for example, the accretion of a denser blob of matter) might be what caused the enhanced emission. On the contrary, we propose a different mechanism: that a huge amount of magnetic energy was released in the accretion flow in a bursting event and was able to accelerate an additional amount of electrons to relativistic energies. Those we do not state which particular acceleration mechanism is working, the most plausible culprits  are magnetic reconnection and turbulence acceleration. Since we dealt with a time-dependent process we took into account the evolution of this population of accelerated relativistic particles as they cool by synchrotron emission and are advected towards the hole. Interestingly, given the length-scales of the accretion flow in Sgr A*, both of these processes, namely cooling by synchrotron and advection, are of the same order that the detected flare. In the following figure we show the flare data and the fitting we obtain with our model.

Screenshot from 2020-02-26 12-46-27

Despite some degeneration in the parameters, our model is able to explain the flare emission and fit the data very accurately. We also make predictions in X-rays and mm wavelengths (Event Horizon Telescope band) that might help to test our model against others. We expect that this works motivate further investigation on particle acceleration and bursting events in RIAFs, and in particular in Sgr A*.

Link to the article in arXiv

This work was supported by FAPESP (Fundação de Amparo à Pesquisa do Estado de São Paulo) under grant 2017/01461-2.

Paper: Gamma-ray observations of low-luminosity active galactic nuclei

The majority of the activity around nearby (z ~ 0) supermassive black holes (SMBHs) is found in low-luminosity active galactic nuclei (LLAGN), the most of them being classified as low ionization nuclear emission regions (LINERs). Although these sources are well studied from radio up to X-rays, they are poorly understood in gamma-rays. In this work we take advantage of the all sky-surveying capabilities of the Large Area Telescope on board Fermi Gamma ray Space Telescope to study the whole Palomar sample of LLAGN in gamma-rays.

Gamma-rays from the accretion flow

The observational properties of LLAGN favor a scenario for their central engines which is quite different from that of more luminous AGN: since the SMBHs are accreting at low rates they are in the radiatively inefficient accretion flow (RIAF) mode (Narayan & Yi 1994) rather than radiatively efficient geometrically thin accretion disks (Shakura & Sunyaev 1973).  RIAFs store most of the viscous energy and advect it into the SMBH. The viscous heating affects mainly the ions, while the radiation is produced primarily by the electrons.

The presence of both RIAFs and jets in LLAGN produces a rich multiwavelength spectral energy distribution (SED) in which a gamma-ray component should be expected. There are multiple possible origins for the gamma-ray emission. In the vicinity of the event horizon, the ion temperature of a RIAF can reach 10¹² K (Yuan & Narayan 2014), thus enabling proton-proton collisions and the production of neutral pions, which subsequently decay into pairs of GeV photons (Mahadevan et al. 1997). Furthermore, synchrotron self-Compton (SSC) emission is expected from the jet. Quantifying and modeling the gamma-ray emission from LLAGN is the main goal of this work.


Precisely, the four radio-brightest LLAGN in the sample are identified as significant gamma-ray emitters, all of which are recognized as powerful Fanaroff-Riley I galaxies. These results suggest that the presence of powerful radio jets is of substantial importance for observing a significant gamma-ray counterpart even if these jets are misaligned with respect to the line of sight (see figure below, left panel). We also find that most of the X-ray-brightest LLAGN do not have a significant gamma-ray and strong radio emission, suggesting that the X-rays come mainly from the accretion flow in these cases (see figure below, right panel).

Lgamma_Lradio_noerrorbar  Gamma_Xrays_noerrorbar

When comparing different emission models to the SEDs of these gamma-ray-bright LLAGN, we find that they are well described by a jet-dominated model in the form of a one-zone synchrotron-SSC jet scenario, similar to what is expected when observing blazars (see figure below, left panel). We also find that models invoking the origin of the electromagnetic radiation in a RIAF fail to explain the observed gamma-ray emission but are able to reproduce the radio-to-X-ray emission (see figure below, right panel). Since the gamma-ray-bright LLAGN consist of powerful radio galaxies, it is not entirely unexpected that their emission can be dominated by a jet even if misaligned to the observer’s line of sight.


The original discussion of this work can be found in de Menezes et al. (2020).

This work was supported by FAPESP (Fundação de Ampara à Pesquisa do Estado de São Paulo) under grants 2016/25484-9, 2018/24801-6 (R. de Menezes), 2017/01461-2 (R. Nemmen) and 2019/10054-7 (I. Almeida); and many other institutions.





Attending the biggest event of AI and Machine Learning in Latin America: Khipu

From November 10th to November 15th, Khipu happened in Montevideo, Uruguay. Khipu was the best biggest event of artificial intelligence and machine learning in Latin America. The speakers of this event are all AI leaders in their fields and big companies such as Google, Deepmind, Apple and Facebook were sponsoring the event. The competition to be in one of the 200 participants of this event was highly competitive. It was one in a lifetime experience that I will try to sum up here. 

On the very first day, they welcomed us with a small lecture about motivation in AI fields and with a toast event. In this toast event, I met Nando de Freitas (Principal Scientist at Deepmind and of the main speakers there) and Cho Kyunghyun (Research at Facebook AI and NYU Professor), we talked about my project, about the event and about Brazil! They were really nice and friendly to me and my friend.

All of the lectures were interesting but I will give special attention to some of them: Ian Goodfellow talking about GANs, Yoshua Bengio talking about Perspectives in AI, Nando de Freitas talking about Reinforcement Learning, Cho Kyunghyun talking about Recurrent Neural Networks and Jeff Dean closing the event with an amazing talk about the challenges in AI. All the lectures are available here:

Another important lecture was “How To Write a Good Paper”. So important and very helpful ideas!

