Q: What was your thesis about and who was your advisor?
A: My PhD thesis (1989) addressed several aspects of molecular line observations applied to the study of star formation, combining both theoretical and observational work. I began by theoretically analyzing the expected effects of protostellar infall on molecular line profiles as observed with radio telescopes. Although this problem was later extensively studied by several groups using different approaches, our method has proved to be particularly useful for the analysis of images obtained in recent observations with radio interferometers. I also investigated the influence of temperature gradients on the derivation of physical parameters of outflows from molecular line observations.
On the observational side, I studied the properties of molecular outflows in the emblematic HL/XZ Tau region using the 12-m radio telescope at Kitt Peak. In addition, I analyzed the dense gas in a large sample of regions with outflows using the Haystack radio telescope, which at that time played a prominent role in advancing our understanding of star-forming regions through ammonia observations. While my thesis focused on the use of single-dish radio telescopes, my subsequent work has been mainly devoted to observations with large interferometers.
I did my PhD thesis at the University of Barcelona under the supervision of Robert Estalella, with whom I maintained both a close friendship and collaboration until just over a year ago, when he sadly passed away suddenly while still very active. My thesis was the first doctoral dissertation on star formation at that university and was conducted at a time of limited national scientific funding, but I consider it helped lay the groundwork for subsequent generations of researchers in radio astronomy of star formation at that university. Following a visit by Jorge Cantó to Barcelona, an opportunity arose to initiate a collaboration with UNAM, which enabled me to undertake a one-year (1985-1986) predoctoral research stay in Mexico.
That stay proved decisive in shaping my academic career and establishing my long-term interest in star formation. There, Luis Felipe Rodríguez guided my first steps in forefront research, teaching me how to design radio astronomical observations to address key questions in the field, while Jorge shared his ability to distill the physical essence of problems and approach them with simplicity and rigor. I traveled to the United States to carry out observations at Green Bank, Kitt Peak, Haystack, and the VLA, and there I met my collaborators Chema Torrelles and Paul Ho for the first time. Many of the central themes of my doctoral thesis emerged during this intense period, which effectively combined theory and observation and led to a fruitful long-term collaboration.
Q: Back in the 1980s it was not always easy to identify the driving sources of outflows. You showed that ammonia observations were helpful in that regard.
A: At that time, very little was known about the properties of deeply embedded protostars. In fact, Class 0 objects had not yet been formally identified, and the search for genuine protostars was famously compared by Gareth Wynn-Williams to the quest for the Holy Grail. As a result, considerable effort was devoted to locating these very young stellar objects.
Ammonia is a particularly useful molecule in this context. It is a reliable tracer of high-density gas and has several transitions around 1.3 cm that are easily observable, allowing accurate determination of the physical conditions of the gas. In a survey of a large number of outflow regions, I found that the driving sources of the outflows tend to be located toward the peak of the ammonia emission in maps obtained with a large single-dish telescope. This positional accuracy was often sufficient to discriminate among different close by IRAS source candidates. This approach also proved useful for selecting more restricted regions for detailed follow-up studies. However, higher angular resolution observations with the VLA later revealed that, on smaller scales, the situation is usually more complex, with an irregular ammonia emission distribution and multiple objects within the IRAS positional error ellipse.
Q: In the early 1990s you and your team identified a number of sources of Herbig-Haro objects and molecular outflows in the 3.6 cm radio continuum, including a rare double source in the L723 cloud. What were the advantages of this approach?
A: In the late 1980s, the VLA became equipped with new 3.6 cm receivers funded by NASA to support the tracking of the Voyager encounter with Neptune in 1989. These receivers were significantly more sensitive than those previously available at other wavelengths, enabling the detection of radio sources that had remained undetected in earlier studies. One of the first regions where these new observations revealed multiple YSOs associated with a single IRAS point source was L723.
The molecular outflow in L723 displays a peculiar four-lobe morphology centered on the IRAS source, leading to a debate as to whether this structure represented two independent bipolar outflows, each with a pair of lobes, or a single bipolar outflow with strongly limb-brightened shells delineating its two lobes. In VLA observations carried out in January 1990, shortly after the installation of the 3.6 cm receivers, we detected a double radio source within the large IRAS positional error ellipse, strongly favoring the interpretation of two independent outflows driven by two distinct sources. Subsequent, more sensitive VLA observations with subarcsecond resolution — carried out as part of the PhD work of Carlos Carrasco-González — revealed an even more complex picture, with multiple centimeter sources near the outflow center and possibly three independent outflows at different evolutionary stages.
