1 Director’s Message 22 Making Good Use of Bad Weather: Jennifer Lotz Finding Metal-poor Stars Through 4 The First Repeating Fast Radio Burst the Clouds in a Spiral Galaxy Vinicius Placco Benito Marcote, Kenzie Nimmo, and 27 Science Highlights Shriharsh Tendulkar John Blakeslee 9 NGC 2071-IR: A Who-dunnit 45 The Legend of Zorro Begins Mystery Ricardo Salinas and Steve B. Howell Tom Geballe 48 A Galactic Dance 12 Speckle Imaging Takes Gemini to Gemini Press Release Its Diffraction Limit 50 On the Horizon Rachel Matson and Andy Stephens Gemini staff contributions 15 Exoplanets Can’t Hide Their Secrets 61 News for Users from Innovative New Instrument Gemini staff contributions Gemini Press Release 69 Gemini Outreach Programs Sparkle 18 Neptune’s Moon Triton Fosters in Both Hemispheres Rare Icy Union Manuel Paredes and Alyssa Grace Jennifer Hanley 73 Papa ‘Ōlelo Hawai‘i Kilohōkū ON THE COVER: Alison Peck Gemini North Multi- Object Spectrograph GeminiFocus January 2020 image of NGC 5394/5, otherwise known as the and 2019 Year in Review Heron Galaxy. This four- GeminiFocus is a quarterly publication color composite captures of the Gemini Observatory an intimate moment in 670 N. A‘ohoku Place an elegant dance by two Hilo, Hawai‘i 96720, USA interacting galaxies some Phone: (808) 974-2500 / Fax: (808) 974-2589 160 million light years Online viewing address: distant. To read more http://www.gemini.edu/node/27 about this compelling Editor: Peter Michaud interacting pair, turn to page 48. Associate Editor: Stephen James O’Meara Designer: Eve Furchgott/Blue Heron Multimedia Credit: Gemini Observatory/NSF’s Any opinions, findings, and conclusions or National Optical-Infrared recommendations expressed in this publication are those of the author(s) and do not necessarily Astronomy Research reflect the views of the National Science Laboratory/AURA Foundation or the Gemini Partnership. ii GeminiFocus January 2020 / 2019 Year in Review Jennifer Lotz

Director’s Message A New Decade for Gemini Observatory Begins

Happy New Year to everyone in the Gemini Observatory community! The past year has en- compassed a number of “firsts” and milestones for me, personally, as Gemini Director: I host- ed my first Gemini Observatory Open House at the 2019 winter American Astronomical So- ciety meeting; visited Korea and Korea Astronomy and Space Science Institute (KASI) for the first time (and got some very important lessons on how to use sujeo, the super-skinny metal Korean chopsticks); met with Argentinian astronomers for the first time in their country at Reunión annual de la Asociación Argentina de Astronomía and at the Universidad Nacional de La Plata; worked on the basics of Chilean Spanish (but still have a long way to go); got a crash course on the nuances of Hawaiian politics and history; and, last but not least, kicked- off the October launch of the National Science Foundation’s (NSF’s) National Optical-Infrared Astronomy Research Laboratory. The best parts of the year were my interactions with Gemini’s global community, and learn- ing about the fantastic scientific discoveries led by our users: observations from the Gemini Near-InfraRed Spectrometer (GNIRS) pinned down the mass of the supermassive black hole of a gravitationally-lensed quasar at the edge of the Universe (Fan et al., 2019); ultra- sharp near-infrared images from Gemini’s multi-conjugate adaptive optics (MCAO) imager GeMS/GSAOI uncovered the age of one of the oldest star clusters in our Galaxy (Kerber et al., 2019); the visiting high-resolution spectrograph IGRINS discovered an extremely rare molecular composition of carbon monoxide and nitrogen in the ices of Triton, Neptune’s largest moon (Tegler et al., 2019); the Gemini Planet Imager Exoplanet Survey (GPIES) of over 500 stars concluded its five-year run and revealed very different pathways for the formation of Jupiter-like planets and the smallest brown-dwarf stars (Nielsen et al., 2019);

January 2020 / 2019 Year in Review GeminiFocus 1 ultra-high-resolution speckle imaging with Texas at Austin) will use the GNIRS to search visiting ‘Alopeke at Gemini North traced the for and characterize the expected reversal orbit of a Jupiter-sized exoplanet in a close of the 20-year long-term downtrend of the binary star system and conclusively dem- temperature of Uranus’ thermosphere. Let- onstrated, for the first time, which star the ters of Intent for the 2020 LLPs are due Febru- planet orbits (Steve B. Howell et al., 2019); ary 4th; these include new opportunities to and over the past few months, Gemini North use the multi-object spectroscopy mode on and South have joined the chase of our first FLAMINGOS-2 and to apply for Subaru Inten- known interstellar comet, 2I/Borisov (Guzik sive Programs as an extension of our Subaru et al., 2019). Telescope time exchange program.

Gemini Observatory had its most scientifi- Gemini Observatory’s staff and collaborators cally productive year ever in 2019! We closed have also achieved significant milestones in out the year with a record number of Gemini development, operations, and user support publications — over 250, a sharp increase over the past year that we expect to pave the from the previous year. Some of this rise in way for Gemini’s science in the next decade. publications can be attributed to the increas- We released the first phase of DRAGONS ingly popular and productive Fast-Turn- (Data Reduction for Astronomy from Gemini around proposal program, with over 10% of Observatory North and South) to support 2019 publications and an average oversub- all of the Gemini facility instrument’s imag- scription rate of ~ 2.2. We have also seen in- ing modes with a modern, Python-based creasing demand for Gemini’s Director’s Dis- software package. The Gemini South MCAO cretionary Time, accounting for an average GeMS upgraded natural guide star sensor is of 12% of the refereed papers over the past performing well, and will enable more effi- several years, compared to a nominal 5% of cient observations over three times the pre- the allocated time. vious available sky area.

The Large and Long Program (LLP), started in A number of ongoing facility and visiting in- 2014 to support more ambitious and longer- strument development projects made signif- term projects, also had a banner year, with icant progress: the Gemini High-resolution the largest number of LLP publications. This Optical SpecTrograph (GHOST) is undergo- year we started three new LLPs: ZF2K: The First ing final testing at National Research Coun- Exploration of the K-Band Window and a Com- cil Canada’s Herzberg Astronomy and Astro- plete Census of Massive Galaxies at 4 < z < 6, physics before shipping to Gemini South; the led by Casey Papovich at Texas A&M Universi- new visiting high-resolution spectrograph ty, will obtain medium-band K imaging over MAROON-X (Principal Investigator (PI) Jacob 0.5 square degrees to detect 4 < z < 6 and Bean) is in commissioning at Gemini North; higher-redshift emission-line objects; Obser- SCORPIO, the facility 8-channel imager/ vational Characterization of Recurrently Active spectrograph, passed its Critical Design Re- Main-Belt Comets and Near-Earth Main-Belt view; and a state-of-the-art MCAO system Comet Candidates, led by Henry Hsieh (Plan- at Gemini North, integral field unit upgrades etary Science Institute), will characterize the for GNIRS, and the visiting Gemini InfraRed activity and nuclei of a number of known Multi-Object Spectrograph (PI Suresh Siva- main-belt comets (MBCs) and near-Earth nandam), all held successful Conceptual MBC (NEMBC) candidates; and Monitoring Design Reviews. Finally, the GPI instrument Seasonal Reversal in Uranus' Upper Atmo- team has secured independent funding from sphere, led by Laurence Trafton (University of Heising-Simons Foundation (PI Quinn Ko-

2 GeminiFocus January 2020 / 2019 Year in Review nopacky, University of California San Diego) in the Astro2020 white paper will signifi- and the NSF (PI Jeffrey Chilcote, University of cantly increase Gemini’s photon-collecting Notre Dame) to upgrade GPI and move it to power by the end of the decade, enabling Gemini North. unknown discoveries to come. In these early days of 2020, I was happy to see What the Future Holds so many in the US community at what was my second Gemini Observatory Open House The next year — and the next decade — are during the AAS winter meeting in Honolulu, shaping up to be transformative for Gemini Hawai‘i. Looking ahead, one of the highlights Observatory and astronomy as a whole. We of 2020 will undoubtedly be the next Gemini cannot yet know how new discoveries and Science Meeting: “20th Anniversary and Be- facilities will disrupt the way we do and think yond,” in Seoul, Korea, from June 21-25, 2019. about astronomy. Therefore, Gemini Obser- Registration is now open, and I can’t wait to vatory’s strengths of flexibility, diversity, and see you all there. agility will continue to serve us well as we prepare for the decade of discovery to come. Although the unrest in Chile and protests at Maunakea have provided challenges for our Over the next several years, we will enhance staff and to doing science over the past year, I our ability to provide efficient and rapid am grateful for the privilege to be part of our observations through the development of journey of discovery about the Universe and updated user interfaces and proposal tools, for everyone in the Gemini community that automated dynamic scheduling, and the makes that journey possible. Clear skies and spectroscopic DRAGONS pipelines. We will happy new year! deliver the first MCAO system to Maunakea by the middle of the next decade, with night- Jennifer Lotz is the Gemini Observatory Director. ly, queue-ready operations. The pathway to She can be reached at: [email protected] full ground-layer adaptive optics described Gemini Focus in Transition: A New Era in User Communications

With the publication of this issue of GeminiFocus, a new era in communications with our in- ternational user community begins. Beginning in Quarter 2 of 2020, a joint publication of all of NSF's NOIRLab facilities will launch that will encompass all of the news that users expect in GeminiFocus, plus similar news from Kitt Peak National Observatory, Cerro Tololo Inter-Amer- ican Observatory, the Community Science & Data Center and the Vera C. Rubin Observatory. Gemini users throughout the international Gemini Partnership will continue to receive the in- formation needed to carry out the cutting-edge science we’ve come to expect from our com- munity and, with the additional content, to expand everyone’s horizons. The Gemini e-newscast will for the time being continue to provide the latest time-critical infor- mation in email format for users, such as proposal deadlines, instrument availability and other important events. We welcome your input as we embark on this transition to better serve the entire Gemini Partnership.

January 2020 / 2019 Year in Review GeminiFocus 3 January 2020 Benito Marcote, Kenzie Nimmo, and Shriharsh Tendulkar

The First Repeating Fast Radio Burst in a Spiral Galaxy

Observations with the European VLBI Network and the Gemini North telescope have localized for the second time in history a Fast Radio Burst (FRB) source that repeats. Known as FRB 180916.J0158+65, it originates from a prominent star-forming region in a spiral galaxy that resembles our Milky Way. Surprisingly, this source and its host galaxy are radically different from those of the first repeating FRB. The observed diversity in hosts and local environments may point to multiple classes of FRBs with different progenitors.

Fast Radio Bursts (FRBs) are extremely bright radio flashes of millisecond duration and ex- tragalactic origin. Astronomers have known of their existence for only about a decade. The first FRB was discovered in 2007, in archival pulsar data from the 64-meter Parkes radio telescope in New South Wales, Australia. These data revealed a single, bright signal lasting only a few milliseconds (now known as the Lorimer Burst; Lorimer et al., 2007). Since then less than a hundred FRBs have been discovered. Despite estimates that some 1,000 FRBs occur in the sky every day, their nature is one of the most topical questions in astrophysics today (Petroff, et al., 2019; Cordes and Chatterjee, 2019).

Zeroing in on the First Repeater Given the short intrinsic duration of the source’s radio flashes, we can measure the disper- sion delay that the radio waves suffer. The delay is proportional to the column density of electrons from the source to the observer, a quantity called dispersion measure (DM). Tak-

4 GeminiFocus January 2020 / 2019 Year in Review ing into account how electrons are distrib- they uncovered the precise location of FRB uted in the Milky Way and the Universe, the 121102, confirming its extragalactic nature; DM can provide a rough distance estimate the source was found within a low-metallic- to the source. ity star-forming region of an irregular dwarf galaxy some 3 billion light years distant All FRBs show dispersion measures that sig- (Chatterjee et al., 2017; Marcote et al., 2017; nificantly exceed the expected values from Tendulkar et al., 2017; and Bassa et al., 2017). the electrons in our Milky Way. This indi- cates that FRBs originate at cosmological Interestingly, the radio bursts from FRB distances. Those detected lie billions of light 121102 have an extremely high rotation years distant, and are around a trillion times measure — a rotation of the plane of polar- more luminous than the brightest pulsars ization that occurs during the propagation in our Galaxy. There is no clear solution to of electromagnetic waves in a magnetized scale pulsar emission mechanisms to match plasma (Michilli et al., 2018). They were also the luminosity and recurrence rate of FRBs. found spatially coincident with a luminous A large number of possible scenarios have persistent radio counterpart (Chatterjee et been proposed: from giant magnetar flares al., 2017; Marcote et al., 2017). This extreme and colliding neutron stars, to exotic models environment suggests a possible connec- invoking axions and cosmic strings (see e.g., tion between FRBs and other energetic Platts et al., 2018; Petroff et al., 2019). transients, such as long gamma-ray bursts (Metzger et al., 2017). However, the observa- An important step forward in the field oc- tions are also consistent with models invok- curred in 2012 with the discovery of multiple ing extreme objects such as neutron stars or bursts from the same source — FRB 121102 massive black holes (Chatterjee et al., 2017; (Spitler et al., 2014 and 2016; Scholz et al., Michilli et al., 2018). 2016). This discovery rules out cataclysmic models, at least for this particular source. A Within the last year, three new localizations handful of similar repeating FRBs have been have been reported; so far, all are non-re- discovered since. It remains unclear if all peaters. In all three cases, the observed host FRBs have the capability of repeating, or if galaxies are radically different from the first there are two distinct classes of FRBs: repeat- repeating FRB: they are all located in massive ing and non-repeating. To date, only a small galaxies: two reside in the outskirts of ellip- fraction of the observed population of FRBs ticals, and one in a spiral galaxy (Bannister et repeat; perhaps more observing time for al., 2019; Ravi et al., 2019; and Prochaska et longer durations and more constant moni- al., 2019). toring with a more sensitive instrument is re- The large discrepancies between the local quired to detect bursts, we just do not know. environment and host of the first repeater, While single-dish radio telescopes are pow- FRB 121102, when compared with those of erful FRB detectors, they do not have the the apparently non-repeating sources, deep- resolution to localize their host galaxy. Since ened the idea of two distinct classes of FRBs: FRB 121102 exhibits repeating bursts, this repeating and non-repeating. Cleary, we re- allowed for follow-up observations with the quired more localizations of both repeating Karl G. Jansky Very Large Array (VLA), the Eu- and non-repeating FRBs to clarify the nature ropean VLBI Network (EVN), Gemini North, of these events. and the Hubble Space Telescope. In 2017

January 2020 / 2019 Year in Review GeminiFocus 5 Localizing a Second its a low DM, placing it somewhere between Repeating FRB the Galactic halo and a redshift up to ~ 0.1. We observed the field of FRB 180916. The Canadian Hydrogen Intensity Mapping J0158+65 on June 19, 2019, with the EVN, Experiment telescope and Fast Radio Burst combining data from a total of eight radio detector (CHIME/FRB) at the Dominion Ra- telescopes in real time to reach unparalleled dio Astrophysical Observatory in British Co- resolution and sensitivity at 1.7 gigahertz lumbia has proven to be the most prolific (GHz). In parallel, we also recorded from the FRB-detecting machine. Since 2018, the tele- 100-meter Effelsberg telescope in Bad Mün- scope's large collecting area, wide band re- stereifel, Germany, high time and frequency ceiver, and enormous field of view has led to resolution data to directly search for single, the discovery of many new repeating FRBs bright radio bursts coming from the source. (CHIME/FRB Collaboration et al., 2019a,b), in- cluding eight in August 2019. During this EVN run, we detected four bursts from FRB 180916.J0158 + 65, with each burst One of the discovered repeating sources is lasting for, at most, a few milliseconds. As FRB 180916. J0158 + 65. The CHIME/FRB Col- shown in Figure 1, the resolution reached laboration refined the source’s position to a in this observation allowed astronomers to few arcminutes in the sky. This source exhib- pinpoint the origin of the bursts in the sky with an accuracy of about 3 mil- liarcseconds (Marcote et al., 2020). Our team found no persistent radio coun- Figure 1. terparts consistent with The interferometric this position, unlike with localization of FRB FRB 121102 (the first re- 180916.J0158+65 using peater). In archival images the EVN. Panels a to d from the Sloan Digital Sky show the images of the Survey and PanSTARRs, four detected bursts. Panel e shows the this position placed it continuum radio image at the edge of a diffuse, of the field, where no seemingly elliptical gal- significant persistent axy. Was this repeating radio counterparts are FRB, which is in the same reported. Panel f shows kind of environment as the derived positions the non-repeating FRBs, for each of the bursts drastically different from (orange circles) and the that of the first repeater? averaged final position (black square), to which With the GMOS imager/ all plots are referred: spectrograph on the α (J2000) = 01h 58m 8-meter Gemini North 00.75017s (± 2.3 mas), δ telescope, we observed (J2000) = 65˚ 43‘ 00.3152“ this field between July (± 2.3 mas). Error bars and September 2019 with represent 1-σ uncertainty. the g and r photometric

6 GeminiFocus January 2020 / 2019 Year in Review filters, but also with long- Figure 2. slit optical spectroscopy. Gemini North image FRB 180916. J0158 + 65 was and optical spectra found to be at the apex of the source. Panel of a prominent V-shaped A shows the r image star-forming region of a of the host galaxy and a zoom-in of the spiral galaxy located at a star-forming region redshift of 0.0337, or about where FRB 180916. 149.0 Megaparsecs. Figure J0158 + 65 is located 2 shows both the optical (highlighted by the image and the spectra at white cross and red both the location of FRB circle, respectively). 180916. J0158 + 65 and from The uncertainty in the core of the galaxy. the position of FRB 180916.J0158 + 65 is smaller than Towards the the resolution of the optical image. Understanding Panel B shows the of FRBs spectrum extracted from a 2-arcsecond The host and local envi- aperture around ronment of FRB 180916. the position of FRB J0158 + 65 is markedly dif- 180916.J0158 + 65 ferent and less extreme (orange) and a than that of the first re- 5-arcsecond aperture peating FRB, which was around the core of the located inside a low-metal- host galaxy (blue). licity star-forming region of Significant emission a dwarf galaxy, and associ- lines are labeled. ated with a very compact (< 0.7 parsecs) per- supernova or a massive black hole. The for- sistent radio counterpart of unclear origin. mer models could still explain FRB 180916. This new host also contrasts with the mas- J0158 + 65 by invoking an older source, of sive elliptical galaxies where two of the three approximately 300 years, whereas the latter localized non-repeating FRBs were located, seems to be less likely in this case given the where little or no star-formation is present. location in the host galaxy (see Marcote et However, it may be consistent with the star- al., 2020, for further details). forming galaxy associated with the third lo- The origin of FRBs remains unclear, and a calized non-repeater. The observed diversity large number of precise localizations will be in hosts and local environments may point required to establish the ultimate physical to multiple classes of FRBs with different conditions required to produce these kinds progenitors. of bursts. The proximity of FRB 180916. Many scenarios were proposed to explain J0158 + 65, the closest FRB so-far localized, FRB 121102, the first repeating FRB. Several allows dedicated observations across the of them proposed that the bursts originate full electromagnetic spectrum, from radio from a young and rapidly rotating magne- to very high energy gamma rays, to search tar, either interacting with a superluminous for prompt or persistent multiwavelength

January 2020 / 2019 Year in Review GeminiFocus 7 counterparts and to constrain magnetar- Burst of Extragalactic Origin,” Science, 318: 777, based models. 2007

Finally, not only are FRBs an intriguing new Marcote, B., et al., “The Repeating Fast Radio astrophysical transient, but they also pro- Burst FRB 121102 as Seen on Milliarcsecond An- vide the opportunity to investigate the his- gular Scales,” The Astrophysical Journal Letters, tory of the Universe by probing the baryonic 834: L8, 2017 content on large cosmological scales. Marcote, B., et al., “A repeating fast radio burst source localized to a nearby spiral galaxy,” Na- ture, 577: 190, 2020 Benito Marcote is a permanent Support Scien- Metzger, B., Berger, E., and Margalit, B., “Milli- tist of the European VLBI Network and located second Magnetar Birth Connects FRB 121102 at JIVE in the Netherlands. He can be reached at: [email protected] to Superluminous Supernovae, and Long-dura- tion Gamma-Ray Bursts,” The Astrophysical Jour- Kenzie Nimmo is a PhD researcher at the Univer- nal, 841: 14, 2017 sity of Amsterdam and ASTRON in the Nether- Michilli, D., et al., “An extreme magneto-ionic en- lands. She can be reached at: [email protected] vironment associated with the fast radio burst Shriharsh Tendulkar is a former postdoctoral source FRB 121102,” Nature, 553: 182, 2018 fellow at the McGill Space Institute and Depart- Petroff, E., Hessels, J. W. T., Lorimer, D. R., “Fast ra- ment of Physics, McGill University. He can be dio bursts,” Astronomy and Astrophysics Review, reached at: [email protected] 27: 4, 2019 Platts, E., et al., “A Living Theory Catalogue for References Fast Radio Bursts,” Physics Reports, 821: 1P, 2018 Prochaska, J. X., et al., “The low density and mag- Bannister, K. W., et al., “A single fast radio burst netization of a massive galaxy halo exposed by localized to a massive galaxy at cosmological a fast radio burst,” Science, 366: 231, 2019 distance, ” Science, 365: 565, 2019 Ravi, V., et al., “A fast radio burst localized to a Bassa, C. G., et al., “FRB 121102 Is Coincident massive galaxy,” Nature, 572: 352, 2019 with a Star-forming Region in Its Host Galaxy,” The Astrophysical Journal Letters, 843: L8, 2017 Scholz, P., et al., “The Repeating Fast Radio Burst FRB 121102: Multiwavelength Observations CHIME/FRB Collaboration, et al., “A second and Additional Bursts,” The Astrophysical Jour- source of repeating fast radio bursts,” Nature, nal, 833: 177, 2016 566: 235, 2019 Spitler, L. G., et al., “Fast Radio Burst Discovered CHIME/FRB Collaboration et al., “CHIME/FRB De- in the Arecibo Pulsar ALFA Survey,” The Astro- tection of Eight New Repeating Fast Radio Burst physical Journal, 790: 101, 2014 Sources,” The Astrophysical Journal Letters, 885: L24, 2019 Spitler L. G., et al., “A repeating fast radio burst,” Nature, 531: 202, 2016 Chatterjee, S., et al., “A direct localization of a fast radio burst and its host,” Nature, 541: 58, Tendulkar, S. P., et al., “The Host Galaxy and 2017 Redshift of the Repeating Fast Radio Burst FRB 121102,” The Astrophysical Journal Letters, 834: Cordes, J. M., and Chatterjee, S., “Fast Radio L7, 2017 Bursts: An Extragalactic Enigma,” Annual Review of Astronomy and Astrophysics, 57: 417 Lorimer, D. R., et al., “A Bright Millisecond Radio

8 GeminiFocus January 2020 / 2019 Year in Review Tom Geballe October 2019

NGC 2071-IR: A Who-dunnit Mystery

Two recently retired Gemini staff members (author Tom Geballe and Dolores Walther) have utilized Gemini North to obtain the sharpest composite infrared images ever of the chaotic core of one of the nearest star-forming clouds. These images, combined with key infrared spectral signatures of two of the embedded protostars, are helping astronomers determine the causes of the mayhem.

