Contact Information:

Hai Fu

Associate Professor
Department of Physics & Astronomy
University of Iowa
Iowa City, IA 52242

Office: 751 Van Allen Hall
Phone: +1 (319) 335-0402
Email: hai-fu at uiowa dot edu

I am an associate professor at the Department of Physics & Astronomy, University of Iowa. I obtained my B.S. in Astronomy from Nanjing University in 2003 and my Ph.D. in Astronomy in 2008 from the University of Hawai'i. Afterwards, I continued doing research as a postdoc fellow at California Institute of Technology and University of California, Irvine. I joined the faculty of the University of Iowa in 2013.

Galaxies like our Milky Way are the building blocks of the Universe. After decades of research, we now know that, over the past 13.7 billion years, massive galaxies like our Milky Way have grown from tiny overdensities in an essentially homogeneous universe to large ensembles of stars, gas, and dark matter. But it's a complex process, because galaxies do not evolve in isolation and they exhibit large diversities. To make progress in this quest, astronomers divide the galaxy population into various categories at different epoches and study each category in great detail, hoping that eventually we will piece together a coherent story. Previously, I studied quasar extended emission-line regions (EELRs), the coevolution of black holes and galaxies, double-peaked emission-line active galactic nuclei, and the brightest dusty starforming galaxies. I have observed these facinating objects with integral-field spectroscopy, adaptive optics, radio interferometers, and space-based telescopes. Currently, my research interests focus on understanding the effects of galaxy mergers with SDSS-IV/MaNGA integral-field spectroscopy, probing dark matter in high-redshift star-forming galaxies with gas kinematics from ALMA, and tracing the large-scale gas supply of high-redshift starburst galaxies with quasar absorption-line spectroscopy.

I am fortunate to work with a group of students who are fascinated by galaxies. And we are grateful to NSF and NASA for their generous support to our research.

Publications: All Referred, Major Contribution, Google Scholar.

.: Recent News

  • July 2019, the ADS to Bibdesk (API Edition) is brought to you by UIowa Postdoc Rui Xue. As the ADS Classic is now deprecated, this new ADS to Bibdesk python package uses ADS modern's API to update and maintain a user's local BibDesk database. Read this blog to learn about user-developed tools for ADS modern.
  • July 2019, X-ray Properties of Radio-Selected Dual Active Galactic Nuclei: Gross, Fu, Myers, Wrobel, Djorgovski, 2019 ApJ, in press
  • Dec 2018, The Gas Fraction Evolution of Star-Forming Galaxies Traced by The CO Tully Fisher Relation: Isbell, Xue, & Fu, 2018 ApJL 869 37
  • Aug 2018, Flat Rotation Curves Found in Merging Dusty Starbursts at z=2.3 through Tilted-ring Modeling: Xue, Fu, et al. 2018 ApJL 864 11
  • Aug 2018, SDSS-IV MaNGA: Galaxy Pair Fraction and Correlated Active Galactic Nuclei: Fu et al. 2018 ApJ 856 93
  • Jan 2018, Gravbox -- the first AR sandbox for astronomers. The NSF-funded sandbox is brought to you by nine UIowa undergraduates (AST-1614326). Isbell et al. 2018 AAS Meeting #231, id. 316.06
  • .: Main Research Areas

    Dusty Starburst Galaxies

    Fu et al. 2012, 2013, 2016 2017

    We live in a rapidly evolving Universe: Ten billion years ago, galaxies were forming stars ten times more fiercely than they do today. Thanks to the Herschel Space Telescope, for the first time we have identified a sample of gravitationally lensed massive starbursts at this peak epoch of cosmic star formation. Our high-resolution multi-phase observations in combination with gravitational lensing have helped us gain a comprehensive understanding of these unusual galaxies. We found that these starburst galaxies are ten times more efficient in forming stars than normal starforming galaxies. Such a high efficiency can choke future star formation by rapidly exhausting cold molecular gas, which is the fuel for star formation. Therefore, it provides an effective mechanism to turn a massive star-forming galaxy into a quiescent galaxy in the early universe. However, it remains unclear how these galaxies maintain such a high star formation efficiency. Solving this question relies on future high-resolution observations.