Another highlight in this event was Nando de Freitas talking about my project with me when he visited my poster and commenting about my project in his lecture. It was a really special moment for sure. He even posted in Twitter and retweet my tweet talking about the project. I received a supportive comment from Cho Kyunghyun as well.

It is impossible to talk about Khipu without talking about Jeff Dean (GoogleAI leader and one of Tensorflow creators). I talk with him for a few minutes in a coffee break, he is also very friendly and we talked about my project and about GoogleAI opportunities. In his awesome lecture about challenges in AI, he also comments about my project and I found that one of the biggest highlights in the week. I took a picture with him and posted on Twitter, he commented.

I even met some amazing people from DeepMind and GoogleAI there, they gave me a lot of advice and talked about the life in DeepMind and in GoogleAI. I even met Feryal Behbahani from DeepMind and Women in ML and Oriol Vinyals from DeepMind (main AlphaStar researcher)

This event was so important with so many big names in AI. I am very thankful for Khipu because this event because I have so many opportunities now and also big thank you because they paid all the expenses with travel and accommodation. I really look forward to Khipu 2020! It is a really life changing experience.  

Credits of pictures: @Khipu_AI

Paper: Black hole growth in high redshift radio-loud quasars studied with the ESO Very Large Telescope

Our paper, VLT/SINFONI study of black hole growth in high redshift radio-loud quasars from the CARLA survey”has been accepted for publication in the Monthly Notices of the Royal Astronomical Society, and the pre-print appeared on astro-ph today. The paper was led by former graduate student Murilo Marinello, and this work formed a major part of his PhD dissertation published in April this year.

The new study focused on 35 distant, radio-loud quasars, the majority of which were selected from the Clusters Around Radio-Loud AGN (CARLA) survey. These quasars were known to have large black hole masses, emit luminous radio emission, and tend to be found in dense regions of the early universe. Therefore, they are believed to be good candidates for the distant progenitors of massive (elliptical) galaxies that dominate the universe today. The masses of their supermassive black holes had previously been estimated using the virial black hole mass method applied to their SDSS spectra. Due to their high redshifts, however, the only emission line available to make these measurements in these optical SDSS spectra was the CIV line. This line is known to be affected by non-gravitational effects (winds or outflows) and is thus not optimal for the virial black hole mass estimate. In this project, we therefore re-observed the quasars in the near-infrared using the SINFONI spectrograph on the Very Large Telescope in Chile. This allowed us to access the redshifted Ha broad emission line, and thus determine the black hole masses more accurately. This makes a big difference, as can be seen in the figure below showing the nice symmetric Ha line on the top and the distorted CIV line at the bottom for one of our quasars:


Together with a determination of the accretion rates of the quasars, which can be estimated from their luminosities, the new black hole masses were used to also derive the growth histories of these supermassive black holes. One major finding was that if these quasars had always been accreting at the same rates as measured at the current time, it would not have been possible for them to obtain their high observed masses within the cosmic time available since the Big Bang. The logical conclusion is thus that these quasars must have experienced a phase of much faster growth in the past. This can be nicely illustrated in the following figure:


The red points are the CARLA quasars from this study. The black solid lines show the growth tracks we found to be the ones describing their most likely histories. These tracks consist of two phases: a rapid growth phase starting from a one thousand solar mass black hole seed at z ~ 20, growing at the Eddington limit to a hundred million or more solar masses at z ~ 6, followed by a second, slower phase at the observed lower Eddington ratios until z ~ 2-3. As such, it is possible that the CARLA quasars are direct descendants of the luminous quasars found at z ~ 6-7.

In the local universe, there is a strong correlation between the masses of the supermassive black holes and the masses of their host galaxies. Since the more massive galaxies are also found, on average, in more massive dark matter halos, there is an indirect connection between the mass of the black hole and the mass of its halo. We therefore also tested whether the black holes in the CARLA quasars already “know” that they are located in dense galaxy environments:


We found a weak, low significance correlation between the black hole masses and the surface density of galaxies that surround them (the latter is a measure of the environment or halo mass of the CARLA quasars), and therefore do not find strong evidence that the most massive CARLA quasars are also in the most dense environments or massive halos. However, these galaxy surface densities had been previously determined with the Spitzer Space Telescope as part of the CARLA project, and are not very precise. In the future, we will therefore focus on trying to obtain more precise measurements of the environments of the CARLA quasars, and test again for possible correlations between black hole mass and environment.


Massive presence of the group at the Brazilian Astronomical Society Meeting

The Black Hole Group had a massive presence at the 43th meeting of the Brazilian Astronomical Society—Sociedade Astronômica Brasileira, SAB.

After Marta Volonteri’s review talk on the cosmic evolution of massive black holes, Fabio gave a presentation about the current status of his analysis of Fermi LAT observations of the Galactic Center—we are finishing the first paper of the series which should be submitted very soon.

After Fabio’s talk, we’ve had Ivan’s talk on his numerical hydrodynamical simulations of radiatively inefficient accretion flows and their winds.

Roberta Pereira presented her poster on applying deep learning to predict the future of accreting black holes, which are an extreme example of spatiotemporally chaotic systems.

Finally, Gustavo won one of the best poster prizes at the meeting, and was awarded a talk at the meeting. Wait, is that an award? 🙂

Gustavo Soares’s talk at the 43th SAB meeting.