Star formation can be a messy process. When gravity causes a portion of a calm interstellar gas cloud to collapse, and a star is born, some of that infalling gas is violently blown back into the surrounding cloud, disrupting much of it. In the process, small portions of the cloud are briefly shock-heated to temperatures of thousands of degrees. If only a single protostar at a time is engaged in this destructive activity, astronomers can usually identify it. But when more than one protostar in a cloud is doing this at the same time, understanding what is going on, including determining which protostars are respon- sible for which parts of the disruption, is a challenge. Such is the case with one of the nearest star-forming clouds to the Sun, NGC 2071. The core of this cloud, known as NGC 2071-IR because of its bright infrared emission, has long fascinated Dolores Walther, who retired in 2017 as head of Gemini North’s crew of Science Operations Specialists. Walther had always wanted to use Gemini and its powerful infrared instruments to get a better look at NGC 2071-IR and solve some of its mysteries. I joined in the study and co- published the results with her in the April 20, 2019, issue of The Astrophysical Journal.

January 2020 / 2019 Year in Review GeminiFocus 9 Figure 1. Gemini North / NIRI infrared image of NGC 2071-IR in the combined light of the broad J, H, and K filters and a narrow band filter at the wavelength of a line of shock-excited molecular hydrogen

(H2 ). The resolution of the image is ~0.40”, at least three times higher than any previously obtained images of the region. The field of view is 2’ x 2’. North is up and East is to the left. Most of the finely structured

emission is from H2 , while the more smoothly distributed emission is

the combined light of H2 and starlight scattered off of dust particles. Identified protostars have been labelled. Color composite image created by Gemini Science Operations and IRS 6 A&B, is emitted by shock-heated Specialist Jennifer Miller. To help identify the young culprits respon- molecular hydrogen, where gas ejected sible for disrupting the core of NGC 2071, from a protostar is colliding with quiescent Walther and I used the Gemini North Near- gas in the surrounding cloud. A fainter V InfraRed Imager and spectrometer (NIRI) extension can be seen at lower right. Both in 2017 and 2018 to dig deeper into the extend far beyond the edges of the image. complex region. Figure 1 shows our results Such “bipolar outflows” of gas are common- — the sharpest composite infrared image ly observed from stars accreting material ever obtained of the region. We combined from their natal clouds. images taken individually through several While it was originally supposed that IRS 1, filters — one of which was sensitive to the by far the most luminous and likely the most

emission of hot molecular hydrogen (H2). massive protostar in NGC 2071-IR, was gen- Putting them together created a coherent erating this bipolar outflow, these new data, picture. along with radio and infrared observations Stars, light from glowing gas, and light re- published by other scientists, strongly sug- flected off of dust particles are readily ap- gest that the less luminous IRS 3 is the culprit. parent in the image. The complex V-shaped To the left of the center of the image lies the

structure extending from the center of the brightest region of H2 line emission in NGC image toward the upper left, whose left arm 2071-IR, which others had recently suggest- extends across the positions of IRS 2 A&B ed might be associated with IRS 1. In our pa-

10 GeminiFocus January 2020 / 2019 Year in Review per, Walther and I concur with this sugges- One especially mysterious source, detect- tion. In addition, we propose that the rather ed in the infrared for the first time by NIRI, compact and amorphous appearance of this but found earlier at radio wavelengths by region is due to the outflow of material be- the Very Large Array (in New Mexico) and ing directed almost exactly toward the Sun. dubbed VLA-1, shows signs of activity at ra- The Gemini image gives us a view somewhat dio wavelengths. Located between IRS 1 and akin to looking down the barrel of a cannon IRS 3, but apparently much more deeply bur- that has just been fired. ied in the star-forming cloud than either of them, it may be an important key to under- We also captured the most detailed infrared standing the entire region. spectra ever obtained of IRS 1, the bright fuzzy object at the middle of the image, and Identifying all of the active protostars within IRS 3, the much fainter object located close NGC 2071-IR will allow us to complete the to IRS 1, just to its upper right. The emission picture of how these violent activities are lines of atomic and molecular hydrogen, sculpting the surrounding cloud. The active ionized iron, and hot carbon monoxide that ones not only could be preventing more we found in their spectra attest to both stars stars from forming, but also could be dis- generating intense outflowing winds. rupting the abilities of other younger proto- stars to collect nearby gas, placing limits on Other protostars within NGC 2071-IR could how massive they can become. Walther and I also be producing outflows that are disrupt- hope that additional spectra will give a clear- ing the cloud. If we are correct, NGC 2071- er understanding of the activities within this IR may be generating more outflows simul- fascinating cloud. taneously than any cloud core in the solar neighborhood. However, spectra do not ex- ist of most of the other stars in Figure 1. For additional details see Walther, D. M., and We are hoping to be granted additional time Geballe, T. R., The Astrophysical Journal, 875: to obtain infrared spectra of all of them. A 153, 2019, or arXiv 1903.03583 crude spectrum of IRS 7, located far from IRS 1 and IRS 3, that we obtained 30 years ago, Tom Geballe is Emeritus Astronomer at Gemini shows strong evidence of outflow activity. Observatory. He can be reached at: [email protected]

January 2020 / 2019 Year in Review GeminiFocus 11 October 2019 Rachel Matson and Andy Stephens

Speckle Imaging Takes Gemini to Its Diffraction Limit For nearly a decade, speckle imaging at Gemini Observatory has produced an abundance of notable results, including a monumental breakthrough in exoplanet research in binary star systems. Now with the recent addition of an innovative pair of powerful high-resolution speckle instruments to permanently reside at Gemini North and South, the Observatory stands on the forefront of high-resolution ground-based exploration.

Recently our team at the NASA Ames Research Center authored a high-impact journal ar- ticle that featured key Gemini data on the transit by a giant exoplanet of one of the com- ponents of the Kepler-13AB binary star system. system. This study, led by Steve Howell, not only classified the Jupiter-sized exoplanet (Kepler-13b) in this close binary system but, in a first for ground-based imaging, conclusively determined which star the planet orbits. The Gemini press release on our finding is reprinted starting on page 15 of this issue, and The Astronomical Journal paper is available here. To execute this type of diffraction-limited science and uncover the hidden secrets of close exoplanet binary star systems, in which about one half of all exoplanets reside, our team designed under Howell’s leadership an innovative pair of twin instruments that perform high-resolution “speckle imaging” — collecting a thousand 60-millisecond exposures ev- ery minute; after processing this large amount of data, the final images are free of the adverse effects of atmospheric turbulence which can bloat, blur, and distort star images.

12 GeminiFocus January 2020 / 2019 Year in Review Our team aptly named the two permanently mounted instru- ments Zorro and ‘Alopeke (Figure 1), which come from the Span- ish and Hawaiian (respectively) words for Fox — because the in- struments are both speedy and sly like foxes. The instruments were built to take advantage of innovation and crafty approaches to problems with only a fraction of the resources necessary for most 8-meter-class telescope in- struments. The power of these instruments Figure 1. was demonstrated when Howell and his The Rise and Promise of Speckle The ‘Alopeke/Zorro team used ‘Alopeke to probe the Kepler- Imaging at Gemini team at Gemini North 13AB system. ‘Alopeke sharply resolved the Speckle imaging at Gemini began in 2012 preparing ‘Alopeke for two stars (Kepler A and B), and captured a installation. clear drop in the light from Kepler A, proving when the Differential Speckle Survey Instru- Credit: Alison Peck that the planet orbits the brighter of the two ment (DSSI; designed by Elliott Horch) came stars. Moreover, as ‘Alopeke simultaneously to the Observatory as a visiting instrument. provides data at both red and blue wave- This precursor to ‘Alopeke and Zorro was lengths, the researchers could see that the granted 10 hours on Gemini North to ob- dip in the star’s blue light was about twice as serve high-priority planet candidates from deep as the dip seen in red light. NASA’s (now-retired) Kepler mission, whose prime objective was to explore the structure As a very extended atmosphere would more and diversity of exoplanetary systems, in- effectively block light at blue wavelengths, cluding estimating how many planets there the researchers characterize Kepler-13b as are in multiple-star systems. a Jupiter-like gas-giant exoplanet with a “puffed up” atmosphere due to exposure to To search for planets around other stars, the the tremendous radiation from its host star; Kepler Space Telescope would stare at thou- Thus, these multi-color speckle observations sands of stars and look for a slight decrease give us a first tantalizing glimpse into the ap- in brightness, indicating that a planet had pearance of this distant world orbiting a star transited (crossed in front of) the star as in a binary system — something we know viewed from Earth. While the transit method very little about. is very successful at finding planets, other phenomena can mimic the signature of a Our work with Kepler-13b stands as a model planet. Because of this, other methods must for future research on exoplanets in multiple be used to confirm whether a planet caused star systems. The observations highlight the the star’s dimming. ability of high-resolution imaging with large telescopes like Gemini, not only to assess High-resolution speckle imaging enables as- which stars with planets are in binaries, but tronomers to not only resolve other objects also robustly determine which of the stars near the star hosting the planet candidate, the exoplanet orbits. but detect or rule out other, non-planetary

January 2020 / 2019 Year in Review GeminiFocus 13 objects that can panion, speckle imaging provides the po- cause a star’s light sition and separation from the host star, as to dim (speckle can- well as color and contrast information that not see planets). greatly reduces the likelihood of false posi- This is achieved by tives and improves the estimates of the exo- employing statis- planet size. tical techniques to assess whether Zorro and ‘Alopeke: the observed dim- Specifics for Users ming is likely to be a true transit by an ‘Alopeke and Zorro add great new capa- orbiting planet or a bilities, and having identical instruments “false positive.” Us- on both Gemini telescopes allows collect- ing this technique, ing homogeneous datasets over the whole the DSSI observa- sky. The speckle mode provides diffraction- tions at Gemini limited (0.016 arcsecond Full-Width at Half- North in 2012 helped confirm over a dozen Figure 2. Maximum at 500 nm and 0.025" at 800 nm) planet candidates, including the five-planet Speckle image resolution imaging at optical wavelengths system Kepler-67; DSSI would eventually reconstruction of Pluto over a narrow field of view (~6 arcseconds). provide more than 2,100 observations of Ke- and Charon obtained The wide-field mode provides high-sensitiv- in visible light at 692 pler planet candidate host stars. ity natural-seeing imaging with virtually no nanometers (red) with Based on the success of DSSI, and the need readout delay in the standard Sloan broad- the Gemini North to validate and characterize the 4,000 exo- band filters over a moderate field of view 8-meter telescope using planet candidates discovered to (~60 arcseconds). the Differential Speckle date by NASA’s Kepler/K2 Space Both instruments are considered Survey Instrument (DSSI). Telescope and the Transiting Exo- "permanent resident" visiting Resolution of the image is planet Survey Satellite (TESS), How- about 20 milliarcseconds instruments, meaning they are ell initiated the design of two new average. This is the first available throughout the semes- speckle instruments: ‘Alopeke and speckle reconstructed ters for regular queue and Fast image for Pluto and Zorro, which our team went on Turnaround proposals. This makes Charon from which to build at NASA Ames Research them great for programs that astronomers obtained not Center. The twin instruments need simultaneous photometry only the separation and each use two electron-multiply- in two filters, variability studies, position angle for Charon, ing CCDs and combinations of and rapid events like occultations, but also the diameters of narrow-band (40- to 50-namom- which also benefit from the flexibility of the two bodies. North is eter-wide) filters to provide simultaneous Gemini’s queue scheduling. up, east is to the left, and two-color diffraction-limited photometric the image section shown and astrometric information at optical wave- here is 1.39 arcseconds lengths. Differential Speckle Imaging across. at Gemini Each instrument can also identify back- Credit: Gemini ground objects and companion stars — to Some Science Highlights Observatory/NSF/NASA/ within < 0.1 to 1.2 arcseconds of, and up to AURA Speckle imaging at Gemini Observatory is a 10 magnitudes fainter than, the exoplanet’s forefront technology allowing researchers to host star — that can contaminate exoplanet push the limits of high-resolution imaging (Fig- transit detections. For any detected com- ure 2). The following science references pro-

14 GeminiFocus January 2020 / 2019 Year in Review vide a sampling of past successes while hint- massive stars, halo binaries, massive young ing at what is possible with these instruments. stellar objects, and

• Pluto + Charon imaging (Howell et al., • Deriving/improving mass-luminosity rela- 2012). tionships for low-metallicity stars and M- dwarfs. • TRAPPIST-1 (Howell et al., 2016) • Half of all exoplanet host stars are binary (Matson et al., 2018) Rachel Matson is an astronomer at the United • Kepler-13AB: see press release below. States Naval Observatory and is a member of the ‘Alopeke/Zorro team. She can be reached at: Other science being pursued: [email protected] • Ages of moving groups (and imaged plan- ets in the moving groups) via dynamical mass determinations using Gemini speck- Andy Stephens is an instrument scientist at Gem- ini North and is a member of the ‘Alopeke/Zorro le + GPI team. He can be reached at: • Light curves of white dwarfs [email protected] • Studying multiplicity of nearby M-dwarfs,

Gemini Press Release Exoplanets Can’t Hide Their Secrets from Innovative New Instrument A cunning new instrument at Gemini Observatory has achieved what was once thought impossible — namely, the characterization of an exoplanet orbiting a binary star and determining which star of the pair it orbits.

In an unprecedented feat, an American research team discovered hidden secrets of an elusive exoplanet using a powerful new instrument at the 8-meter Gemini North tele- scope on Maunakea in Hawai‘i. The findings not only classify a Jupiter-sized exoplanet in a close binary star system, but also conclusively demonstrate, for the first time, which star the planet orbits. The breakthrough occurred when Steve B. Howell of the NASA Ames Research Center and his team used a high-resolution imaging instrument of their design — named ‘Alopeke (a contemporary Hawaiian word for Fox). The team observed exoplanet Kepler-13b as it passed in front of (transited) one of the stars in the Kepler-13AB binary star system some 2,000 light years distant. Prior to this attempt, the true nature of the exoplanet was a mystery.

January 2020 / 2019 Year in Review GeminiFocus 15 Artist's conception of the Kepler-13AB binary star system as revealed by observations including the new Gemini Observatory data. The two stars (A and B) are large, massive bluish stars (center) with the transiting "hot Jupiter" (Kepler-13b) in the foreground (left corner). Star B and its low mass red dwarf companion star are seen in the background to the right. Credit: Gemini Observatory/NSF/AURA/ Artwork by Joy Pollard

“There was confusion over Kepler-13b: was lence — which can bloat, blur, and distort it a low-mass star or a hot Jupiter-like world? star images. So we devised an experiment using the sly “About one half of all exoplanets orbit a star instrument ‘Alopeke,” Howell said. The re- residing in a binary system, yet, until now, we search was recently published in The Astro- were at a loss to robustly determine which nomical Journal. "We monitored both stars, star hosts the planet,” said Howell. Kepler A and Kepler B, simultaneously while looking for any changes in brightness dur- The team’s analysis revealed a clear drop ing the planet’s transit,” Howell explained. in the light from Kepler A, proving that the “To our pleasure, we not only solved the planet orbits the brighter of the two stars. mystery, but also opened a window into a Moreover, ‘Alopeke simultaneously provides new era of exoplanet research.” data at both red and blue wavelengths, an unusual capability for speckle imagers. Com- “This dual win has elevated the importance paring the red and blue data, the research- of instruments like ‘Alopeke in exoplanet ers were surprised to discover that the dip in research,” said Chris Davis of the National the star’s blue light was about twice as deep Science Foundation, one of Gemini’s spon- as the dip seen in red light. This can be ex- soring agencies. “The exquisite seeing and plained by a hot exoplanet with a very ex- telescope abilities of Gemini Observatory, as tended atmosphere, which more effectively well as the innovative ‘Alopeke instrument blocks the light at blue wavelengths. Thus, made this discovery possible in merely four these multi-color speckle observations give hours of observations." a tantalizing glimpse into the appearance of ‘Alopeke performs “speckle imaging,” collect- this distant world. ing a thousand 60-millisecond exposures Early observations once pointed to the tran- every minute. After processing this large siting object being either a low-mass star or amount of data, the final images are free of a brown dwarf (an object somewhere be- the adverse effects of atmospheric turbu- tween the heaviest planets and the lightest

16 GeminiFocus January 2020 / 2019 Year in Review stars). But Howell and his team’s research is the highest-quality image that a telescope almost certainly shows the object to be a Ju- can produce, effectively obtaining space- piter-like gas-giant exoplanet with a “puffed based resolution from the ground — making up” atmosphere due to exposure to the tre- these instruments superb probes of extraso- mendous radiation from its host star. lar environments that may harbor planets. ‘Alopeke has an identical twin at the Gem- The discovery of planets orbiting other stars ini South telescope in Chile, named Zorro, has changed the view of our place in the which is the word for Fox in Spanish. Like Universe. Space missions like NASA’s Kepler/ ‘Alopeke, Zorro is capable of speckle imag- K2 Space Telescope and the Transiting Exoplan- ing in both blue and red wavelengths. The et Survey Satellite (TESS) have revealed that presence of these instruments in both hemi- there are twice as many planets orbiting spheres allows Gemini Observatory to re- stars in the sky than there are stars visible to solve the thousands of exoplanets known to the unaided eyes; to date the total discovery be in multiple star systems. count hovers around 4,000. While these tele- "Speckle imaging is experiencing a renais- scopes detect exoplanets by looking for tiny sance with technology like fast, low noise dips in the brightness of a star when a planet detectors becoming more easily available," crosses in front of it, they have their limits. said team member and ‘Alopeke instrument “These missions observe large fields of view scientist Andrew Stephens at the Gemini containing hundreds of thousands of stars, North telescope. "Combined with Gemini's so they don’t have the fine spatial resolu- large primary mirror, ‘Alopeke has real poten- tion necessary to probe deeper,” Howell said. tial to make even more significant exoplanet “One of the major discoveries of exoplanet discoveries by adding another dimension to research is that about one-half of all exo- the search." planets orbit stars that reside in binary sys- First proposed by French astronomer An- tems. Making sense of these complex sys- toine Labeyrie in 1970, speckle imaging is tems requires technologies that can conduct based on the idea that atmospheric turbu- time sensitive observations and investigate lence can be “frozen” when obtaining very the finer details with exceptional clarity.” short exposures. In these short exposures, “Our work with Kepler-13b stands as a stars look like collections of little spots, or model for future research of exoplanets in speckles, where each of these speckles has multiple star systems,” Howell continued. the size of the telescope’s optimal limit of “The observations highlight the ability of resolution. When taking many exposures, high-resolution imaging with powerful tele- and using a clever mathematical approach, scopes like Gemini to not only assess which these speckles can be reconstructed to form stars with planets are in binaries, but also the true image of the source, removing the robustly determine which of the stars the effect of atmospheric turbulence. The result exoplanet orbits.”