These 3.6 cm observations marked the beginning of the discovery of a population of protostars that had remained hidden in earlier studies. Although the radio emission from these objects is weak, observations at centimeter wavelengths can reach a very high sensitivity, becoming one of the most effective tools for detecting the deeply embedded Class 0 protostars. An important advantage is the ability to achieve subarcsecond angular resolution, revealing features such as protostellar multiplicity on scales of ~100 au, nearly two orders of magnitude finer than was possible with IRAS.
By the early 1990s, it had become quite clear that the centimeter continuum emission associated with outflow-driving sources arises from free-free emission produced by partially (~10%) ionized, jet-shaped sources. However, the origin of the ionization was unclear. Classical HII regions associated with massive young stars are ionized by stellar UV photons, but low-luminosity YSOs emit radio continuum radiation despite producing orders of magnitude fewer ionizing photons than required to explain the emission as a classical HII region.
A few years earlier, Salvador Curiel and collaborators had proposed a model in which a neutral stellar wind becomes ionized through shocks against a dense ambient medium. This model predicted, under simplifying assumptions, a correlation between the outflow momentum rate and the centimeter luminosity. Earlier attempts to detect this correlation had been inconclusive primarily due to the limited sensitivity of existing observations. In 1992, using the new 3.6 cm VLA receivers, we found the first observational evidence for this correlation, supporting the shock-ionization scenario. Subsequent studies with larger samples confirmed this result and also revealed a strong correlation between centimeter luminosity and bolometric luminosity over a wide range of masses, from brown dwarfs to massive protostars.
Q: You have also worked on the detection of HO maser sources.
A: In the 1990s, we conducted an extensive single-dish survey of nearly 200 regions to investigate the relationship between HO masers and high-density gas traced by NH and CS emission. We found several interesting correlations between the physical properties of the regions and those of the maser emission. In particular, the correlation between the NH and CS line widths and the maser luminosity suggests that water maser emission is intrinsically associated with enhanced thermal and turbulent energy in the surrounding cloud. This result is consistent with models in which HO masers originate in shocked regions, such as those produced by YSO outflows.
Because maser emission is non-thermal and exhibits very high brightness temperatures, it can be observed with VLBI at extremely high angular resolution. In a series of projects enthusiastically promoted by Chema Torrelles, we used VLBA observations of water masers to trace substructures such as filaments, bubbles, disks, winds, and jets, as well as their proper motions, down to astronomical-unit scales. This approach allowed us to undertake subsequent studies, such as tracking the onset and development of outflow collimation.
Q: In a highly cited paper from 1998 you and your collaborators determined the spectral indices of centimeter continuum outflow sources. What did you learn?
A: Centimeter continuum emission is a powerful tracer of the base of protostellar jets and, therefore, of the youngest and most deeply embedded YSOs. However, VLA observations cover relatively large fields of view and usually include several extragalactic radio sources that can be mistaken for YSOs. This problem of contamination from background sources is more severe in highly sensitive observations, thus making a discriminating method more necessary. In that paper, we showed that the sources located toward the outflow centers (their likely driving sources) systematically exhibit positive spectral indices, characteristic of thermal emission, whereas background sources typically show negative spectral indices indicative of non-thermal emission.
We also presented an updated formulation for estimating the expected number of background sources in interferometric radio observations and summarized the properties of YSO radio jets as understood at the time. Our goal was to provide practical guidelines and a physical framework for interpreting centimeter continuum observations. Motivated by the development of new, ultra-sensitive, high-resolution facilities such as the SKA and the ngVLA, this work was later updated and expanded in two review papers published in 2015 and 2018 with Luis Felipe and Carlos.
Q: In 2007 you published a study of the proper motions of jets in the HH 30/HL/XZ Tau region, and analyzed the wiggling as effects of binarity of the sources.