    Our observations on an extremely rare merger of two massive dusty starburst galaxies was published in Nature and was picked up in the News: LA Times, NBC News,,

    Assembly of Super Massive Black Holes

    Fu et al. 2011a, 2011b, 2012, 2015a, 2015b

    During the collision of two galaxies, gas from the galaxy outskirts are driven to the centers of both galaxies because of strong gravitational torques. The gas flow can feed the central black holes in the merging galaxies. Like their host galaxies, the black holes will eventually merge and form a larger black hole. By observing this kind of objects we can learn more about the complicated physical processed in galaxy mergers and how the merger frequency evolves over cosmic time.

    To catch such black hole binaries before they merge, I use the Keck 10-meter telescope to take sharp images of accreting black holes and look for companions in their close vicinity. The figure on the left illustrates an example.

    Here is an interesting article about this work on

    Co-Growth of Galaxies and Black Holes

    Fu et al. 2010a

    A central question in extragalactic research is the mass assembly history of galaxies and their black holes. Understanding this process requires measurements of the rates of star formation and black hole accretion throughout the cosmic ages. A major obstacle in this field is to separate star formation and SMBH accretion.

    With NASA's Spitzer space telescope, we obtained mid-IR spectra of luminous galaxies at redshift 0.7, when the Universe was only half of its current age. The spectra allow us to separate the emission from star formation and black hole accretion and, subsequently, to study the co-growth of galaxies and their black holes.

    We found that intense black hole accretion accompanies about a quarter of the galaxies when they are rapidly forming stars, and the growth in black hole mass is ~0.1% of the mass of newly formed stars. These observations show that the correlation between black hole mass and galaxy mass is maintained in major episodes of star formation.

    Black Hole Feedback & Extended Nebulae

    PhD thesis; Fu & Stockton 2006, 2007a, 2007b, 2008, 2009

    Virtually every galaxy contains a massive black hole at its heart. Although neglibible in mass and size when compared to the host galaxies, these black holes appear to be able to control the growth of the galaxies, because the mass of the galaxies is tightly related to that of the black hole in nearby galaxies including our own Milky Way. To explain this observation, theorists have hypothesized a feedback mechanism from quasars. The feedback can expel most of the gaseous material to large distances impulsively, thus regulating both black hole growth and star formation in the galaxy. But did this actually happen?

    To catch black hole feedback in action, we used the Gemini 8-meter telescope and the Hubble Space Telescope to study the fascinating filamentary nebulae around quasars. Our observations favor a scenario where the driving force producing the nebulae is a galactic-scale quasar superwind, in the form of a roughly spherical blast wave. Since such a mechanism is capable of ejecting a mass comparable to that of the total gaseous medium in our Milky Way, quasar extended nebulae provide local examples of the purported black hole feedback that may have regulated star formation and black hole growth in the early universe.

    This result was featured in

    Near-Earth Asteroid Families

    Fu et al. 2005a

    Asteroids occassionally run into one another and create families of smaller asteroids with similar orbits. Although many asteroid families are known in the main belt, no such families have been identified among the hazardous near-Earth asteroids. These families would create spikes in the collision probability with our Earth (imagine being hit by a few asteroids at the same time!).

    With numerical simulations I found that these hazardous families are only identifiable within ~300,000 years of formation, as they would quickly become well-mixed with background objects because of the disturbance from the planets. Random alignments of objects in the orbital space occur much more frequently than true families. So I developed a novel technique to distinguish between generic asteroid families and random orbital alignments. This technique can potentially discover the very first near-Earth asteroid families, as more and more near-Earth asteroids are identified by surveys like Pan-STARRS.