January 2020 / 2019 Year in Review GeminiFocus 17 July 2019 Jennifer Hanley

Neptune’s Moon Triton Fosters Rare Icy Union Observations from the visiting IGRINS spectrograph at Gemini South reveal for the first time beyond the lab, an extraordinary union between carbon monoxide and nitrogen ices. The discovery offers insights into how this volatile mixture can transport material across Neptune’s moon Triton via geysers, trigger seasonal atmospheric changes, and provide a context for conditions on other distant, icy worlds.

Neptune’s largest moon Triton has been mysterious ever since its discovery in 1846 as the only large retrograde-orbiting satellite: in 1989, the Voyager 2 flyby (Figure 1) showed geologic activity despite extremely cold temperatures, and later ground-based observations showed it and Pluto sharing similar sur- face compositions. Triton is now thought to be a captured dwarf planet from the Kuiper Belt, but further observations are necessary to unmask the moon’s many secrets. Until we can return to the Neptunian system (and there are proposals under- way), our best way to understand Triton is through telescopic observations, laboratory investigations, and chemical modeling. Our research at the As- trophysical Materials Laboratory at Northern Arizona University in Flagstaff, Arizona, has combined these techniques in order to study the composition Figure 1. of Triton’s surface. For the telescopic observations, we utilized the visiting high-resolution Voyager 2 image of Triton near-infrared spectrometer IGRINS — built as a collaboration between the University of showing the south polar region with dark streaks Texas at Austin and the Korea Astronomy and Space Science Institute (Park et al., 2014; produced by geysers visible Mace et al., 2018) — which allowed us to acquire a high signal-to-noise spectrum of Triton on the icy surface. to make an unprecedented discovery beyond the lab. We recently published the synthesis Credit: NASA/JPL of these results in The Astronomical Journal (Tegler et al., 2019).

18 GeminiFocus January 2020 / 2019 Year in Review Laboratory Investigations

While previous studies have shown that carbon monoxide (CO) and nitrogen (N2) ices exist on Triton, we decided to investigate their spectral features — specifically, we wanted to see how the spectra changed as a function of the mix- ing ratio between the CO and N2. In order to study spectroscopic telescopic data, one needs to have an appropriate library of labora- tory spectra. Most laboratory experiments Figure 2 represent the spectrometer beam Figure 2. collect spectra of thin ice samples of only through the sample. Thermometers (T1 and (A) Schematic diagram of microns thick. These experiments are su- T2) and heating elements (H1 and H2) con- the Astrophysical Materials perb at studying intrinsically strong absorp- trol the temperature of the sample down to Laboratory thick cell in cross section as seen from tion bands. Thin film experiments are not as 30 Kelvin (K). Further details concerning the the side. (B) The optical good for studying intrinsically weak absorp- cell are described in Tegler et al. (2019). train in our experiment tion bands. Longer path lengths are needed We measured the absorption coefficient of as seen from above. The to study these bands. In the Astrophysical spectrometer beam is varying mixtures of CO and N2, and noticed Materials Laboratory, we have a unique ex- represented by dashed lines. an unidentified, weak band that wasn’t in ei- perimental setup that enables us to study ice Only the infrared detector ther pure species. This band was strongest samples as thick as 2 centimeters. As a result, was used in the experiments when the ratio of CO to N2 was at 50:50 (Fig- described here. we can study very weak absorption bands. ure 3). The spectra shown in Figure 3 are all Our thick cell is mounted on top of a cryo- taken at 60 K, where the ice mixture is in the cooler. Gas enters the cell from above via b-phase. A maximum band strength for sam- a fill tube (Figure 2a). The dotted lines in ples with nearly equal amounts of CO and

Figure 3. Spectra of CO/N2 ice samples with the CO abundance ranging from (a) 0% to 40% and (b) 60% to 100%. In panel (a), the spectra show the new band near 4467 cm-1. The new band is not present in the pure N2 sample (black line) and increases in strength with increasing CO abundance. The saturated band at 4252 cm-1 is a CO overtone and the weak, broad band at 4654 cm-1 is N2. The strength of the weak, unidentified band at a CO abundance of 60% in panel (a) is nearly the same as its strength at 40% in panel (b) and then decreases in strength with increasing CO abundance. The band is not present in the pure CO ice sample in panel (b). Figure and caption modified from Tegler et al. (2019).

January 2020 / 2019 Year in Review GeminiFocus 19 the CO fundamental and the N2 fundamen- tal. For this to happen, the CO and N2 mol- ecules have to be intimately mixed together.

Triton Observations

One exciting aspect of this work is that if we detect this band on any astronomical object we know that carbon monoxide and nitro- gen must be intimately mixed together at the molecular level. That excitement rose as we used the 8-meter Gemini South Tele- scope in Chile on the night of July 2, 2018, to explore Triton’s icy surface with IGRINS. The Figure 4. N2, and its absence in pure N2 and pure CO, combination of this large aperture telescope A portion of an reinforces the idea that the band is caused with the phenomenal throughput of IGRINS 80-minute, binned- by the CO and N2 molecules being near each over long exposure times, coupled with the spectrum of Triton other, and probably interacting. high spectral resolution gives the ability to taken with the 8.1- bin to get desired signal-to-noise ratio. All meter Gemini-South this was necessary to even have a chance to telescope and the Molecular Understanding detect this weak feature. We summed our IGRINS spectrometer individual Triton spectra to obtain a single (red squares). The Individually, carbon monoxide and nitrogen broad absorption in spectrum with a total exposure time of 80 ices each absorb their own distinct wave- the Triton spectrum is minutes. consistent with the broad lengths of infrared light, but the tandem Since our objective was to detect the spec- absorption of the two- vibration of an ice mixture absorbs at an ad- molecule combination ditional, distinct wavelength. Looking at the trally broad CO-N2 combination band at band at 2.239mm (4466.5 pure species, we are able to identify the fun- 2.239 microns (mm) (4466.5 cm-1), we used in- cm-1) in our laboratory damental vibrational frequencies, as well as verse variance weighting to bin the spectrum transmission spectrum their overtones and combinations. However, into blocks of 64 pixels, and thereby improve (blue line). Both broad this band (first noted but not identified by the signal-to-noise ratio of the Triton spec- bands are inconsistent trum. The binned spectrum had a resolution with the telluric (black Quirico and Schmitt, 1997) did not align with squares at top of figure) any known features. Since the band had of λ/Δλ = 2,500. and the solar (black maximum strength in samples with nearly As can be seen in Figure 4, there is a broad squares at bottom of equal amounts of CO and N2, and was ab- feature in the Triton spectrum (red squares) figure) spectra. sent in pure N2 and pure CO, we realized it located at the same position as the band in Figure and caption must arise from both molecules simultane- our laboratory spectrum of 8% CO and 92% modified from Tegler, et ously. We refer to the band as a two-mole- N2 ice sample at 60 K (blue line). For com- al. (2019) cule combination band. parison, we show absorption due to Earth's We were able to quantitatively show the atmosphere (black squares seen at top) and new band was the result of the simultaneous reflected sunlight,i.e., Fraunhofer lines (black excitation of adjacent CO and N2 molecules. squares at bottom of figure). The telluric and Specifically, we found the energy (wave- solar spectra are binned to the same resolu- number) required to excite the weak, un- tion as the binned Triton spectrum, i.e., λ/Δλ identified band was equal to the sum of the = 2,500. The vertical dotted line marks the energies (wavenumbers) required to excite wavelength of maximum absorption by the

20 GeminiFocus January 2020 / 2019 Year in Review broad band in our Triton spectrum. The band We expect that these findings will shed light in our Triton spectrum coincides with the on the composition of ices and seasonal vari- 2.239 mm (4466.5 cm-1) band in the labora- ations on other distant worlds beyond Nep- tory spectrum. tune. Astronomers have suspected that the The strength of absorption of Triton’s N2 mixing of carbon monoxide and nitrogen ice and CO ice bands varies with longitude, by exists not only on Triton, but also on Pluto, roughly a factor of two, with the strongest where the New Horizons spacecraft found absorption being on the leading part of the the two ices coexisting in Sputnik Planitia sub-Neptune hemisphere (longitude ~50˚ (Protopapa et al., 2017) — an icy basin that East; see Grundy et al., (2010). We observed has apparently caused Pluto’s entire crust when Triton was at a sub-Earth longitude of to shift over time. The same may be true 113˚ East, not far from the maximum in N2 for more recently discovered small planets and CO absorption. like Eris and Makemake, both of which host volatile ices like those on Pluto and Triton. This Gemini finding is the first direct spec- Looking ahead troscopic evidence of these ices mixing and absorbing this type of light on either world. On distant Triton, carbon monoxide and ni- trogen freeze as solid ices. They can form Jennifer Hanley is an astronomer at Lowell Obser- their own independent ices, or condense to- vatory. She can be reached at: gether in the icy mix detected in the Gemini [email protected] data. Our discovery, for the first time beyond the lab, of an extraordinary union between References carbon monoxide and nitrogen ices is impor- tant, as it could be involved in Triton’s iconic Grundy, W. M., et al., “Near-infrared spectral moni- toring of Triton with IRTF/SpeX II: Spatial distribu- geysers — first seen in Voyager 2 spacecraft tion and evolution of ices,” Icarus, 205: 594-604, images as dark, windblown streaks on the 2010 moon’s south polar region back in 1989 (Fig- Mace, Gregory, et al., “IGRINS at the Discovery Chan- ure 1). nel Telescope and Gemini South, SPIE, 10702: 18 pp., 2018 Since Voyager 2’s discovery of the geysers, theories have focused on an internal ocean Park, Chan, et al., “Design and early performance of IGRINS (Immersion Grating Infrared Spectrometer),” as one possible source of erupted material. SPIE, 9147: 12, 2014 Or, the geysers may erupt when the sum- Protopapa, S., et al., “Pluto's global surface composi- mertime Sun heats this thin layer of volatile tion through pixel-by-pixel Hapke modeling of New ice on Triton’s surface, potentially involving Horizons Ralph/LEISA data,” Icarus, 287: 218-228, the mixed carbon monoxide and nitrogen 2017 ice revealed by the Gemini observation. That Quirico, Eric., and Schmitt, Bernard, “A spectroscop- ice mixture could also migrate around the ic study of CO diluted in N2 ice: Applications for Tri- surface of Triton in response to seasonally ton and Pluto,” Icarus, 128: 181-188, 1997 varying patterns of sunlight. Tegler, S. C., et al., “A New Two-Molecule Combina- tion Band as Diagnostic of Carbon Monoxide Di- Seasons progress slowly on Triton, as Neptune luted in Nitrogen Ice On Triton,” The Astronomical takes 165-Earth years to orbit the Sun. A sea- Journal, 158: 17, 2019 son on Triton lasts a little over 40 years; Triton passed its southern summer solstice mark in 2000, leaving about 20 more years to conduct further research before its autumn begins.

January 2020 / 2019 Year in Review GeminiFocus 21 April 2019 Vinicius Placco

Making Good Use of Bad Weather: Finding Metal-poor Stars Through the Clouds

The Gemini telescopes played a key role in identifying low- metallicity stars in the Galaxy by gathering medium-resolution spectroscopic GMOS data for 666 bright (V <14) stars under poor weather conditions. In-depth studies of these stars provide a unique opportunity to witness not only the chemical and dynamical evolution of the Milky Way but also to identify and distinguish between a number of possible scenarios for the enrichment of star- forming gas clouds in the early Universe.

Low-metallicity stars are the Rosetta Stones of stellar astrophysics. Encoded in the atmo- sphere of these low-mass, long-lived relics are the signatures of nucleosynthetic process- es, by which the first light elements were cooked up; this could have occurred as early as a few tens of millions of years after the Big Bang. The first generation of stars to be born in the Universe were formed (mostly) out of hydrogen and helium. These are thought to be massive (tens to hundreds of solar masses), short-lived, and to end their lives in an ex- plosive event that would seed the up-to-then chemically pristine Universe with most of the chemical species we know today. By studying the mass distribution of these so-called Population III (Pop. III) stars it is possible to constrain models for the chemical evolution of the Universe at high-redshifts and the formation and evolution of our Galaxy. However, most (if not all) of the Pop. III stars are long gone, and the only way to infer their existence is by observing the low-mass stars formed right after.

22 GeminiFocus January 2020 / 2019 Year in Review Extremely Metal-poor Stars: candidates from broadband or narrowband Windows into the Early Universe photometry. Even though these methods can successfully identify metal-poor star The only way to understand and character- candidates, they become more and more ize the first generation of stars is to look for uncertain as metallicities decrease. As a re- their direct descendants that would still be sult, medium-resolution (R = λ/Δλ ≈ 1,500) alive today: second-generation low-mass, spectroscopy becomes a valuable tool not low-metallicity stars. A subset of these, the only for pre-selection of targets to be fol- Extremely Metal-Poor (EMP; [Fe/H] < -3.0) lowed-up in high-resolution (R ≈ 30,000) but stars, with iron abundances of 1/1,000 of also for parameter determination and stellar the solar value, are believed to carry in their population studies. atmospheres the chemical fingerprints of Recently, our team published two studies in the evolution of as few as one Pop. III mas- The Astronomical Journal (Placco et al., 2018; sive star. Apart from the very low iron abun- Placco et al., 2019), aiming to increase the dance, the majority (more than 60%) of inventory of EMP star candidates observed the observed EMP stars show a very strong with medium-resolution spectroscopy. We molecular carbon signature in their optical observed these stars over the course of sev- spectrum. Such high carbon abundances are en semesters (from 2014A to 2017A) with a one of the expected yields of the final stages variety of telescopes, including the Gemini of evolution of zero-metallicity Pop. III stars North and South telescopes, the Southern and can help trace back the nature of the Astrophysical Research telescope, Kitt Peak first stars in the Universe. National Observatory’s Mayall telescope, and the European Southern Observatory’s Finding the Needle in the New Technology Telescope. In total, 2,551 Haystack stars were observed. Identifying such pristine objects is a chal- We selected the (bright) candidates from Figure 1. lenging endeavor. EMP stars are intrinsically two sources — the RAdial Velocity Experi- Equatorial and Galactic coordinate distribution rare (less than 30 stars identified to date with ment (RAVE) and the Best & Brightest Survey (B&B) — and used the Gemini North and of the stars observed [Fe/H] < -4.0) and can only be properly char- with Gemini North and South telescopes to observe 666 stars out of acterized as such via spectroscopic studies. Gemini South in poor In addition, metal-poor stars are generally the 2,551. Figure 1 shows the distribution of weather conditions. found in higher fractions in the halo populations of the Galaxy, making most of them faint and "expensive" in terms of tele- scope time. Thus, it is important to have reliable selection criteria in the search for the brightest metal-poor star candidates for high-resolution spectroscopic follow-up.

Since changes in metallicity af- fect the colors in optical wave- lengths in predictable ways, we pre-selected a number of such January 2020 / 2019 Year in Review GeminiFocus 23 equatorial and Galactic coordinates for the size of the symbols is proportional to the Gemini targets, color-coded by catalog. All exposure time for each object, in seconds. of these spectra, interestingly, were gath- It is interesting to note the large spread in ered exclusively as part of the Poor Weather counts for stars with similar exposure times proposal cycle offered by the Gemini Obser- in a narrow range of magnitudes (e.g., blue vatory. filled circles at V ~13.5). Similarly, there are cases where it took up to four times longer to gather the same counts for stars with Big Eyes and Cloudy Nights similar magnitudes (e.g., red filled squares The targets selected from the RAVE and B&B at V ~12.5 and Counts ~1,000). These are catalogs were bright enough to be observed telltale signs of the highly variable weather under poor, but usable, conditions, as part of conditions (mostly image quality and cloud the Poor Weather programs at Gemini. Such cover) in which these stars were observed. programs are executed only when noth- In total, seven GMOS Poor Weather programs ing in the regular queue is observable and were executed (three in the North and four hence considered "weather loss" for time ac- in the South) spanning four semesters (from counting purposes. The targets followed-up 2015A to 2016B). Those programs had 310 as part of this effort had no observing condi- hours of allocated time. By adding all the tion constraints (CC = Any, IQ=Any, SB=Any/ exposure times, there were about 89 hours Bright, and WV = Any), and spectra were of on-target observations for the 666 stars, taken using the Gemini Multi-Object Spec- averaging about 8 minutes per exposure. trograph (GMOS; North and South) B600 Adding ~12 minutes for acquisition and cali- gratings and 1-arcsecond slits. brations, these were 20-minute observing Figure 2 shows the total counts at 4000 Ång- blocks, giving an average of three stars per stroms in the observed spectra as a function hour. As a result, assuming 666 targets took of the visual magnitude of the stars. The 222 hours of observing time, the efficiency Figure 2. Total counts at 4000 Å as a function of visual magnitude. The size of the symbols is proportional to the exposure time for each object, in seconds.

24 GeminiFocus January 2020 / 2019 Year in Review Figure 3. Example spectra for 25 stars with [Fe/H] < -2.5 observed with Gemini/GMOS (North and South). The shaded areas highlight regions of interest for the determination of metallicity (blue - [Fe/H], Ca II K absorption feature), carbon abundance (red - A(C), CH G-band), and temperature (green - (K) Teff hydrogen Balmer lines). The spectra are ordered by increasing values of [Fe/H].

was around 72%, meaning that only 28% of (A(C)). Figure 3 shows the GMOS spectra of the already poor weather was lost, which is a 25 stars with [Fe/H] < -2.5 observed under great accomplishment for the program and poor weather conditions. The shaded areas the Observatory. highlight absorption spectral features used to determine [Fe/H] (Ca II K absorption fea-

ture), A(C) (CH G-band), and Teff (hydrogen Scientific Gain from Balmer lines). The values for each parameter Weather Loss are also listed. From the 666 stars, metal- The spectra gathered at Gemini/GMOS are of licities could be determined for 656 (98%), sufficient quality (signal-to-noise ratios and including 477 stars with [Fe/H] < -1.0 (73%), spectral resolution) to allow for the deter- 285 stars with [Fe/H] < -2.0 (43%), and 9 stars mination of stellar atmospheric parameters: with [Fe/H] < -3.0 (including one at [Fe/H] = -3.65). Carbon abundances were determined effective temperature (Teff), surface gravity, metallicity ([Fe/H]), and carbon abundances for 653 stars.

January 2020 / 2019 Year in Review GeminiFocus 25 Alliance" (RPA) — a multi-stage, multi-year effort to provide observational, theoretical, and experimental constraints on the nature and origin of the astrophysical r-process (rapid neutron-capture).