A: This is the only paper I have led that is based entirely on optical observations. The study began at the suggestion of Alex Raga, who noticed that the HH 30 and HL/XZ Tau jets, although similar in their knotty appearance to other well-known jets — such as HH 111 or HH 34, which you, Bo, know very well — they lacked a clearly defined ”head” with a terminal bow shock. Using what was then a large-format CCD installed on the ING telescope at La Palma, we carried out observations led by Rosario López, aimed at covering the full extent of the jets. In this work, we succeeded in detecting the likely jet heads and, in combination with observations from other telescopes, obtained preliminary measurements of their proper motions, supporting their association with the jets.
I became particularly intrigued by the pronounced wiggling observed in the long HH 30 jet and later we performed more precise observations using the NOT telescope at La Palma. I initially attempted to fit the observations using an existing model from the literature describing a cylindrical jet launched from an orbiting source, but this approach produced unreliable results. I then realized that the model neglected the centrifugal effect of the orbital motion — an essential ingredient — which naturally leads to a conical rather than a cylindrical jet. Recognizing this limitation, I persuaded Alex to work on a more realistic model. He quickly developed a simple model capable of accounting for either orbital motion or precession, which he published with Elena Masciadri, who was a postdoctoral researcher at UNAM at the time. Using this model as a foundation, we were able to successfully explain the observed wiggling, infer the presence of a binary system, and demonstrate that the HH 30 disk is indeed a circumbinary disk. While our initial results — together with subsequent Plateau de Bure observations by Stéphane Guilloteau and collaborators — favored a binary with a separation of ~15 au as the origin of the wiggling, more recent ALMA observations suggest that the wiggling may instead be caused by precession in a much tighter binary system, leaving the detailed binary configuration still open to debate.
Q: You have also worked on protoplanetary disks. What was your approach to this topic?
A: I have always been interested in studying star formation with the best possible combination of angular resolution and sensitivity, as this often reveals faint and unexpected features that are invisible in lower-quality data. As a natural progression, my research evolved from molecular clouds, high-density cores and large-scale flows to close protobinary systems, radio jets, and ultimately protoplanetary disks. This evolution also reflects a broader trend in the field, driven in part by the advent of ALMA, while the JWST has recently renewed interest in jets.
I have consistently sought to balance observational capabilities with tailored, physically motivated models, making extensive use of large radio interferometers and multi-wavelength data. This philosophy has guided the work of our group at the Institute of Astrophysics of Andalusia in Granada, which includes José F. Gómez (an experienced radio astronomer) and Mayra Osorio (my wife, whose expertise lies in theoretical modeling). More than a decade ago, Mayra spearheaded the incorporation of protoplanetary disk studies into our group by combining our expertise in radio interferometry with the modeling legacy of Paola D’Alessio on accretion disks. I am particularly proud of the results obtained in the study of ”peculiar” disks, many of which formed the basis of the PhD theses of Enrique Macías and Ana K. Diaz-Rodriguez. These include some of the first studies of transitional disks, notably the first imaging at 7 mm and modeling of the now iconic disk around HD 169142 (2014, 2017); the discovery of an extremely compact dust disk around XZ Tau B with a radius of only 3 au, yet potentially capable of forming planets (2016); imaging of the accretion disk around the massive protostar driving HH 80-81 (2018); and a comprehensive centimeter and millimeter study of the disks in the close protobinary system SVS 13 (2022).
Q: What are you and your students currently working on?
A: In our most recent work, which constitutes the core of the PhD thesis of Guillermo Blázquez-Calero and has just been published in Nature Astronomy, we have fully clarified the nature of molecular bullets first discovered by Rafael Bachiller 35 years ago. Previous studies had established that these bullets consist of multiple knots similar to those observed in optical, infrared, and radio jets. Our ALMA observations revealed weak emission at intermediate velocities between the extremely high-velocity (~100 km/s) and standard high-velocity (~10 km/s) gas components. The morphology and kinematics of this emission agree remarkably well with the predictions of a model of a momentum-conserving bow shock produced by a variable-velocity jet, developed by Alex Raga and collaborators three decades ago. Although Alex did not live to see the final stages of this work, Sylvie Cabrit and Gary Fuller contributed greatly to its development and completion.
In contrast to current trends favoring large statistical surveys, I strongly believe in the power of comprehensive, detailed studies that combine high-resolution, high-sensitivity observations with realistic physical modeling. Although models are necessarily simplified representations of reality, capturing the essential physics is crucial. This philosophy is something I actively strive to pass on to my students.