The parameters determined using the Gem- ini spectra are extremely useful to tailor target lists for the type of (high-resolution) follow-up conducted by the RPA, and there is already a study published based on an extremely metal-poor star first identified at Gemini (Cain et al., 2018). This star, J2005- 3057, shows enhancements in elements formed by the r-process, such as europium, iridium and thorium, among others. An- other effort currently underway is gathering high-resolution data for the most carbon- Figure 4. The distribution of the carbon abundances enhanced stars identified by Gemini and the Carbon abundances, as a function of the metallicity for these results are also promising. Collectively, these A(C), as a function of the stars is shown in Figure 4. The lower and side discoveries help us paint a more cohesive metallicity, [Fe/H], for the panels show marginal distributions for each picture of how the Universe evolved chemi- program stars observed quantity. The behavior is similar to that ex- with Gemini. The side and cally and how we can reshape our current pected from high-resolution spectroscopic lower panels show the understanding of stellar evolution and gal- marginal distributions for samples, which makes this subset impor- axy formation. In the near future, such bright each quantity. tant for two reasons: 1) as a tool for target stars will be perfect targets for high-resolu- selection, and 2) to have an independent tion spectroscopic follow-up with GHOST, estimate of quantities, such as the fraction which will be a great asset in pushing these of carbon-enhanced metal-poor stars as a efforts forward. function of [Fe/H], which is a crucial obser- vational constraint to Galactic chemical evo- References lution models. Cain, M., Frebel, A., Gull, M., et al., The Astro- physical Journal, 864: 43, 2018 What Have We Learned and Placco, V. M., Beers, T. C., Santucci, R. M., et al., What's Next? The Astrophysical Journal, 155: 256, 2018 Placco, V. M., Santucci, R. M., Beers, T. C., et al., The objectives of such follow-up studies, The Astrophysical Journal, 870: 122, 2019 which can include Gemini Poor Weather ob- Schlaufman, K. C., and Casey, A. R., The Astro- servations, are two-fold: 1) build statistics of physical Journal, 797: 13, 2014 metallicities and carbon abundances deter- mined from medium-resolution spectros- Steinmetz, M., Zwitter, T., Siebert, A., et al., The copy, which are crucial for studies of stellar Astrophysical Journal, 132: 1645, 2006 populations and formation of the Milky Way, and 2) select interesting stars for further, Vinicius Placco is Research Assistant Pro- fessor at the Department of Physics at the more targeted, high-resolution spectrosco- University of Notre Dame and is located at py efforts. One effort that is feeding directly Notre Dame, Indiana. He can be reached at: from the Gemini data is called the "R-Process [email protected] 26 GeminiFocus January 2020 / 2019 Year in Review John Blakeslee

Science Highlights Figure 1. Gemini North GMOS color Recapping some of the most recent and significant research composite image of Comet 2I/Borisov, produced from results achieved by the Gemini user community. data obtained in the g, r, and i filters on the night of November 11-12, 2019. Credit: Gemini Observatory/ NSF’s National Optical- JANUARY 2020 Infrared Astronomy Gemini Tracks Comet 2I/Borisov from North to South Research Laboratory/AURA Last quarter’s GeminiFocus reported on Director’s Discretionary Time (DDT) observations of interstellar Comet 2I/Borisov taken with the Gemini Multi-Object Spectrograph (GMOS) at Gemini North in early Sep- tember 2019, not long after it was discovered. In the ensuing months, the comet has traced a southward arc across the sky, and Gemini has been following its journey from both hemispheres. While diverse DDT programs were activated to study 2I/Borisov through October, more recent observations have been ob- tained via Fast Turnaround (FT) proposals and a 2019B Target of Opportunity program.

In one Gemini North FT program, Rosemary Pike (Aca- demia Sinica Institute of Astronomy and Astrophysics, Taiwan) and colleagues used GMOS and the Near-In- fraRed Imager and spectrometer (NIRI) to measure the optical and near-infrared (NIR) col- ors of the dust coma and tail for comparison with Solar System comets. Team member Meg Schwamb (Queen's University, Belfast) participated in the November observations via the “eavesdropping” option. Although most of the observations were taken with non-sidereal tracking, the observers also obtained a sequence of sidereally tracked exposures for pho- tometry of reference stars. These exposures were then used to make a color composite im- age, shown in Figure 1, that found its way into the pages of The New York Times.

January 2020 / 2019 Year in Review GeminiFocus 27 Figure 2. Near-infrared K-band image (right) of Comet 2I/Borisov taken on November 30, 2019, with FLAMINGOS-2 at Gemini South when the comet was 2.05 AU from the Sun, compared to an optical V-band image (left) taken a week earlier (when it was 2.12 AU from the Sun) at the Nordic Optical On November 13th, 2I/Borisov crossed into more than 10% of the coma cross-section. Telescope. Unlike in the the Southern Hemisphere, and the most Thus, the ice grains are likely confined to the optical image, the comet recent Gemini observations of it have been region of the nucleus. The study has been appears pointlike in K. made from Cerro Pachón. In a study pub- accepted for publication in Astronomy & As- Image reproduced from lished in the Research Notes of the Ameri- trophysics Letters, and a preprint is available Lee et al., Research can Astronomical Society, Chien-Hsiu Lee online. Notes of the American (NSF’s National Optical-Infrared Astronomy Astronomical Society, The GNIRS observations were taken on Sep- Research Laboratory) and collaborators ana- 3:184, 2019 tember 24th when 2I/Borisov was still 2.6 lyze 2.2-micron (μm) K-band images of the Astronomical Units (AU) from the Sun. It will comet obtained at Gemini South with FLA- be interesting to see how the spectrum has MINGOS-2 in late November. As shown in evolved as the comet reached its perihelion Figure 2, the comet appears point-like at 2.2 distance of 2.0 AU in December and began μm, unlike at optical wavelengths where the its long journey back to interstellar space. appearance is dominated by the extended The observations continue, and we are sure coma. Assuming that the K-band light is re- to see more highlights from this first inter- flected directly by the nucleus, and adopt- stellar comet before it’s gone for good. ing an albedo of 7% at this wavelength, the study derives an equivalent radius of 1.5 ki- lometers (km), similar to previous estimates. GPI Imaging of Debris Disks in A higher albedo would translate into a more Scorpius-Centaurus diminutive nucleus. The Gemini Planet Imager (GPI) has been Gemini has also observed 2I/Borisov spec- cranking out the results from Gemini South troscopically, in both the optical and NIR. A for the past six years, including a demo- study led by Bin Yang (European Southern graphical analysis, published last year in Observatory) used NIR spectra from the The Astronomical Journal, of large exoplan- Gemini Near-InfraRed Spectrometer (GNIRS) ets and brown dwarf companions from the at Gemini North, as well as from NASA’s In- first 300 stars observed in the GPI Exoplanet frared Telescope Facility, to search for diag- Survey (GPIES). The GPIES program also in- nostic absorption features of water ice. The cluded a disk campaign, with the goal of data show a moderately red, featureless discovering debris disks around young stars spectrum in the NIR similar to D-type aster- and characterizing the structure present in oids, 1I/‘Oumuamua, and many Solar System spatially resolved scattered-light images. In comets. No water ice absorption features a study recently accepted for publication in were detected, and spectral modeling indi- The Astronomical Journal, the GPIES team cated that large ice grains must comprise no

28 GeminiFocus January 2020 / 2019 Year in Review presents the first resolved images of debris solved debris disks; the disks are misaligned disks around four members of the Scorpius- by about 60 degrees. Depending on the or- Centaurus (Sco-Cen) association. bital eccentricity, it is possible that the mor- phological irregularities seen in both debris Sco-Cen is the nearest OB association to the disks could result from external dynamical Sun, with member distances ranging from perturbations of the other star in the system. about 110 to 140 parsecs and ages of 10-16 The large separation prevents an estimation million years. It is a particularly useful labo- of either the inclination or eccentricity of the ratory for studying debris disks, as the infra- binary orbit. red excess observed in young massive stars tends to be greatest around this age. Three The new results contribute to the census of of the disks newly imaged with GPI appear disks and the panoply of disk structures ob- symmetric in morphology and brightness served around hot young stars at this criti- distributions, but vary in inclination and ra- cal stage in the development of planetary dial extent. systems. A total of 15 stars in the Sco-Cen association now have debris disks that have The disk around the fourth star, HD 98363, been resolved in scattered light, and at least shows significant asymmetry that could in- seven of these show evidence for asymme- dicate the presence of a sizable planet. How- try. Figure 3 displays a gallery of images of ever, HD 98363 also has a wide co-moving scattered light disks and giant planets in the stellar companion, separated by 7,000 AU, association. The rich diversity of debris disks that has its own debris disk at a different seen around stars within a single young en- inclination and with differing morphologi- vironment is remarkable, and we can expect cal peculiarities. This makes HD 98363 A/B even more results to emerge from GPIES and the first binary system with two spatially re- its follow-up programs in the near future. Figure 3. Map of the Scorpius- Centaurus association in Galactic coordinates with stars having resolved scattered-light disks and imaged giant planets indicated. Proper motions are represented by the vectors. Green points represent previ- ously resolved debris disks, while gold points are the four new sys- tems. Red points indicate stars with imaged giant planets; HD 106906 has both a resolved debris disk and an imaged planet. The majority of the images are from GPI. Credit: Hom et al., The Astronomical Journal, in press

January 2020 / 2019 Year in Review GeminiFocus 29 the intracluster medium. In particu- lar, hundreds of cluster candidates have been identified in this way by the South Pole Telescope (SPT), a 10-meter radio dish located at the South Pole, designed for large-area surveys at millimeter and submilli- meter wavelengths. Because the SZ signal does not provide the redshift, additional observations of the mem- ber galaxies are required.

The SPT-GMOS Survey, led by Mat- thew Bayliss at Harvard (now at MIT), Figure 4. The study is led by Justin Hom of Arizona used the GMOS instrument at Gemini South Color composite image State University, and a preprint is available to measure the redshifts of SZ-selected clus- of the merging cluster online. SPT-CL J0356−5337 at z ter candidates identified by SPT. The survey = 1.036, made by com- measured redshifts for nearly 1,600 member bining Gemini/GMOS- Strong Lensing by Colliding galaxies in 62 SPT clusters, including several South g and i images with strong lensing features. The cluster SPT- with Hubble/ACS F606W. Clusters at High Redshift CL J0356–5337 (or SPT-0356) at z = 1.036, The yellow ellipses mark Clusters of galaxies, the largest self-grav- for which Bayliss and collaborators spectro- cluster members; several itating structures in the Universe, form strongly lensed arcs are scopically confirmed eight members, was via hierarchical assembly, increasing their visible near the center of among the highest-redshift strong lensing the field. masses through the accretion of individual clusters in the sample. Credit: Mahler et al., galaxies and small groups, often funneled arXiv:1910.14006 inward along cosmic filaments. Occasional- In a new study, Guillaume Mahler of the Uni- ly, two massive clusters coalesce, providing versity of Michigan and collaborators pres- an opportunity to study high-speed galaxy ent a strong lensing analysis of SPT-0356 and interactions and shock physics within the expand the sample of likely cluster members colliding intercluster media, the dominant using single-band F606W Hubble Advanced baryonic component in such clusters. If the Camera for Surveys (ACS) imaging combined timing and geometry are favorable, and if with Gemini/GMOS-South g- and i-band each cluster is massive enough to produce imaging. Figure 4 shows a color composite detectable gravitational lensing of back- made from the Gemini and Hubble data, with ground sources, then the event also affords yellow ellipses enclosing galaxies lying on a rare opportunity to constrain the physical the cluster red sequence; the largest ellipse properties of the nonbaryonic cluster dark marks the brightest cluster galaxy (BCG). The matter. Examples of such collisions include red sequence selection is based on the color- the “Bullet Cluster” at redshift z = 0.30 and “El magnitude diagram shown in Figure 5, made Gordo” at z = 0.87. from a combination of Gemini and Hubble photometry. To enable the lensing analysis, Large numbers of distant clusters have now the team used Magellan Observatory to ob- been found via the Sunyaev-Zel’dovich (SZ) tain redshifts of three multiply-imaged back- effect, the apparent decrement in bright- ground galaxies, lensed into the arcs visible ness of the cosmic microwave background near the center of Figure 4, about 9 to 15 arc- (CMB) radiation resulting from the scattering seconds west of the BCG. of CMB photons by high-energy electrons in

30 GeminiFocus January 2020 / 2019 Year in Review The team’s strong lens modeling indicates that SPT-0356 has a two- component mass distribution, with one component centered on the BCG and the other centered on a tight clump of eight galaxies located about 22 arcseconds (170 kilopar- secs) west of the BCG. The two com- ponents have similar masses, with a 3:2 mass ratio being within the range implied by the analysis, although the galaxy distributions appear very dif- ferent. Moreover, the difference in their mean line-of-sight velocities is ejected the object long ago from its home Figure 5. only about 300 km/s, suggesting that most planetary system. Color-magnitude of the relative motion is in the plane of the diagram of galaxies sky. Thus, SPT-0356 appears to be a face-on Now, less than two years later, a second in- in the field of SPT-CL major merger at z > 1, reminiscent of the Bul- terstellar emissary has arrived from the di- J0356−5337 made let Cluster at much lower redshift. However, rection of Cassiopeia, and it bears strikingly from Gemini GMOS- additional data, including deep X-ray obser- little resemblance to the first. If the stars are South and Hubble/ACS vations and more galaxy redshifts to supple- trying to tell us something, their message data. Galaxies selected as being on the red ment those supplied by GMOS, are needed is inconsistent. The new object was discov- sequence are marked ered by the Crimean amateur astronomer to fully characterize this complex system. with red stars; filled Gennady Borisov on August 30, 2019, using The study has been submitted to The Astro- symbols indicate galaxies a 65-centimeter telescope that he built him- within 76 arcseconds physical Journal, and a preprint is available self. Subsequent observations have shown (about 600 kiloparsecs) online. that its orbital eccentricity with respect to of the brightest cluster the Sun exceeds 3.3 (eccentricities above 1.0 galaxy. Spectroscopically confirmed members correspond to unbound hyperbolic orbits; OCTOBER 2019 are indicated by gold ‘Oumuamua had an eccentricity of 1.20). Comet 2I/Borisov Breezes squares. Popularly known as “Comet Borisov” (even Through Solar System, Credit: Mahler et al., though the amateur has discovered seven arXiv:1910.14006 Tail Streaming Behind other, more conventional, comets), the ob- It was in October 2017, just days after this ject received the official interstellar designa- writer joined Gemini, that the first interstel- tion 2I/Borisov from the International Astro- lar object, later designated 1I/‘Oumuamua, nomical Union on September 24, 2019. was spotted making its expeditious escape Gemini Observatory was first alerted to 2I/ from our Solar System. Observations by Borisov by a Director’s Discretionary Time Gemini and many other observatories dem- (DDT) proposal received on the evening of onstrated that ‘Oumuamua was surprisingly September 9th, when the object was in the asteroidal in nature, with no apparent coma northern sky at a distance of 3.4 AU from or tail. Moreover, judging from the dramatic from the Earth and within 43 degrees of the variations in its light curve, this first interstel- Sun. Following careful review, the proposal lar visitor had an unusually large axis ratio, was found to be compelling, with Gemini's perhaps 10:1, suggesting that it may be a large aperture being well suited for inves- scattered shard from a violent collision that tigating possible cometary activity during

January 2020 / 2019 Year in Review GeminiFocus 31 Figure 6. The observational study GMOS-North g,r of this second interstellar composite color image interloper has only just of the interstellar comet begun. Additional Gem- 2I/Borisov, obtained in ini observations have morning twilight on already been obtained, September 10, 2019, and more are currently at a mean elevation of less than 30° from the scheduled in the queue. eastern horizon. The 2I/Borisov is entering alternating red-blue the Solar System from streaks are background “above,” and its visibility stars that appear trailed will gradually improve because the telescope as it crosses the celestial was tracking the comet, which was moving non- equator in mid-Novem- sidereally at a rate of 75 ber and moves towards a arcseconds per hour. The perihelion distance of 2.0 comet was 3.4 AU from AU, near the inner edge the Earth at the time of of the Asteroid Belt, on these observations. December 8th. It reaches a minimum distance of the brief visibility window at the end of the 1.9 AU from the Earth in late December, and night. Consequently, multi-band imaging will continue to be visible from the South- observations with the Gemini Multi-Object ern Hemisphere for much of next year. Thus, Spectrograph (GMOS) on Gemini North Gemini’s access to the entire sky will enable were obtained during morning twilight, less detailed study of 2I/Borisov throughout the than 12 hours after the proposal was sub- entire course of its visit — we are sure to mitted. Figure 6 displays the resulting com- have more highlights on this first interstel- posite color image; an extended coma and lar comet before it leaves our corner of the cometary tail are clearly visible. This makes Galaxy forever. 2I/Borisov the first known interstellar comet.

A study based on these Gemini North ob- servations, with supplementary data from Probing for Patterns in Io's the William Herschel Telescope on La Palma, Volcanoes Using Adaptive finds that 2I/Borisov appears quite similar Optics to typical Solar System comets in terms of photometric color and its derived dust par- Ever since the Voyager spacecrafts revealed ticle properties. Graduate student Piotr Guzik the rampant volcanism on Jupiter’s inner- of Jagiellonian University in Poland led the most large moon Io, planetary scientists study, which is currently in press at Nature have been puzzling over the variations in Astronomy (a preprint is available online). the timing and intensities of the splotchy The object’s g-r color is only slightly redder satellite’s many eruptions. Intense tidal heat- than average for comets, and the estimated ing, the stretching and squeezing of Io’s diameter of 2 km for the comet’s nucelus, crust as it follows its 1.8-day elliptical orbit while highly uncertain, is well within the nor- around the giant planet, supplies the energy mal cometary range. In light of ‘Oumuamua’s to melt interior silicates and produce mag- anomalies, the apparent banality of 2I/Bor- ma, which eventually erupts to the surface. isov is in itself remarkable. However, the variations in the volcanic activ- 32 GeminiFocus January 2020 / 2019 Year in Review ity generally occur on longer timescales, un- in the near-infrared varies by more than an Figure 7. correlated with the orbital period. This con- order of magnitude. This large data set en- AO-assisted near-infra- trasts with the case for other tidally heated abled the team to uncover surprising pat- red images taken with moons such as Saturn’s Enceladus, for which terns in Io’s volcanic activity. For instance, of NIRI on Gemini North the degree of activity varies predictably with the 18 sites with the brightest eruptions, 16 of Jupiter’s moon Io, showing the eruption its distance from the planet. Although Io and are on the trailing hemisphere with respect of Isum Patera in May- Enceladus have very similar orbital eccen- to Io’s orbital motion. This tendency remains June 2018. Isum Patera tricities and periods, a key difference is the unexplained; the likelihood of it occurring is the only bright spot viscosity of the erupting fluid, which is water from a random spatial distribution is much visible in these Kc (2.27 on Enceladus and magma for Io. less than 1%. μm) images; it is seen at the corresponding loca- To understand what drives the variations in In a companion paper published in Geo- tions in the L’ (3.78 μm) the volcanism on Io, a team of astronomers physical Research Letters, de Kleer and images. The bright spot led by Katherine de Kleer of the California colleagues show that the roughly 500-day south of Isum Patera in Institute of Technology has analyzed the variations in the intensity of Loki Patera’s the L’ images is Marduk Fluctus. most detailed data set on the moon’s volca- activity may be related to periodic changes Figure reproduced from nic activity to date. The observations were in the shape of the moon’s orbit. Regular de Kleer et al., The Astro- collected on 271 nights between August gravitational perturbations from Europa nomical Journal, 158: 2013 and July 2018 using the Near InfraRed and Ganymede, which respectively have 2:1 29, 2019. Imager and spectrometer (NIRI) on Gemini and 4:1 orbital resonances with Io, prevent North with the ALTAIR adaptive optics sys- the inner moon’s orbit from circularizing. tem in natural guide star (NGS) mode and Instead, Io’s eccentricity and semimajor axis the Near InfraRed Camera 2 (NIRC2) on the vary cyclically with periods of 480 and 460 Keck II telescope, also using NGS adaptive days, respectively. This evolution in Io’s orbit optics. The Gemini/NIRI data comprise 80% is consistent with the timescale of the quasi- of the total visits; example NIRI images are periodic behavior of Loki Patera. shown in Figure 7. The study has been pub- At first, this link between orbital evolution lished in The Astronomical Journal and fea- and volcanic activity may seem surprising, tured in The New York Times. since the range in the tidal stresses over a In total, the team has detected at least 75 single orbit is larger than the variation in the unique hot spots of volcanic activity. The mean tides resulting from the change in or- most active volcano, known as Loki Patera, bital shape. However, the researchers note was detected 113 times during the five-year that while magma is likely too viscous to campaign, essentially every time it was visi- change its flow significantly on the timescale ble. Three other hot spots were each detect- of one orbit, it can adjust its flow over the ed at least 80 times. Loki Patera appears to longer period associated with the change be erupting continuously, but its brightness in Io’s orbital shape. If there is a connection, January 2020 / 2019 Year in Review GeminiFocus 33 Figure 8. the peak in activity should coincide with the spective. GR predicts that luminous objects Panel A: Astrometric time of maximum orbital eccentricity, and in strong gravitational fields should exhibit measurements the star the data confirm that this is indeed the case. relativistic redshifts. This means that a star S0-2 over its 16-year Higher cadence observations are needed to moving towards us in the vicinity of a black orbit of the supermas- test this hypothesis and rule out shorter pe- hole should appear to have a smaller blue- sive black hole at the riod drivers of Loki Patera's variability. shift, and one moving away from us should center of the Milky Way, compared with the best- have a larger redshift, than would be the fitting projected General case if the law of Newtonian gravity pre- Relativistic orbit model. Three Maunakea Observatories vailed. In the most stringent test of this pre- The black hole is located Track Relativistic Star around a diction to date, the team analyzed over two at the origin of the coor- Black Hole decades of astrometric and spectroscopic dinate system, and the data, obtained using adaptive optics, on a dashed line shows the If Einstein were alive today, he might be one star known as S0-2 as it followed its eccen- intersection of the orbital of the few people tired of actually winning. tric 16-year orbit around Sagittarius A* (Sag plane with the plane of Setting aside his long quarrel with quantum the sky. The black points A*), the supermassive black hole at the cen- mechanics and all that business about a uni- represent new observa- ter of our Galaxy. Figure 8 shows the full set fied field theory, his formulation of General tions from 2017-2018, of positional and velocity data. while the gray points are Relativity (GR) has proven to be one of the earlier measurements. most successful descriptions of nature ever The star reached its closest approach to Sag Panel B: radial velocity proposed. From the deflection of starlight in A* in May 2018, when it was at a distance of measurements over the 1919 to the detection of gravitational waves only 120 AU and moving at 2.7% of the speed period 2000-2018 and the in 2015, Einstein’s General Relativity has tri- of light. During the critical months surround- best-fitting model (col- ored curve). Open, gray, umphed over every observational test to ing pericenter passage, the team used three and black circles repre- date. Now a team of researchers led by An- different spectroscopic instruments at three sent previous, rederived, drea Ghez at the University of California Los different observatories, including the Near- and new measurements, Angeles has tested GR in a new regime, the infrared Integral Field Spectrometer (NIFS) respectively. Panel C: strong gravitational field near a supermassive on Gemini North, the OH-Suppressing Infra- residuals from the best- black hole. The result: chalk up another one Red Imaging Spectrograph (OSIRIS) on the fitting velocity model. for the iconic physicist. Keck II telescope, and the Infrared Camera Figure adapted from Do and Spectrograph (IRCS) on the Subaru tele- et al., Science, 365: 664, Although simple conceptually, the test was scope. "The velocity of the star was chang- 2019. incredibly exacting from a technical per- ing quickly every night! So having all three observatories participate was es- sential," said Tuan Do (also of UCLA), the lead author of the study. Combin- ing data from mul- tiple instruments also allowed the team to carefully check for instru- mental biases.

34 GeminiFocus January 2020 / 2019 Year in Review As shown in Figure 9, GR provides an ac- curate description of the star’s positional and velocity data throughout its very large swing in velocity near its closest approach to Sag A*. In contrast, the observations rule out Newton’s law of gravity with a high sta- tistical significance. “The GR model is 43,000 times more likely than the Newtonian mod- el in explaining the observations,” the study concludes. The measurements also provide strong constraints on the black hole’s dis- tance and mass, 8.0 kiloparsecs and 4.0 mil- lion solar masses, respectively.

Of course, no one wins forever, and at some point, namely the event horizon of a black hole, GR must also fail. However, although S0-2 plunged precipitously near Sag A*, the not entirely dissimilar. Like the perfect guitar Figure 9. minimum distance was roughly 1,000 times riff, reverberation mapping requires precise Top: Zoom in on the radial velocity data from larger than the radius of the event horizon. timing and can be quite challenging to ex- 2018, encompassing the Thus, it may be some time before observa- ecute in practice. In addition, the virtue of both lies in their conceptual simplicity. maximum and minimum tional limits encroach on the limits of GR’s of the observed radial validity. Meanwhile, such observations con- Reverberation mapping works by applying velocity. Measurements tinue to enlighten our understanding of the the familiar virial theorem to the broad line from the three different observatories are dynamics and evolution of the center of our region (BLR) of an AGN. Assuming that the Galaxy. The study appears in the journal indicated; Gemini/ motion of the gas in the BLR is primarily influ- NIFS and Keck/OSIRIS Science. enced by the central black hole, the mass of each provided nine 2 the black hole MBH will be proportional to σ R, measurements during where σ is the velocity dispersion determined this critical period, over JULY 2019 from the Doppler width of a broad emission which the observed line and R is the characteristic radius of the velocity changed by 6,000 km/s. Bottom: Reverberations from an BLR. The radius is determined from the delay radial velocity residuals Intermediate-mass Black Hole time τ between variations in the intensity of with respect to the in a Bulgeless Dwarf the continuum light from the AGN, which best-fitting General excites the gas within the BLR, and the line Relativistic model. For some, the term “reverberation mapping” emission itself: R = cτ, where c is the speed Figure from Do et al., might suggest the idea of pinpointing the of light. Because lines of different ionization Science, 365: 664, 2019. locations of the various garage bands in show different delays, the same line should the neighborhood (all with their amplifiers be used for determining both σ and τ. Typical turned way up) based on the distribution AGNs powered by supermassive black holes and intensity of the vibrations emanating of millions of solar masses (MB) have delay from one’s walls and window panes. But in times measured from Balmer lines ranging actuality, it denotes a powerful technique from a few days to many months. for determining the masses of the black holes embedded within the active galactic A new study published in Nature Astron- nuclei (AGNs) at the centers of many galax- omy has measured the mass of the black ies. Interestingly, the two phenomena are hole associated with one of the lowest lu-

January 2020 / 2019 Year in Review GeminiFocus 35 Figure 10. GMOS spectrum of the AGN in the low-mass spiral NGC 4395, showing the narrow [NII] and Hα lines superposed on the broad Hα emission used for the reverberation measurement (left). The narrow [SII] lines at longer wavelength were used as proxies for the central stellar dispersion (right). Figure reproduced from Woo et al., Nature Astronomy, 2019, in press (arXiv 1905.00145).

Figure 11. The new NGC 4395 black hole measurement is plotted in the context of the relation between central black hole mass minosity AGNs known. The AGN resides minutes based on a combination of broad- and stellar velocity within a nuclear star cluster at the center of and narrow-band imaging collected at sev- dispersion for more massive systems. The the nearby dwarf spiral NGC 4395, and the eral small telescopes, the implied black hole stellar velocity dispersion study was led by Jong-Hak Woo of Seoul Na- mass is about 9,100 MB. Previous estimates for NGC 4395 is shown tional University. Using spectroscopic data ranged from 5 to 40 times higher, but were as a previously published from the Gemini Multi-Object Spectrograph much more poorly constrained. The new re- upper limit (open red (GMOS) at Gemini North, Woo’s team mea- sult is securely within the realm of the elu- square) and the proxy sured a line-of-sight velocity dispersion of sive “intermediate-mass” black holes, which value adopted from the width of the [SII] 426 kilometers per second (km/s) from the may be the seeds from which supermassive emission line (solid red width of the broad Hα line (Figure 10). Com- black holes grow. square). Plotted values bined with a reverberation time delay of 83 There are well established relations for mas- for the higher mass galaxies are sive galaxies between central stellar dynamical black hole mass and the prop- measurements in erties of the stellar bulge; it inactive galaxies is interesting to ask how NGC (open black circles) 4395, a pure disk galaxy without and reverberation any bulge, fits into these. The mapping in active new study estimated the cen- galaxies (filled blue circles). The solid tral stellar velocity dispersion

and dashed lines σ★ from the width of the nar- are, respectively, row [SII] emission line, finding fits to the σ★ ≈ 18 km/s, consistent with a combined high- previous upper limit. Using this mass sample and value, they place NGC 4395 on to the dynamical the diagram of M versus ve- measurements BH only. locity dispersion for high-mass Figure reproduced galaxies (Figure 11), conclud- from Woo et ing it is broadly consistent with al., Nature a simple extrapolation to lower Astronomy, 2019, masses. This suggests that the in press (arXiv observed relations between M 1905.00145). BH

36 GeminiFocus January 2020 / 2019 Year in Review and central dispersion does not originate GPIES is sensitive to young, self-luminous Figure 12. from the process of hierarchical growth, but planets with masses above about 2 Jupiter GPIES sensitivity contours that the galaxy mergers that produce central masses and orbital semi-major axes from 3 for companion mass (in bulges preserve a relation that may already to 100 AU. The detections thus far include units of Jupiter masses) be present for the seed intermediate-mass six giant planets and three brown dwarfs. and orbital semi-major axis (Astronomical Units) black holes. Although only about 40% of the stars in- for planetary (left) and cluded in the analysis have masses greater Testing this scenario will require more stud- brown dwarf (right) com- than 1.5 MB, all of the detected planets orbit panions. The six giant ies of the incidence and masses of black stars above this mass. This is even more strik- planets and three brown holes in the centers of low-mass galaxies. In ing because it would be easier to see such dwarfs detected in the addition, such studies can determine wheth- planets orbiting fainter, lower mass stars. survey are overlaid on er the familiar supermassive black holes like- the contours. Although While there have been previous indications ly originated from “light” seeds of order 100 the majority of these of a correlation with stellar mass, the GPIES to 1,000 M (possible remnants of massive companions were not B results confirm to better than 99.9% confi- Population III stars) or “heavy” seeds of order discovered by GPIES, dence that high-mass stars are more likely their host stars were part 104 M or more (formed via the direct col- B to host planets within the explored range of of the unbiased sample lapse of giant gas clouds). As demonstrated planetary masses and orbital separations. and were not selected by the impressive results on NGC 4395, rever- because of the pres- beration mapping remains the most promis- Accounting for the detection sensitiv- ence of the companions; ing method for building up the required data ity curves and combining their results with thus, the detections are samples to address these questions. those from radial velocity studies (sensitive included in the statistical analysis. The curves indi- to companions at smaller radii), the team cate the numbers of stars concluded that the most likely location for Divergent Demographics of in the sample for which giant planets to occur is between 1 and 10 the sensitivity allowed Planets and Brown Dwarfs in AU from their host stars. The occurrence rate detection of compan- the GPI Exoplanet Survey drops steeply at larger separations. The num- ions with the plotted ber of giant planets also declines significant- combinations of param- Soon after the Gemini Planet Imager (GPI) was ly with increasing planetary mass. eters; very few stars had commissioned at Gemini South, the interna- sensitivity sufficient to tional team behind the instrument embarked Although brown dwarfs are often consid- detect planets of masses

on a major systematic survey for substellar ered transitional objects between planets < 3 MJup , but two were companions and protoplanetary disks around and stars, they appear to have quite different detected. Figure reproduced from the youngest, closest stars in the southern demographics than giant planets, as shown Nielsen et al., The Astro- in Figure 12. The study concludes that only sky. Earlier this year, the GPI Exoplanet Survey nomical Journal, 158: (GPIES) observed its 531st target star, bringing about one in ten stars hosts a brown dwarf 13, 2019. the main survey to a close after more than four years, although follow-up observations of promising candidates have continued. Now, the team has published preliminary results from a statistical analysis of the first 300 stars surveyed. The study, published in the July issue of The Astronomical Journal, was led by Eric Nielsen of Stanford Uni- versity and represents the largest direct imaging survey for giant planets pub- lished to date.

January 2020 / 2019 Year in Review GeminiFocus 37 companion at separations of 10 to 100 AU. powerful probe of a galaxy’s dynamical struc- This is a factor of ten below the inferred oc- ture is integral field spectroscopy (IFS). Wide- currence rate of giant planets around high- field IFS studies provide insight into global mass stars. Moreover, although the numbers dynamics and past interactions, while IFS are low, the distributions in both mass and data on the innermost regions can constrain semi-major axis are consistent with being the central supermassive black hole (SMBH) flat for brown dwarfs, in contrast with the mass and the shapes of the stellar orbits in falling distributions for giant planets. In ad- the vicinity of its sphere of influence. dition, the detected brown dwarfs all orbit The MASSIVE Galaxy Survey is systemati- stars with masses below 1.5 M , again unlike B cally targeting all early-type galaxies in the the giant planets. northern hemisphere with stellar masses 11 Based on these results, earlier suggestions greater than 3 × 10 MB within a distance that wide-separation giant planets and of about 100 megaparsecs for detailed ki- brown dwarfs may comprise a single under- nematic and photometric analysis. The lat- lying population is unlikely to be correct. est work in the MASSIVE series presents the The divergent trends strongly indicate dis- first results from the high angular resolution parate formation mechanisms. Specifically, portion of the survey, based on deep GMOS- the study concludes that giant planets likely North IFS observations of 20 galaxies. These form “bottom up” through the process of are combined with wide-field IFS data from core accretion while brown dwarfs form “top the Mitchell spectrograph at McDonald Ob- down” like stars via gravitational instabil- servatory to obtain detailed kinematic maps ity. More data are needed to confirm these spanning more than two orders of magni- trends; fortunately, there are another 231 tude in galactocentric radius. The new study stars from the rest of the GPIES survey await- appears in the June issue of The Astrophysical ing final analysis and publication. Journal and is led by graduate student Irina Ene of the University of California, Berkeley. Spatially Resolved Kinematics Figure 13 (next page) shows example maps of the first four moments (v, σ, h , and h ) of of 20 MASSIVE Ellipticals 3 4 the stellar velocity distributions from the Every galaxy has its own story, and every gal- high-quality GMOS IFS data for two galax- axy has been many others in the past (un- ies in the survey. The maps cover the cen- like in the human parallel, this is not purely tral 5 × 7 arcseconds. The figure also shows metaphorical, as galaxies grow via hierarchi- the one-dimensional distributions of these cal assembly). Generally speaking, the most parameters combined with the wider field massive galaxies have led the most interest- IFS measurements. Although both galaxies ing lives. These often reside in dense envi- exhibit strong central rotation, they have rons that have exposed them to frequent strikingly different kinematic profiles. In fact, interactions with assorted neighbors, influ- most of the galaxies in the MASSIVE sample encing in complex ways the coevolution of show only slow rotation (unlike most previ- their component stars, gas, dark matter, and ous IFS studies of early-type galaxies, which supermassive black holes. were weighted towards lower luminosity). Although the detailed formation histories of Interestingly, in galaxies that do rotate, the most galaxies will remain forever uncertain, central rotation is often unaligned with the the key thematic elements may be surmised large-scale kinematics, indicating diverse through a variety of methods. A particularly merger histories.

38 GeminiFocus January 2020 / 2019 Year in Review Figure 13. Example distributions of the first four kinemati-

cal moments (v, σ, h3

and h4 ) measured from the GMOS-N IFS data for two of the MASSIVE survey galaxies. For each galaxy, the top row shows two-dimensional maps, while the bottom row shows two-sided radial profiles from GMOS (magenta circles) and Mitchell (green squares) data. The verti- cal dotted lines mark radii of ± 0.2 arcsecond. Figure reproduced from Ene et al., The Astro- physical Journal, 878: 57, 2019.

The kinematic diversity across the full sample from the higher order moments, particularly is illustrated in Figure 14 (next page), which the kurtosis h4 , can determine the relative shows the velocity dispersion profiles for all importance of these two effects. For this 20 galaxies. Although most of the galaxies purpose, high spatial resolution for resolv- have centrally rising dispersions, the slopes ing stellar kinematics within the sphere of vary greatly, and in some cases change sign influence of the SMBH is essential. with radius. A sharply rising central disper- As a proof of concept, the new study per- sion may indicate the presence of a SMBH forms detailed dynamical modeling of the but can also reflect increasing radial anisot- combined GMOS and Mitchell IFS data sets ropy in the stellar velocities. Information for NGC 1453, the most regular fast-rotating January 2020 / 2019 Year in Review GeminiFocus 39 Figure 14. Velocity dispersion profiles for 20 galaxies in the MASSIVE survey observed at Gemini North with the GMOS-N integral field unit (magenta) combined with wide-field measure- ments from the Mitchell spectrograph at McDon- ald Observatory (green). The diversity in the dis- persion profiles among these high-mass early- type galaxies is evident. The blue lines show the best-fit power laws to the GMOS data. Figure reproduced from Ene et al., The Astro- physical Journal, 878: 57, 2019.

galaxy in the sample. In addition to con- the largest TNO has a maximum angular straining the stellar mass-to-light ratio and size of about 0.1 arcsecond; more typical circular velocity of the dark matter halo, the ones are unresolved at 0.01 arcsecond or analysis finds both a spatially varying veloc- smaller. Except for the two TNOs that have ity anisotropy and a central SMBH with an been visited by spacecraft, the most direct impressively large mass in excess of 3 × 109 measurements of TNO sizes come from stel-

MB. The MASSIVE Survey team, led by Berke- lar occultations. Consequently, planetary ley professor Chung-Pei Ma, is currently run- scientists exercise great vigilance in taking ning the detailed models for the full galaxy advantage of these rare opportunities. sample. The results will provide further in- One such opportunity occurred on March 7, sight into the assembly histories of the larg- 2017. Based on ground-based astrometry, it est galaxies in the local Universe and refine was thought that an occultation of a mag- our understanding of the coevolution of gal- nitude V = 14.6 star by the large TNO Orcus axies and their central black holes up to the would be viewable from parts of the Pacific most extreme masses. and the Americas on that date. With an esti- mated diameter in excess of 900 kilometers APRIL 2019 (km), Orcus likely meets the shape criteria for a dwarf planet. Like Pluto, it is in a 3:2 Vanth Surprises with Double orbital resonance with Neptune, has a semi- Dip During Occultation major axis of 39 AU, and a high eccentric- The sizes and surface compositions of ity. It has one large satellite named Vanth, trans-Neptunian objects (TNOs) are notori- which orbits with a period of 9.5 days. With ously difficult to study. As seen from Earth, the availability of astrometry from the Gaia 40 GeminiFocus January 2020 / 2019 Year in Review space mission, it became clear that Vanth, Figure 15. rather than Orcus, would be the one tracing Gemini South DSSI a path of occultation across the Earth’s sur- image of the star pair face on the predicted date. occulted by Vanth, a satellite of the large In anticipation of this event, an international trans-Neptunian team of occultation-chasers led by Amanda object Orcus. This Sickafoose of the South African Astronomi- image consists of 1,000 seconds of speckle data cal Observatory organized a monitoring combined to reveal the campaign with five telescopes located in binary pair responsible Hawai‘i, California, Texas, and Chile. To their for the observed double surprise, the coordinated observations de- occultation. The bright tected two non-simultaneous dips in the primary is at center, stellar brightness at two widely separated and the newly detected companion is at upper telescopes. The detections were made by strument team. The resulting image, shown right (approximately the NASA Infrared Telescope Facility on in Figure 15, reveals that the occultation star is indeed a double, with a separation of 250 2:00 position; the Maunakea, and the Las Cumbres 1-meter other “star” at the 8:00 milli-arcseconds and a brightness differen- telescope at the McDonald Observatory in position is an artifact Texas. The observations could not be ex- tial of about 0.9 magnitude in the red DSSI of the autocorrelation plained by a single object occulting a single bandpass. Figure 16 compares the original analysis used in speckle star; moreover, previous Hubble Space Tele- prediction for the single path of occultation processing). Figure reproduced from scope (HST) data ruled out the possibility of by Vanth with the paths of the two occulta- tions as reconstructed from the binary star Sickafoose et al., Icarus, another satellite of sufficient size to explain 319: 657, 2019. the second dip in stellar brightness. positions in the DSSI data. The reconstruc- tions fit perfectly with the observations. To test the occultation star for possible multi- plicity, the team applied for Fast Turnaround time with the visiting Differ- ential Speckle Survey In- strument (DSSI) at Gem- ini South. The proposal, led by Amanda Bosh of the Massachusetts Insti- tute of Technology, was successful, and the ob- servations were quickly processed by the DSSI in-

Figure 16. The dual paths of Vanth. Left: the predicted path of Vanth’s shadow during the occultation of March 7, 2017, based on Gaia DR1 astrometry. The locations of the telescopes participating in the occultation campaign are indicated by stars. The extent of the shadow is indicated for a physical diameter of 280 km; the shadow of Orcus is off the globe. Right: the actual shadow paths of Vanth as reconstructed using the positions of the two components of the double star determined from Gemini/DSSI imaging. The brighter star was occulted along the upper path, which passed over the observing location in Texas, but was not detected at the location in California. The occultation of the fainter star occurred along a path that passed over the observing location in Hawai‘i; no occultations were detected at the locations in Chile. The paths are drawn for a Vanth diameter of 442.5 km, the size determined from these observations. Figure reproduced from Sickafoose et al., Icarus, 319: 657, 2019. January 2020 / 2019 Year in Review GeminiFocus 41 Several decades ago it was proposed that a substantial fraction of the most distant qua- sars found in flux-limited surveys would be brightened above the survey limit by gravi- tational lensing. If this is the case, the result- ing “magnification bias” would cause a sys- tematic overestimation of the masses of the supermassive black hole population power- ing high-redshift quasars. However, no mul- tiply-imaged lensed systems had ever been found above redshift z = 4.8 (a lookback time of about 12.5 billion years), despite intensive high-resolution follow-up of hundreds of Figure 17. Once the binary nature of the occultation quasars known beyond this redshift. It may be that the extended appearance of multi- Gemini was one of star was revealed by Gemini/DSSI, the two several large telescopes observed occultations, combined with non- ply-lensed quasars, and/or color contamina- that contributed to detections at the other sites, allowed the tion by the lensing galaxy, causes a strong the study of the lensed team to place a tight constraint of 443 ± 10 selection bias against these systems. quasar J0439+1634, km on the diameter of Vanth. Remarkably, selected as a candidate Fan’s team selected J0439 + 1634 as a high- high-redshift object this is 60% larger than previous estimates, redshift quasar candidate based on a com- because it is an r-band and roughly half as large as the estimated bination of imaging data from the Pan- “dropout” with little i size of Orcus. The results also placed a limit STARRS1 survey in the optical, the UKIRT flux (top). The 6.5-m of a few microbars on any possible atmo- Hemisphere Survey in the near-infrared, and MMT and 10-m Keck-I sphere around Vanth. The study has been archival Wide-field Infrared Survey Explorer telescope obtained published in the journal Icarus, and a pre- optical spectra (outlined data in the mid-infrared. Follow-up optical in cyan), while the 8.1-m print is available online. spectroscopy with the 6.5-meter (m) Mul- Gemini North telescope tiple Mirror Telescope and 10-m Keck I tele- obtained an infrared The Mass of the Most Distant scope showed a prominent spectral break spectrum (outlined in consistent with a redshift near 6.5 (lookback red). The width of the Lensed Quasar time of 12.9 billion years). A near-infrared Mg II line near 2100 nm spectrum obtained with GNIRS at Gemini constrains the mass of Observations from the Gemini Near-Infra- the black hole powering Red Spectrograph (GNIRS) have confirmed North detected strong Mg II emission, yield- the quasar. The 2 x the redshift and constrained the mass of ing a firm redshift measurement ofz = 6.51. 8.4-m Large Binocular the brightest quasar yet discovered at red- Figure 17 shows the combined spectrum. Telescope captured an shift z > 5. However, the discovery paper From the width of the Mg II line in the GNIRS adaptive optics corrected led by Xiaohui Fan of the University of Ari- spectrum, the team derived a mass of almost image that suggests the 5 billion solar masses for the black hole pow- quasar is lensed, later zona concludes that the object, known as confirmed by HST. J0439+1634, is not the intrinsically most ering the quasar, and the photometric mea- Credit: Feige Wang luminous quasar at this redshift. Rather, its surements implied an astounding total lumi- 14 (UCSB), Xiaohui Fan apparent brightness has been boosted by nosity of 5.8 × 10 solar luminosities. (University of Arizona) a factor of about 50 by the gravitational However, imaging obtained with adap- magnification of an intervening galaxy. This tive optics on the 2 × 8.4-m Large Binocular makes J0439 + 1634 the most distant known Telescope indicated that J0439 + 1634 was strongly lensed quasar, and perhaps the broader than a point source, suggesting the first of many waiting to be revealed through presence of either a host galaxy or multiple high-resolution imaging.

42 GeminiFocus January 2020 / 2019 Year in Review images. Higher resolution imaging with HST clearly resolved the system into multiple lensed components with a maximum sepa- ration of about 0.2 arcsecond, plus an ex- tended source about 0.5 arcsecond away, in- terpreted as the lensing galaxy. Photometric analysis implied a redshift of about 0.7 and a mass of 6.3 billion solar masses for the lens- ing galaxy. Based on these measurements, the team derived a best-fit lensing model with three quasar images and a total mag- nification factor of 51.3. After correcting for the magnification, the inferred luminosity of J0439 + 1634 drops to “only” 1.1 × 1013 solar luminosities, and its black hole‘s mass be- comes a pedestrian 430 million solar masses. than those of the halo, a subclass of moder- Together these estimates imply an extremely Figure 18. ately metal-poor ([Fe/H] < −1.0), α-enhanced The Gemini GeMS+GSAOI high mass accretion rate, as required to grow ([α/Fe] > + 0.3), bulge globular clusters with J, K color composite image such a large black hole at early times. blue horizontal branches are thought to be of HP 1 (right) is shown within the context of a The results of this study indicate that many among the oldest stellar systems in the Gal- larger field imaged under strongly lensed, high-redshift quasars could axy. In this scenario, the moderate metallici- natural seeing conditions have been missed by past surveys because ties of these ancient star clusters result from by the Visible and Infra- standard color selection criteria will fail the early, rapid chemical enrichment of the red Survey Telescope for when the quasar light is contaminated by Milky Way’s innermost regions. Astronomy (VISTA, left). a lensing galaxy. Thus, changing the tech- One such candidate “fossil relic” of the niques for selecting quasars could signifi- bulge’s early formation is HP 1, a globular cantly increase the number of lensed quasar cluster just 3° away from the Galactic Cen- discoveries. “This discovery demonstrates ter with 3.7 magnitudes of visual extinction. that strongly gravitationally lensed quasars High-dispersion spectroscopy of member do exist at redshift above five, despite the red giants indicates that HP 1 has metallic- fact that we’ve been looking for over 20 ity [Fe/H] ≈ − 1.1 dex and is α-enhanced by years and have not found any others this far about a factor of two. However, the age had back in time,” said Fan. “However, we don’t been uncertain because past photometric expect to find many quasars brighter than studies were unable to reach beyond the this one in the whole observable Universe.” main sequence turn-off (MSTO). The study has been published in The Astro- A new study by an international team of as- physical Journal Letters. tronomers presents a detailed analysis of deep near-infrared observations of HP 1 ob- Excavation of an Ancient Star tained with the Gemini South Adaptive Op- Cluster Deep in Milky Way Bulge tics Imager (GSAOI) using the Gemini Multi- conjugate adaptive optics System (GeMS). Of the roughly 160 globular clusters known The GeMS/GSAOI J and Ks images, shown in in the Milky Way, roughly a quarter appear Figure 18, have spatial resolution of about 0.1 to be associated with the Galactic bulge. Al- arcsecond and probe two magnitudes be- though these are generally more metal rich low the MSTO. The study was led by Leandro

January 2020 / 2019 Year in Review GeminiFocus 43 Kerber of the Universidade de São Paulo and measured radial velocity and the absolute Figure 19. Universidade Estadual de Santa Cruz in Brazil. proper motion given by Gaia (Data Release Results of fitting the 2) in order to constrain the cluster’s orbit. GeMS near-infrared The team combined their GSAOI data with They find that HP 1 passes just 0.12 kpc from CMD of HP 1 using archival F606W (wide V) images from the the Galactic Center at closest approach the Dartmouth Stellar HST’s Advanced Camera for Surveys to de- and reaches a maximum distance of about Evolutionary Database termine relative proper motions and select (DSED) models. Left 3 kpc. It is likely that many of the cluster’s bona fide cluster members. They then fit- panel: CMD showing stars have been stripped away as it has re- ted two different sets of model isochrones all likely member stars peatedly plunged through the bulge during (grey) and those used in to the color-magnitude diagrams (CMDs) the course of its long history. the fit (black). The best-fit to determine the stellar population param- isochrone is indicated eters, distance, and reddening. Figure 19 “HP 1 is one of the surviving members of by a thick green line; the shows the results for one set of isochrones the fundamental building blocks that as- green shading shows using only the GeMS/GSAOI data; the team sembled our Galaxy’s inner bulge,” said the uncertainty range. The red arrow indicates also performed fits to CMDs made with a Kerber. Added coauthor Mattia Libralato of a change in reddening combination of HST and GeMS data. The the Space Telescope Science Institute, “The of ΔE(B − V) = 0.10 mag. analysis indicates an age near 13 billion combination of high angular resolution and Right panels: the result- years, confirming that HP 1 is one of the old- near-infrared sensitivity makes GeMS/GS- ing one- and two-dimen- est globular clusters in the Milky Way and AOI an extremely powerful tool for study- sional constraints for all likely formed less than a billion years after ing these compact, dust-enshrouded stel- model parameters. The the Big Bang. lar clusters.” The study appears in Monthly contours correspond to Notices of the Royal Astronomical Society. confidence levels of 0.5σ, The heliocentric distance of 6.6 kiloparsecs 1.0σ, 1.5σ, and 2.0σ. (kpc) estimated from the isochrone fitting John Blakeslee is the Chief Scientist at Gemini Figure reproduced from agrees well with the distance implied by Kerber et al., Monthly Observatory and located at Gemini South in the extinction-corrected brightnesses of 11 Notices of the Royal Chile. He can be reached at: Astronomical Society, RR Lyrae stars identified within the cluster. [email protected] 484: 5530, 2019. The team combined this distance with the

44 GeminiFocus January 2020 / 2019 Year in Review Ricardo Salinas and Steve B. Howell July 2019

The Legend of Zorro Begins In May, Gemini successfully commissioned Zorro, the Observatory’s new dual-channel, dual-plate speckle interferometer. Now permanently installed at Gemini South, the instrument allows diffraction-limited speckle imagery of binary stars, multiple stellar systems, Solar System objects, and your own favorite target!

The atmosphere forgives no one. It does not matter whether you have a futuristic 30-me- ter telescope or a more modest 1-meter telescope, your image quality will be domi- nated and limited by the same factor: atmospheric turbulence. How can we overcome the tyranny of the atmosphere to unleash the real potential (the diffraction limit) of a telescope?

One solution is to circumvent the atmosphere altogether and put the telescope in orbit — as evidenced by the breathtaking beauty of Hubble Space Telescope and other orbit- ing astronomical observatory images, which testifies to the enormous appeal of this solution. But as much as we would like to put Gemini in orbit, we simply can’t; this would not only be very expensive, but above all, our technicians and engineers would really hate their daily commute!

Reaching the Diffraction Limit from Earth A different solution involving shorter commutes is the one given by adaptive optics. In adaptive optics, the incoming wavefront, distorted by the atmosphere, is measured and then corrected using deformable mirrors. One excellent example is the Gemini Multi- conjugate adaptive optics System combined with the Gemini South Adaptive Optics Imager, reaching near the diffraction limit in the K-band.

January 2020 / 2019 Year in Review GeminiFocus 45 Figure 1. One instrument capable of doing speckle An individual interferometry is the Differential Speckle speckle frame Survey Instrument (DSSI, Horch et al., 2009), (top left), the which visited Gemini North and South on integrated image of 1,000 speckle multiple occasions since 2012. Visiting in- frames (top right), struments expand the capabilities of what the Fourier power the facility instruments can offer, but come spectrum (bottom with a significant burden in logistics: permis- left), and the sions must be obtained, agreements signed, resulting recon- the equipment shipped, a dedicated crew of structed diffrac- people must travel, some facility instrument tion-limited image (bottom right). must be removed, and finally the visiting in- Adapted from strument must go through testing and com- Scott and Howell missioning. Is there another viable solution? (2018). In other words, is it possible to make the visi- Yet another solution, far less expensive than tor feel truly at home? the latter and easily implemented at optical wavelengths, is speckle interferometry. First proposed by French astronomer Antoine Enter Zorro! Labeyrie in 1970, speckle interferometry is Zorro (and its sibling ‘Alopeke at Gemini Figure 2. based on the idea that atmospheric turbu- North) is a new dual-channel, dual-plate- The design of Zorro. A lence can be “frozen” when obtaining very scale (field of view) speckle interferometer pickoff mirror deflects short exposures. In these short exposures, permanently mounted on Gemini South. the light coming from stars look like a collection of little spots, or In simpler words, Zorro can obtain two the tertiary mirror, redi- speckles (Figure 1), where each of these recting it into Zorro. diffraction-limited images with different Inside Zorro, the light is speckles has the size of the telescope’s dif- filters simultaneously. Besides the speckle split by a dichroic into fraction limit. When taking many exposures, mode (which gives a field of view of only a red and blue channels to and using a clever mathematical approach, few arcseconds), Zorro also has a wide-field their respective cameras these speckles can be reconstructed to form mode with a field of around 1 arcminute. The equipped with electron- the true image of the source, removing the speckle mode reaches the diffraction limit of multiplying CCDs. effect of atmospheric turbulence. Gemini (15 miliarcseconds at 500 nanome- ters), while the wide-field delivers an image quality between the diffraction limit and the natural seeing. Limited testing has shown images with an image quality of around 0.15 arcsecond.

Zorro (the Spanish word for Fox) is indeed small and clever, like its furry namesake. Mounted between the instrument support structure and the calibration unit at Gemini South, it solves the perennial problem of which facility instrument must be displaced by not displacing any. Since it doesn’t re- quire a port of its own, Zorro is free to take up residence as a “permanent visitor.”

46 GeminiFocus January 2020 / 2019 Year in Review The commissioning of Zorro occurred May 20-23, 2019, when the team from NASA Ames who designed and built the instru- ment (Steve Howell, Nic Scott, Rachel Mat- son, and Emmett Quigley) came to Gemini South to assemble, install, and calibrate the instrument. Despite some battles with the weather, the first science run started imme- diately after commissioning. data. Zorro observed this system during its Figure 3. first science run and ruled out the presence The Zorro commissioning Science with Zorro of any other unresolved stellar companion, team — from left to right: confirming the inferred size of the planet. What kind of science can benefit from the Rachel Matson, Steve diffraction limited images delivered by Zor- But the research done with Zorro does not Howell, and Nic Scott (NASA Ames) — together ro? The main science driver of the renais- stop there. Its exquisite image quality can with Gemini South Sci- sance of speckle interferometry has been also be used to study the whole zoology of ence Operation Specialist, the study of stars hosting exoplanets. The binary stars, multiple stellar systems, Solar Joy Chavez, look happy study of exoplanets has been revolutionized System objects, and maybe even to do some after achieving first light. with dedicated space missions like NASA’s extragalactic science. The selected target was Kepler (now retired), K2, and Transiting Exo- the star N Velorum. Zorro is now commissioned and ready to do planet Survey Satellite (TESS), which have science. What can you do with images hav- discovered thousands of new exoplanets via ing a spatial resolution of ~15 miliarcsec- the transit method — that is, the little dips in onds? We wait for your observing proposals the light curve of a star when a planet passes by the end of September! in front of (transits) it. As impressive as these missions are, they have one problem: be- Figure 4. cause they observe large fields of view con- Kelt-25 as seen by Zorro. The Zorro observations taining hundreds of thousands of stars, their show Kelt-25 has no stel- pixel scales are necessarily coarse, several lar companions, thereby arcseconds or more. confirming the nature of the newly discovered But what if the transited star is actually a transiting giant planet binary star? The properties of the planet de- KELT-25b. rived from the light curve can change radi- Credit: Joey Rodriguez, cally whether the planet is transiting one Sam Quinn, and Josh or the other star. This is where the power of Pepper (KELT-TESS); Steve Zorro is manifest. Following up stars with Howell, Nic Scott, and transits observed by Kepler/K2 and TESS and Rachel Matson (NASA looking for close stellar companions, it can Ames) confirm and clarify the nature and proper- ties of detected exoplanets. Ricardo Salinas is an Assistant Scientist at Gemini One example is the newly discovered giant South. He can be reached at: planet KELT-25b with a 4.4 day orbit around [email protected] its parent star. This discovery was possible Steve Howell is the Space Science & Astrobiology with a combined analysis of the Kilodegree Division Chief at NASA Ames. He can be reached Extremely Little Telescope (KELT) and TESS at: [email protected]

January 2020 / 2019 Year in Review GeminiFocus 47 January 2020 Gemini Press Release A Galactic Dance “Everything is determined… by forces over which we have no control. … Human beings, vegetables, or cosmic dust, we all dance to a mysterious tune, intoned in the distance by an invisible piper.” — Albert Einstein

Figure 1. Image of the interacting galaxy pair NGC 5394/5 (also known as the Heron Galaxy) obtained with NSF’s National Optical- Infrared Astronomy Research Laboratory’s Gemini North 8-meter telescope on Hawaii’s Maunakea using the Gemini Multi-Object Spectrograph in imaging mode. This four-color composite image has a total exposure time of 42 minutes. North is up. Credit: Gemini Observatory/NSF’s National Optical-Infrared Astronomy Research Laboratory/AURA

Galaxies lead a graceful existence on cosmic timescales. Over millions of years, they can en- gage in elaborate dances that produce some of nature’s most exquisite and striking grand designs. Few are as captivating as the galactic duo known as NGC 5394/5, sometimes nick- named the Heron Galaxy. The image in Figure 1, obtained by the Gemini Observatory of NSF’s National Optical-Infrared Astronomy Research Laboratory, captures a snapshot of this compelling interacting pair.

48 GeminiFocus January 2020 / 2019 Year in Review The existence of our Universe is dependent larger galaxy (and a few in the smaller gal- upon interactions — from the tiniest sub- axy). Also visible is a dusty ring seen in sil- atomic particles to the largest clusters of gal- houette against the backdrop of the larger axies. At galactic scales, interactions can take galaxy. A similar ring structure appears in millions of years to unfold. This new image this previous image from the Gemini Obser- released in December captures a moment vatory, likely the result of another interact- in the slow and intimate dance of a pair of ing galaxy pair. galaxies some 160 million light years distant A well-known target for amateur astrono- and reveals the sparkle of subsequent star mers, the light from NGC 5394/5 first piqued formation fueled by the pair’s interactions. humanity’s interest in 1787, when William As in all galactic collisions, these galaxies are Herschel used his giant 20-foot-long tele- engaged in a ghostly dance. Astronomers scope to discover the two galaxies in the have concluded that the two partners have same year that he discovered two moons of already “collided” at least once, though the Uranus. Many stargazers today imagine the distances between the stars in each galaxy two galaxies as a heron. In this interpreta- preclude actual stellar collisions. Neverthe- tion, the larger galaxy is the bird’s body and less, galactic collisions can be a lengthy pro- the smaller one is its head — with its beak cess of successive gravitational encounters, preying upon a fish-like background galaxy! with each galaxy’s gravity deforming the NGC 5394 and NGC 5395, also known col- other’s overall shape. Over time the galax- lectively as Arp 84 or the Heron Galaxy, are ies can morph into exotic forms that bear no interacting spiral galaxies 160 million light resemblance as to how we see them today. years from Earth in the constellation of One by-product of the pair’s turbulent in- Canes Venatici. The larger galaxy, NGC 5395, teractions is that hydrogen gas coalesces is 140,000 light years across, and the smaller into regions of star formation. In this image, one, NGC 5394, is 90,000 light years across. these stellar nurseries appear as reddish See the full image release here. clumps scattered in a ring-like fashion in the

January 2020 / 2019 Year in Review GeminiFocus 49 Gemini staff contributions On the Horizon Figure 1 (top). The GHOST spectrograph This review highlights instrumentation development efforts in the lab prior to the made in 2019 to advance the Observatory’s capabilities to do installation of the inner enclosure. leading science, especially in the era of multi-messenger and Credit: NRC-HAA time-domain astronomy. Figure 2 (bottom). The GHOST spectrograph with JANUARY 2020 inner enclosure and assemblies for blue and red detectors. GHOST on the Move Credit: NRC-HAA For the past six months, the assembly, alignment, and test of Gemini’s High-resolution Op- tical SpecTrograph (GHOST) in Victoria, British Columbia, has gone very close to plan; we expect to ship the instru- ment to Gemini South in February 2020.

The newest instrument chosen for the Gemini South tele- scope, GHOST was designed, and is being built and tested, by a partnership of organizations: Australian Astronomical Optics (AAO)-Macquarie University, the National Research Council Canada (NRC)-Herzberg, the Australian National University (ANU), and Software Design Ideas. During the latter half of 2019, the AAO, which designed and built GHOST’s Slit Viewer Assembly and Optical Fiber Cable, made multiple visits to NRC-Herzberg, where they partici- pated in each sub-assembly’s integration and testing with the spectrograph.

The spectrograph (Figures 1 and 2) has performed ex- cellently during the Acceptance Testing of the past few months. Test results for resolution, throughput, and stabil- ity all look great in the lab. We will repeat the verification of these and other performance requirements after all is re-assembled at Gemini South. Having developed the data reduction and instrument control software, the ANU and Software Design Ideas were also key participants in

50 GeminiFocus January 2020 / 2019 Year in Review the recent testing success. Figure 3 hints at some of GHOST’s capabilities.

The Cassegrain Acquisition Unit, also de- signed and built by the AAO, was previously shipped to Chile from Australia and tested in advance of the upcoming arrival of the spec- trograph. After the spectrograph, slit viewer, and optical cable arrive in Chile, we expect to have all sub-assemblies of the GHOST in- strument fully integrated and functioning in the second quarter of 2020 in preparation for required removing the AO’s three large opti- Figure 3. cal components and dismantling the original commissioning. Blue and red GHOST NGS system. But it all worked out, thanks to images of a mercury the careful preparations made by the NGS2 lamp, with the spectral First Light with NGS2 team. orders labeled and 1.1 x free spectral The Canopus adaptive optics (AO) bench of Commissioning took place last October. range in each order the Gemini Multi-conjugate adaptive optics Apart from Gemini personnel, the team had highlighted. Continuous System at Gemini South recently received a the great pleasure to work with François wavelength coverage significant upgrade: its new Natural Guide Rigaut (Australian National University) and from 359 nm to well Star Wavefront Sensor, also known as the Benoit Neichel (Laboratory of Astrophys- beyond 1 micron (Requirement: 363 - Natural Guide Star Next Generation Sensor ics of Marseille) during the commissioning 950 nm). Significant (NGS2). The original system consisted of three nights (Figure 5, next page). Collaboration wavelength overlap moving probes to pick up guide stars in the from the weather was a weak point, serious- between orders (with field, channel the light into fibers, and proj- ly hampering progress. However, the team overlapping orders ect it onto a quad-cell for tip-tilt detection. tested the full system, and put it through its between arms). This system worked but in practice was cum- paces. Credit: Greg Burley bersome to use, mainly due to each probe’s The first results have been very positive. tiny field of view and the large light losses AO performance under reasonable weather in the system. This implied large acquisition conditions achieved an image quality of 83 Figure 4. times and a brightness limit for the stars that milliarcseconds, indicating that the fully in- The new NGS2 unit after significantly restricted sky coverage. tegrated system worked well (Figure 6). Ac- installation into the For the above reasons, a team from the Aus- quisition of the three natural guide stars was Canopus optical bench. tralian National University spearheaded an alternative approach, making use of novel electron-multiplying CCD technology that allows imaging the whole field of view. Up to three guide stars can be selected on that image. For tip-tilt wavefront sensing on each of the stars, small windows centered on each star are then read out at high speed, making use of the extreme low noise characteristics of the electron-multiplying CCD.

The new NGS2 was incorporated into the Canopus optical bench last September (Fig- ure 4). This was no trivial exercise, because it

January 2020 / 2019 Year in Review GeminiFocus 51 Figure 5. of NGS2 was primarily done by a team at the A tense moment Australian National University in collabora- in the control room when all tion with Gemini engineers and astrono- the AO loops mers. Without such a strong collaboration were closed the project would not have prospered. with NGS2 for the first time. Official first SCORPIO Making Steady light on NGS2! Progress Credit: René Rutten With the exciting build phase of the Spec- trograph and Camera for Observations of very quick, achieving a gain of several min- Rapid Phenomena in the Infrared and Optical utes over the original NGS system for every (SCORPIO) — a powerful next generation in- acquisition (Figure 7). Figure 6 (below, left). strument for Gemini South — underway, the During the brief A key driver for the NGS2 project was to work SCORPIO team has been busy in the last quar- moments of good with fainter guide stars. Whereas the original ter of 2019. On November 22nd, Gemini staff weather on the NGS system could guide down to about R = and subcontractor FRACTAL attended the commissioning nights, the full MCAO system 15.5 magnitude under good conditions, the project’s Quarterly Review at the Southwest with NGS2 could be new system has been proven to work even Research Institute (SwRI) facilities in San Anto- tested on the central beyond R = 18 magnitude; a remarkable im- nio, Texas. Gemini staff noted that significant condensation of stars provement that significantly increases sky progress has been made on several fronts, in the Tarantula Nebula coverage, bringing many more objects into including the Slit Viewing Camera (SVC), ther- (RMC 136). reach of the GeMS/GSAOI instrument. mal design, Failure Mode and Effects Analy- sis, and Center of Gravity issue. Discussions Figure 7 (below, right). During the upcoming observing runs with included assembly-of-instrument and focal- NGS2 control panel, GeMS/GSAOI, we will gain more experience plane-mechanisms failure modes. showing the field of view on NGS2’s performance. We will update the on the NGS2 EMCCD web pages with the latest information for A total of five Manufacturing Readiness Re- camera, and the identified three natural guide users. Meanwhile, we invite interested users views (MRRs) have also taken place: an Elec- stars. Individual guide to exploit the system with new and star window images are previously inaccessible targets. shown in real time on the bottom left. This system The NGS2 project has been made has proven to be much possible thanks to the tight col- easier, faster, and better laboration between many people. to operate. The initiative, a large part of the funding, and the design and build

52 GeminiFocus January 2020 / 2019 Year in Review tronics MRR on November 6th at SwRI at- fiber cable, slit view assembly, and associ- Figure 8. tended remotely by Gemini; NIR Collimators ated electronics to the HAA in early July. On December 18, 2019, and NIR Cameras Opto-mechanics MRR on Both groups, along with the Australian Na- staff from Gemini, November 20th, with SwRI contractor Offici- tional University and Software Design Ideas, FRACTAL, SwRI, Johns na Stellare in Madrid; Mounts and Mechanics have been working together the past several Hopkins Univeristy (JHU), and SwRI contractor MRR on November 21st in Madrid; VIS Col- months to integrate the various hardware, Winlight Systems met in limators and VIS Cameras Opto-mechanics optics, electronics, and software into a func- Madrid for SCORPIO’s VIS MRR on December 18th with SwRI contrac- tioning spectrograph (Figure 10, next page). Cameras Opto-mechanics tor Winlight Systems; and Cooled Electron- MRR. Standing in the back Now that this major milestone is nearly com- ics Box and SVC MRRs on December 18th in (from left to right): Kelly plete, the combined teams expect to enter Madrid (Figure 8). Smith (SwRI), Gerardo into the test phase in October. If all testing Veredas (FRACTAL), Coming up in 2020, the SCORPIO Science goes to plan, the team expects to ship short- Thomas Hayward Team will gather in March for a SCORPIO Sci- ly after the start of 2020 to Chile, where the (Gemini), Manuel ence Meeting in Washington, D.C., closely AAO-built acquisition unit is ready to connect. Maldonado (FRACTAL), Robert Barkhouser followed by a project Quarterly Review tak- (JHU), Vincent Lapere ing place at Gemini South facilities in La Ser- GNAOI Request for Proposals (Winlight Systems), Jean- ena, Chile; there the SCORPIO team will get François Gabriel (Winlight the opportunity to visit Cerro Pachón and Gemini Observatory announces an oppor- Systems), Ernesto Sánchez participate in nighttime observations. Gemi- tunity for the Gemini North Adaptive Optics (FRACTAL), Pete Roming ni is working closely with the SCORPIO team Imager (GNAOI) — a planned instrument to (SwRI), and Stephen Smee (JHU). And standing in to ensure continued steady progress. be used with both the Gemini North multi- front (from left to right): conjugate adaptive optics system (GNAO; Marisa García-Vargas OCTOBER 2019 which is now under development) and with (FRACTAL), Todd Veach a planned future ground-layer adaptive op- (SwRI), Massimo Roberto GHOST: Project Build Phase tics system. We expect that this imager will (JHU/Space Telescope Nearing Completion Science Institute), Ana Pérez (FRACTAL), The National Research Council Canada’s and Stephen Goodsell Herzberg Astronomy and Astrophysics (Gemini). (HAA) Research Centre is in the final few Credit: Marisa García- weeks of the build phase for the Gemini Vargas, FRACTAL High-resolution Optical SpecTrograph Figure 9. (GHOST). HAA staff fine-aligned the spec- Alan McConnachie of trograph optics and are now able to take HAA taking some test test spectra (Figure 9). The Australian As- spectra with GHOST in the tronomical Optics (AAO) group at Mac- HAA integration lab. quarie University delivered the science Credit: Scot Kleinman January 2020 / 2019 Year in Review GeminiFocus 53 be a low-cost low-risk placed in a vacuum tank in a thermally sta- design using a single ble enclosure, which the instrument team HAWAII-4RG detector, assembled in the Gemini North Pier Lab, and intend for GNAO four levels below the telescope. After a year to provide a 2-arc- of monitoring the temperature stability in minute field of view the enclosure, commissioning the Front End with a Strehl ratio (which mounts on the Instrument Support of no less than 30% Structure and holds the optical fiber posi- over the entire field tioner), and integrating the spectrograph of view under median itself in the Gemini North Pier Lab, the team seeing conditions in K has begun commissioning. See a press re- Figure 10. band. A Request for Proposals to design this lease about MAROON-X available from the Scott Roberts of HAA imager has been released and is available on University of Chicago. showing Telescope the Gemini website here. Scientist Tom Hayward of Gemini the impressive GIRMOS Conceptual Design outer enclosure for MAROON-X Deployed at GHOST. Review a Success Gemini North Credit: Scot Kleinman The Gemini InfraRed Multi-Object Spectro- MAROON-X, a new visiting high-resolution graph (GIRMOS) is a powerful new visiting spectrograph at Gemini North (Figure 11), instrument being designed and built for the Figure 11. will be available to users in 2020. Construct- Gemini telescope by a Canadian consortium The MAROON-X ed at the University of Chicago, MAROON-X of universities led by the University of Toron- instrument team and is expected to be able to detect Earth-size to and HAA. This instrument will overcome Gemini staff pose with planets in the habitable zones of mid- to a key limitation in existing adaptive optics the instrument, installed late-M dwarfs using the radial velocity de- (AO) facilities; where exisitng integral field on Gemini North on tection method. spectrographs are designed to observe only September 23rd, and single objects with adequate atmospheric ready for first light. The important wavelength range for the correction, GIRMOS is being designed to Although the Front instrument is 700-900 nanometers and the End was successfully have the ability to observe multiple sources resolving power approximately 80,000. To commissioned in simultaneously with high spatial resolution achieve this precision the instrument must December, now if you while obtaining spectra at the same time look closely, you can be intrinsically stable and the optical setup (Sivanandam et al., 2018). see the optical fiber fixed, so the entire instrument has been that runs down to the GIRMOS accomplishes this by tak- spectrograph in the ing advantage of the latest develop- Gemini North Pier Lab ments in multi-object AO (MOAO) below. Left to right: Paul McBride, John Randrup, and integral field spectroscopy. It Rody Kawaihae, Harlan exploits the AO correction from both Uehara, Eduardo Tapia a telescope-based AO system (either (all Gemini staff), GeMS or the prospective Gemini Andreas Seifahrt, North AO system) and its own addi- David Kaspar, Julian tional MOAO system that feeds mul- Stuermer (all University tiple 1- to 2.4- micron integral field of Chicago), and Alison Peck and John White spectrographs (R = ~3,000 and 8,000) (both Gemini staff). that can each observe an object in- Credit: Julian Stuermer dependently within a 2-arcminute field of view. 54 GeminiFocus January 2020 / 2019 Year in Review GIRMOS is in the very early stages of devel- exceptional instrument team supporting 50 Figure 12. opment, and the team, led by Suresh Siva- nights of observing, we were not able to fit GIRMOS PI Suresh nandam (Principal Investigator; University in all of the great science that was proposed. Sivanandam (University of Toronto, Dunlap Institute) and Darren Er- of Toronto, Dunlap If you missed your chance to use IGRINS in ickson (Project Engineer; HAA), have been Institute; in shadow at 2018, never fear! We are delighted to an- the front of the room) working extremely hard to complete a con- nounce that IGRINS will join us once again presents the agenda ceptual design for the instrument and to at Gemini South for several semesters, start- to the Review Panel, identify the resources needed to make the project members, and ing with 2020A. If you were not able to get project a success. We are very happy to re- participating Gemini your proposal in for the 20A deadline, don’t port that they passed their Conceptual De- staff at a GIRMOS despair, keep your eye out for IGRINS in the sign Review on September 18, 2019, follow- Conceptual Design next several Calls for Proposals. Review held at HAA ing a very exciting few days of presentations in Victoria, British and discussions at the Dominion Astrophysi- Columbia. cal Observatory, in Victoria, British Columbia 20th Anniversary Gemini Credit: Marcin Sawicki (Figure 12). We look forward to continuing to Science Meeting (Saint Mary’s University) work with this great team as they move for- ward to the next stage of the project. Con- Gemini Observatory invites its interna- gratulations to the team! tional user community to Seoul, Korea, for a special 20th anniversary Gemini Science Meeting (GSM) celebrating 20 years of sci- Multiple Opportunities ence operations and a look forward to even to Use IGRINS more exciting things to come. Hosted by the Partnership’s newest member, the topics You probably remember when the visiting will include the latest scientific results from Immersion GRating INfrared Spectrometer Gemini, news on current instrumentation (IGRINS) came to Gemini South in 2018. This projects, updates on operations develop- cross-dispersed near-IR spectrograph — ments, and lively discussion of Gemini’s stra- with a resolving power of R = 45,000, cover- tegic plans for the coming decade. The GSM ing the H and K windows (from 1.45 to 2.5 mi- will take place June 21-25, 2020, followed by crons), in a single exposure, providing both the K-GMT Users’ Meeting on June 26th. (See broad spectral coverage and high spectral poster, next page.) For information and up- resolution — had a very high oversubscrip- dates, see the Gemini Science Meeting 2020 tion rate. A large number of very impressive website. programs were observed, but even with the

January 2020 / 2019 Year in Review GeminiFocus 55 56 GeminiFocus January 2020 / 2019 Year in Review JULY 2019 GHOST Project Achieves Major Milestone During the first two weeks of July, the com- bined Australian and Canadian GHOST teams worked together to reach a major milestone in Victoria, British Columbia: the integration of subassemblies created by each organi- zation for Gemini’s High-resolution Optical SpecTrograph (Figure 13). The Australian Astronomical Optics Macquarie University team brought with them the Slit Viewer As- • An acquisition and guiding slit. Figure 13. sembly with electronics, as well as the Opti- • A simultaneous wavelength calibration National Research cal Fiber Cable to be connected to and tested light injection port. Council Canada team members John Pazder, with the Spectrograph, which the National The fiber system also includes two asso- Andre Anthony, and Scott Research Council Canada team had recently ciated devices: (1) a mode-scrambling, assembled. A spectrum captured with this Macdonald (from left to noise-reducing agitator that creates vari- right) fit check GHOST’s instrument is shown in Figure 14. Software able conditions for propagation of light in red camera optics onto Design Ideas, and staff from the Australian all of the optical fibers; and (2) a calibrator the focus stage. National University, provided software sup- that is the reference source for simultane- Credit: David Henderson port during this effort. ous wavelength calibration via a Thorium- This work bought the fiber system, Slit View- Xenon lamp. er Assembly, and spectrograph together for The Slit Viewer Assembly uses a beam splitter the first time. to direct 99% of the slit output to the spec- The fiber system, which sits between the Figure 14. Cassegrain Unit and the Slit Viewer Assem- Image of bly, includes the following components: spectrum • 62 individual fibers that connect the captured from Cassegrain Unit to the Slit Viewer As- the location where the sembly. GHOST blue • The microlens IFU units that consist of detector will be two low-resolution arrays and one high- positioned. resolution array, each with a separate ar- Credit Tony ray for sky. Farrell • A flexible conduit for the optical cable that minimizes stress on the fibers, there- by reducing Focal-Ratio Degradation. • Spectrograph slit optics that form a slit from each object. The slits are 1 micro- lens wide and either 7 or 19 microlenses long in the standard- or high-resolution modes, respectively.

January 2020 / 2019 Year in Review GeminiFocus 57 trograph and 1% to the slit imaging system. It The team will work the remainder of the also removes the need for an on-instrument year to complete the final integration and wavefront sensor for flexure compensation, testing before shipping to Chile near the with the telescope’s peripheral wavefront end of the year. sensor being used for fast tip/tilt and focus corrections. SCORPIO: Moving Toward Its The spectrograph subsystem is a gravity- Build Phase stable asymmetric white-pupil échelle spec- trograph, with two arms and volume-phase On June 5-7, the SCORPIO project held its holographic grating cross-dispersers. It com- Critical Design Review (CDR) at the South- prises the following key elements: west Research Institute (SwRI) headquarters in San Antonio, Texas. Team members from • An optical table that maintains spectrograph SwRI, FRACTAL (an instrument design firm stability and provides thermal mass for the in Madrid, Spain), Space Telescope Science environmental enclosure sub-system. Institute, Johns Hopkins University, George • A Slit Viewer Assembly unit, discussed Washington University, and Gemini Observa- above, that directs 99% of the light from tory, participated in the review, presenting the slit to the collimator. material to an eight-member external review • A collimator mirror that collimates the committee. John Troeltzsch from Ball Aero- beam from the Slit Viewer Assembly and space and the National Center for Optical- directs it to the échelle grating. infrared Astronomy Management Oversight • An échelle grating that disperses the light Council chaired the very experienced exter- into the échelle orders. nal review panel. • Two transfer mirrors: one convex fold mir- The reviewers recognized and congratu- ror and the white pupil relay mirror. The lated the team for the tremendous amount transfer mirrors and the collimator mirror of work and effort spent in progressing the together form the white pupil relay that project since the Preliminary Design Review. reimages the pupil of the dispersed light In the following weeks, Project Executive at the échelle onto the Volume Phase Holo- Scot Kleinman took the identified concerns, graphic gratings. issues, and risks from both the external re- • A beam splitter that separates the light into view committee and the internal Gemini blue and red channels. staff reviewers and crafted a comprehensive CDR Executive Report that contained rec- • Blue and red gratings that act as both the ommended actions to close out the Design cross-dispersers, to separate the échelle or- Phase of the project and reduce risk going ders, and to introduce an anamorphic fac- forward into the Build Phase. tor for more efficient use of the effective area of the detector in the cross-dispersion We remain confident that the SCORPIO team direction. will build a successful instrument for Gemini. • Blue and red multi-element camera lenses. SCORPIO is a complex and challenging in- strument to create, and the finished product • Blue and red detectors that collect the full promises to become a major capability at the wavelength ranges of each camera, mount- Observatory, aiding scientific discovery in ed in separate cryostats. the coming decades. • Focus controls for each camera.

58 GeminiFocus January 2020 / 2019 Year in Review APRIL 2019 GEMMA’s Out of the Gate In the new year, the Gemini in the Era of Multi-Messen- ger Astronomy (GEMMA) program is off to a good start. Gemini North Adap- tive Optics (GNAO) Principal Investigator Gaetano Sivo formed an external Gemini AO working group to pro- vide community experience and expertise regarding the Observatory’s AO program, including developing sci- ence cases, technical rec- ommendations, and best practices. The Real Time Computer project is performing some tech- Preparations for MAROON-X Figure 15. nology trade studies and considering wheth- Computer-aided design er some components can be designed and MAROON-X is the new radial velocity spec- rendering of the vacuum built in-house. trograph being built at the University of chamber and cameras on Chicago and expected to be deployed at the MAROON-X bench. The time-domain astronomy (TDA) project Gemini North within the next year (Figure The actual spectrograph is also moving along, convening a working 15). This high-resolution, bench-mounted is expected to arrive at group to review user stories related to the Gemini North in May. spectrograph has been designed to deliver concept of operations. In addition, Public 1 meter/second radial velocity precision for Information and Outreach plans to hold a M dwarfs down to and beyond V = 16, and Time-Domain Astronomy Summit later this is expected to have the capability to detect year; the goal is to bring together scientists Earth-size planets in the habitable zones us- and communications and education profes- ing the radial velocity method. sionals to create a roadmap on how to com- municate the concepts of MMA and TDA to Following the success of the Front End com- non-scientists. The program continues to de- missioning, we are planning to install and fine short- and long-term benefits of the in- align the spectrograph in the dedicated en- dividual projects to the future of Observatory closure in the Pier Lab in May 2019. If all goes operations and the astronomy community. well, we hope to complete commissioning Dave Palmer has joined the team to work as in time to include this exciting new visiting Project Manager for both the GNAO and RTC instrument in the 2020A Call for Proposals. GEMMA efforts. He will be working with Act- Watch this space for more information as in- ing GNAO Project Manger Stephen Goodsell tegration and commissioning progresses on during the Conceptual Design Stage, allow- Maunakea. ing Stephen to step back from the role after the Conceptual Design Review.

January 2020 / 2019 Year in Review GeminiFocus 59 Second On-sky Testing of GHOST on the side-looking port, as well. This mainly consisted of checking that the GHOST op- The Gemini High-resolution Optical Spec- erations software was properly accounting Trograph (GHOST) team completed the sec- for the additional reflection produced by the ond round of on-sky testing at Gemini South tertiary (science-fold) mirror and producing in November, 2018. The team successfully the correct coordinate transformations and demonstrated proper operation of the at- ADC corrections, among other things. With mospheric dispersion correctors (ADCs), the some minor tweaks, GHOST worked success- instrument on the side-port of the instru- fully on the side-port. ment support structure, and the interactions between GHOST and the Observatory Con- The team used the prototype optical fiber trol System (OCS) software. cable for this round of Cassegrain unit test- ing. The science optical fiber cable is nearing GHOST uses ADCs to correct for the disper- build and test completion. Upon comple- sion of light by the atmosphere. Rather than tion, the cable and Cassegrain unit build a full-field ADC, GHOST features mini-ADCs team (the Australian Astronomical Optics for each fiber positioner, which offers im- Group at Macquarie University) will ship proved efficiency. The build team tested these components to the spectrograph each of the mini-ADCs to ensure that the build team at the National Research Council hardware and software were working cor- Herzberg in Victoria, Canada, where they will rectly to provide the optimal dispersion cor- be paired with the spectrograph for testing rection as expected. Each ADC was tested in the second half of 2019. The Australian Na- over a range of target zenith distance and tional University team, along with a contrac- position angles. These tests demonstrated tor, Software Design Ideas, is providing the that the ADCs are working as expected and instrument control and data reduction soft- produce the required correction. ware for GHOST; they were also instrumental These tests also marked the first on-sky test- in the November Cassegrain unit testing, as ing of GHOST interoperability with the Gem- were Gemini project team members from ini OCS. GHOST target configurations, in both North and South sites. both high- and standard-resolution modes, were created in the Gemini Observing Tool. SCORPIO Update The telescope systems then used these tar- get configurations to determine the tele- At the end of February, Southwest Research scope pointing. GHOST also used them to Institute (SwRI) hosted a progress meeting place the fiber positioners on the requested in San Antonio, Texas, to assess the matu- targets. While these successful tests were a rity of the SCORPIO project’s Critical Design major milestone in our internal software de- Review (CDR) documentation set. SwRI has velopment process, they also improved the provided Gemini with drafts of the Critical efficiency of the on-sky tests by greatly re- Design documents and the team continues ducing the time for target acquisitions at the to work on providing additional structural telescope. and thermal analysis required for the review. A readiness assessment will take place at the The team operated GHOST on both the beginning of April. The project has now re- up-looking and side-looking ports. While ceived the instrument’s four science grade GHOST is expected to operate primarily on visible detectors. the up-looking port during normal opera- tions, we wanted to ensure proper operation

60 GeminiFocus January 2020 / 2019 Year in Review Gemini staff contributions

News for Users A summary of news events throughout 2019 of relevance to the Gemini user community.

JANUARY 2020 DRAGONS First Public Release After many years in the making, it is with great excitement that we announce the first public release of Gemini's new Python-based data reduction platform, DRAGONS (Data Reduction for Astronomy from Gemini Observatory North and South). DRAGONS’ capabilities were vital in enabling scientists to quickly reduce data critical to observations of the interstellar Comet C/2019 Q4 (Borisov). Click this link to access a related article for more details; also see Science Highlights in this issue starting on page 27.

2020 Large and Long Programs Call for Proposals Gemini Observatory announces opportunities for new Large and Long Programs. Eligible Principal Investigators (PIs) from Canada and the US are invited to propose scientific inves- tigations to begin observations in Semester 2020B. Large and Long Programs either require significantly more time than a partner typically approves for a single program or extend over two to six semesters, or both. Eligible PIs are also invited to submit Large and Long Program proposals for Subaru Intensive Programs, via the Gemini-Subaru time exchange. Letters of Intent are required to be submitted no later than February 4, 2020, with complete proposals due April 1, 2020. Details on Large and Long Programs, the proposal process, and specifics on the 2020 proposal cycle are available on the Large and Long Program webpages.

January 2020 / 2019 Year in Review GeminiFocus 61 Gemini Short Surveys the number of future allocated programs, and manage blocks of scheduled instru- Gemini staff would like to thank all of our ments (such as GPI, GRACES, or GSAOI). user communities for the 694 replies to the Phase I, Phase II, End of Semester, and Phase We are very pleased by the high satisfaction III surveys over the year 2019. Such a massive rate we are getting from the majority of our response rate is invaluable to ensure we give users, and by the warm comments of appre- useful support and deliver good quality data. ciation on the quality of our support work. This motivates us to continue to find creative Every response is compiled and all com- ways to improve our work, and collaborate ments read and reported, in an anonymous with the researchers that depend on Gemini form, to all the Gemini staff. The most recur- for their science. We hope to continue to sat- rent problems are identified and escalated isfy the scientific needs of the researchers of to make sure they are addressed as soon as the Gemini community, and to fix the issues possible. For example: that are an obstacle to our common success. 1. Many people report that they are strongly irritated by the “Observations” and “Band Registration Open for 2020 3” sections in PIT (more precisely, the way targets are entered, and requested Gemini Science Meeting in time is defined). We took note of those Seoul comments, and used them to define the Registration is now open for the Gemini Ob- requirements for the new software tool servatory 20th Anniversary Science Meeting, that will handle Gemini proposals. Known to be held in Seoul, Korea, June 21-25, 2020. as the new OCS Upgrades Program, this Early registration at a discounted rate is avail- tool is expected to be completed in 2023. able until February 28th. See the meeting On the other hand, a certain number of website to register and submit your abstract! problems met by the proposers can be avoided by following this tutorial. 2. A significant fraction of PIs who need fur- OCTOBER 2019 ther work on Phase IIs report that defining the observing sequences is too complex Maunakea Access and Gemini and often confusing. The biggest issues North Shutdown happen when an important modifica- tion needs to be made, like changing the On July 15th, protesters blocked the Mau- choice of grating or the observing mode. nakea Access Road in an effort to prevent Once again we will use these comments Thirty Meter Telescope construction equip- to determine how the future Phase II soft- ment from moving to the Maunakea Astron- ware will work. Meanwhile, we strongly omy Precinct. This action quickly precipitat- recommend PIs contact their contact sci- ed a protracted stoppage of all observing entist (you can find their email in the Ob- atop Maunakea, as observatories assessed serving Tool) or to send a helpdesk ticket. the safety and reliability of access to the summit. By August 12th, we had received as- 3. We look for systematic complaints from surances of support from Law Enforcement PIs of programs using a given instrument and statements from the protestors of their or in a certain Band. This information intent to allow access for staff of the exist- helps greatly to determine the semester’s ing observatories. Combined with some im- schedule efficiently, make decisions on

62 GeminiFocus January 2020 / 2019 Year in Review provements made to the “spur road” Figure 1. (a short segment of the old Saddle Assistant Engineer Road and a portion of a lava field) Mariah Birchard via which we now have to access the (left) and Senior mountain, we returned to work on Electronic the planned maintenance shutdown Technician (excluding coating of the primary Alejandro Gutiérrez (right), work on the mirror which has been deferred un- maintenance of til next year). The maintenance was one of the modules completed on August 30th, allowing of the Acquisition a resumption of night-time observ- and Guiding Unit. ing and enabling the TEXES instru- Credit: Manuel ment to visit as scheduled. Access to Paredes the mountain remains intermittently compromised by conditions on the spur road in particular, but for now we are ensure that Gemini remains at the forefront proceeding with operations. of ground-based optical/IR astronomy and best serves the needs of our international Figure 2. user community throughout the coming de- Cover of Gemini’s Gemini South Annual Shutdown cade. We encourage all of our users to have a Strategic Scientific Plan, a Success look and see where Gemini is headed! approved by the Gemini Board of Directors. A successful annual shutdown at Gemini South ran from August 12th to August 27th. Accomplishments during this shutdown in- cluded replacement of the Cassegrain rota- tor encoder, repairs to the helium lines, and maintenance of the Acquisition and Guid- ance Unit (Figure 1). In the spirit of sharing resources ramping up to the National Sci- ence Foundation’s Center for Optical-infrared Astronomy (NCOA), we had some excellent support from a few Cerro Tololo Inter-Amer- ican Observatory technicians, doing cross- training and knowledge sharing and tighten- ing relationships.

Strategic Scientific Plan The Strategic Scientific Plan (SSP; Figure 2), approved by the Gemini Board of Directors during their most recent meeting, outlines the scientific direction and activities of the Gemini Observatory in the 2020s. It also pro- vides a timeline for the Observatory’s major instrumentation and operations develop- ment efforts. The motivation for the SSP is to

January 2020 / 2019 Year in Review GeminiFocus 63 JULY 2019 Figure 3. GMOS-South CCD Gemini North engineer John White working Intervention on the CCD focal The Gemini Multi-Object Spectro- plane array just prior graph (GMOS) at Gemini South has, for to replacing the electronics board. some time, suffered from instabilities in the charge-coupled device (CCD) Credit: Luc Boucher readout. Since the installation of the Hamamatsu CCDs, they have been performing sub-optimally. In particu- lar, we have seen instances where the charge transfer efficiency became too large, causing smearing on the im- ages, which affects the popular Nod & Shuffle mode of the instrument. We have been planning to tackle this issue by changing the existing electronics board inside the cryostat with one of a better design. This design has been proven to work for the GMOS instrument at Gemini Planet Imager Gemini North, which does not experience Temporarily in Lab for Testing the same smearing effect. The critical and very sensitive intervention was carried out Recently the Gemini Planet Imager devel- in June (Figure 3). The technical interven- oped a problem with its Micro-Electrical Figure 4. tion went very well, and the lab tests quickly Mechanical System deformable mirror. In- Gemini staff share showed very promising results. GMOS-S was vestigations in the lab have indicated that information with CASCA put back in normal operation and the effect a critical electronics board related to the participants in the June of charge smearing has not been seen again. power supply has failed. Repairs are under- meeting in Montréal. The CCD array performs to specification. way, and in the meantime, the instrument will remain in the lab and unavailable for science.

CASCA 2019 meeting in Montréal Gemini staff participated in the 2019 CASCA (Canadian Astronomical Soci- ety/Société Canadienne d’Astronomy) meeting in Montreal, which was held at McGill University from June 17-21 (Fig- ure 4). Besides hosting a booth in col- laboration with the Canadian Gemini Office (Stéphanie Côté, Joel Roediger, and Tim Davidge), we were available

64 GeminiFocus January 2020 / 2019 Year in Review to directly work with many users who had Phase I, Phase II, and data reduction ques- APRIL 2019 tions. Scot Kleinman also presented a talk GEMMA-TDA Advisory Group about the future role of Gemini in the Time- Domain Astronomy era. We thank McGill Assembled University for organizing a successful meet- Guided by a Gemini Science and Technology ing, and we look forward to meeting with ev- Advisory Committee action regarding time- eryone again at York University, in Toronto, domain astronomy (TDA) and multi-mes- next year. senger follow-up, we have assembled a rep- resentative team of astronomers from across Gemini North Primary Coating the Partnership to advise us on our devel- oping plan for TDA. This advisory group, During Shutdown chaired by Abhijit Saha of the National Opti- The Gemini North primary mirror will get a cal Astronomy Observatory (NOAO), has had new coat in the course of an extended mid- two meetings as of early March. year shutdown, which is scheduled to start The members are: Abhijit Saha (US; NOAO) - on July 23rd. The same coating recipe will Chair; Andres Jordan (CL); David Sand (US); be used as is currently on the mirror, which Basilio Santiago (BR); Meg Schwamb (Gemi- comprises four distinct layers deposited ni); Federica Bianco (US); Myungshin Im (KR); by sputtering different magnetron targets. Maria Drout (CA); Craig Heinke (CA); Victoria Closest to the glass substrate, a 65-Ångstrom Alonso (AR); Alexander Vanderhorst (US); (Å)-thick layer of nickel chromium (NiCr) acts and Andy Adamson, Bryan Miller, and John as an adhesive layer between the glass and Blakeslee (all Gemini, in attendance). the overlying reflective silver layer. The silver Not all of the members of this time-domain is sputtered onto the NiCr, at a much greater advisory group work on time-domain sci- thickness of 1100 Å. Next a wafer-thin layer ence; the mission of the group includes of NiCr is sputtered on top of the silver; with protecting the completion of non-TDA pro- a thickness of only 6 Å. Finally, an overcoat grams in the coming Large Synoptic Survey of silicon nitride is applied by sputtering a Telescope era when we expect to have an boron doped silicon target with nitrogen increased number of Target of Opportunity process gas. The thin NiCr appears to facili- proposals. We are grateful to Abi and the tate the growth of a dense and protective group for helpful commentary to date. silicon nitride layer, and slows any corrosion. The current coating has lasted well, but at six years since the last coating, it’s time to TOPTICA Laser: replace it. Available Every Night! Other jobs in the shutdown include replac- With a fully commissioned TOPTICA laser, ing and upgrading the helium supply hoses we are back in operation for Laser Guide in the Cassegrain wrap, replacing the glycol Star (LGS) mode at Gemini North. The 19A coolant hoses, and some instrumentation semester will be a “transition” period from work, including dealing with a bubble in scheduled laser blocks to a fully-integrated the oil interfaces in the Gemini Multi-Object LGS queue operations model. This will al- Spectrograph lens system. low for LGS programs to be observed on any night when conditions allow, giving Gemini Principal Investigators access to LGS adap-

January 2020 / 2019 Year in Review GeminiFocus 65 tive optics observing through- Figure 5. out the semester (Figure 5). Gemini Science Operations Specialist Michael Hoenig (back) The Next Generation and Gemini Senior Laser Technician Jeff Donahue Natural Guide Star discussing LGS operations Sensor for GeMS for the TOPTICA laser in the Gemini Base Facility Gemini South’s multi-conjugate Control Room in Hilo. adaptive optics system pro- vides for an adaptive optics (AO) Credit: Jeff Donahue and the mechanical arrangement is complex corrected field of about one arcminute. To in operation. achieve this important capability, the system relies on a constellation of five laser guide Therefore, some years ago, Gemini entered Figure 6. stars and up to three natural guide stars in into a collaboration with the Australian Na- order to sense and correct for atmospheric tional University (ANU) to develop a better The NGS2 unit on the system designed around the now available test bench in La Serena. turbulence. The original design of the natural high-speed Electron Multiplying (EM) CCD The high-speed EM CCD guide star sensor has been in operation now camera is on the left- for several years. It is based around three me- cameras. Using an EM CCD imager will re- hand side. The orange chanical probes picking up stars in the field. sult in much improved sensitivity. The mov- lines are fibers to mimic Each probe channels the light onto optical ing probes will no longer be necessary, as multiple guide stars that fibers leading to avalanche photodiodes for the full patrol field will be imaged onto the are imaged onto the CCD, while regions of interest around the detector. fast centroiding. Unfortunately, the sensitiv- ity of this system leaves much to be desired, selected stars will be read out at high speed Figure 7. to provide centroiding information to the AO real-time control system. This results in The NGS2 test team, from left to right: Cristian much simpler acquisition procedures, and Moreno, Mariah Birchard, achieves much better sky coverage, since Gaetano Sivo, Brian Chinn, fainter stars will become accessible. François Rigaut, Ian Price, Ignacio Arriagada, Pedro In February a major milestone was Gigoux, Natalie Provost, achieved on this project. The system de- Gianluca Lombardi, and veloped by ANU arrived in La Serena, Eduardo Marin. René Chile, where it was installed on a test Rutten is on the business bench and integrated with the AO real- end of the camera. time control system (Figures 6 and 7). Credit: René Rutten The results have been excellent, proving that the system will work as designed. Much work remains to be done. In- tegration of the new sys- tem, named “NGS2,” in the existing multi-conjugate AO system will not be a trivial task. If all goes well, we expect to do this early in Semester 2019B.

66 GeminiFocus January 2020 / 2019 Year in Review Semester 2018B Outcomes conditions despite the loss of five nights to a major earthquake in January 2019. Note We’re now in the thick of Semester 2019A that in 18B we took data on the last of the and taking stock of the outcome of 18B. Pre- traditional “rollover” programs; from now liminary completion results for programs on, regular queue Band 1 programs (except in the regular queue (in other words, ex- Target of Opportunities, Fast Turnaround, cluding Targets of Opportunity and block- Director’s Discretionary, and Large and Long scheduled instrument modes) are shown Programs) have one semester of “persis- in Figure 8. Band 1 programs at both sites tence,” and so some of those will continue to fared rather well, three quarters of them accumulate data as we continue into 2019A. reaching 100% completion. In the North, Band 3, which typically takes the more re- laxed observing conditions, fared relatively Gemini North Survives Wild worse — another reflection of the fact that Weather 18B was better than either of the preceding B semesters in Hawai‘i. As we reported in our recent e-newscast, on February 10, 2019, a low-pressure system In the South, the completion rate was better (Figure 9, next page) subjected Maunakea than it has been for many semesters, thanks to some of the highest wind speeds ever to a healthy percentage of stable, good recorded. While there’s reason to be skepti- Figure 8. For Gemini South (upper) and Gemini North (lower) the completion histogram for Semester 2018B. Horizontal axis shows the program completion in 10% bins, and vertically the colored bars show the fraction of programs in Bands 1, 2, and 3, which reached that completion percentage. Main features are described in the text. Credit: Andy Adamson

January 2020 / 2019 Year in Review GeminiFocus 67 cal of the widely-reported peak icant failures would be expected. The recent gust speed of 191 miles per additions to the support building, namely hour (mph), winds in excess of the many solar panels and base-facility op- 150 mph (just below Category eration environmental sensors, were de- 5 Hurricane force) were reliably signed to the same wind speed standard as recorded on the summit on that the rest of the building, and all survived the day (Figure 10). wind event intact and remained functional.

Winds of that speed at this el- This wasn't a particularly unusual storm sys- evation, pushing on a structure tem; it was a “Kona low," a low-pressure sys- of the scale of the Gemini dome, tem which usually settles to the west of the is sufficient to produce a force of islands (hence the name) but which this time around 280 tons sideways. The was to the north. To put the wind speeds in Figure 9. Gemini telescope facility is rated to survive perspective, an extreme winter storm on The low-pressure such winds with no distress to materials or Mount Washington in New Hampshire, USA, system to the north of structure. Even somewhat stronger winds of in 1934, produced a wind gust of 231 mph, the Hawaiian Islands, on February 10, 2019, order 160 mph would not threaten the struc- and in 1996 Cyclone Olivia produced a wind Hawaiian Standard ture, as deformations would remain below gust of 253 mph, setting a new world record. Time. The circulation the elastic limit. However, at 200 mph, signif- center is clearly visible in the lower-level cloud pattern (in grey). Image taken from the MKWC satellite archive; go there and select 11-Feb UTC to see animations.

Figure 10. Top panel: The CFHT/ Gemini observed weather data from the Maunakea Weather Center site , at the time (16:43 HST) of the highest gust experienced there — 161 mph (top row, middle, red). Bottom panel: This screenshot from the Maunakea Weather Center shows a wind speed of 96 knots (110 mph) recorded by the CFHT/Gemini weather tower on February 10th at 16:40 HST (bottom frame).

68 GeminiFocus January 2020 / 2019 Year in Review Manuel Paredes and Alyssa Grace April 2019 Gemini Outreach Programs Sparkle in Both Hemispheres Gemini’s two leading public outreach endeavors — AstroDay Chile and Journey Through the Universe continue to uphold one of Gemini’s primary missions: to share the wonders of the Universe with the public.

Astroday Chile 2019: Preparing our host communities for the July 2nd Total Solar Eclipse With excitement mounting over the upcoming July 2nd total solar eclipse over La Serena, Chile, AstroDay Chile on March 23rd was primed to educate its ~3,000 visitors about this special event. Held at the Seminario Conciliar School of La Serena, the program provided educational material and talks about the eclipse, and taught participants how to view the partial phases safely (Figure 1). Numerous other activities and exhibitions were also fea- tured. To help make AstroDay Chile 2019 a success, 23 organizations joined in on the excite- ment of bringing astronomy to the people.

Coordinated by Gemini South’s Public Information Office, the event offered to students, families, and the public, a wide variety of activities, such as science workshops, lectures, 3D cinema, water-rocket launches, solar viewing, and portable planetarium presentations.

Two key partners helped organize this year’s event: the Association of Universities for Re- search in Astronomy (AURA), and the Municipality of La Serena. All of the major obser- vatories in Chile — including the European Southern Observatory (ESO), ALMA, and Las Campanas — and most of the astro-tourist facilities in the surrounding Coquimbo Region also united to participate in this year’s program.

The images shown on page 70 illustrate some of the the activities of the day.

Manuel Paredes is the Communications Coordinator at Gemini South. He can be reached at: [email protected]

January 2020 / 2019 Year in Review GeminiFocus 69 Figure 1. Figure 2. A mother helps her kids Families took inject air into a water rocket for an advantage of amazing lift off! This workshop was AstrodayChile’s the one most preferred by children and special solar- adults during AstroDay Chile. viewing event to learn how to safely observe the total solar eclipse on July 2nd in the Region of Coquimbo.

Figure 3 (left). After sunset, many participants formed lines to see the Moon and stars through telescopes supplied by local amateur astronomers and the Cerro Tololo Inter-American Observatory. Through their kindness and help, AstroDay fulfilled its promise to share the wonders of the Universe.

Figure 4 (right). AstroDay Chile was a good venue for the AURA staff to work together in outreach activities. Seen here, from left to right, Gemini Electronics Engineer Vanessa Montes, and Kathy Vivas and Cesar Briceño (both astronomers from Cerro Tololo Inte- American Observatory, interact with the public to explain the science and technologies that AURA centers currently apply in Chile.

Figure 5. The Gemini/AURA booth was one of the most visited, thanks to the help of the kids from the Gemini Robotics Club. They in turn explained to children how robotics can be used to control remote systems, such as the one that controls the Base Facility Operations at Gemini South. Credit: All photos on this page by Manuel Paredes

70 GeminiFocus January 2020 / 2019 Year in Review Journey Through the Universe: Hawai‘i 2019 The 15th year of Journey Through the Universe, Gemini Observatory’s flagship education and outreach program, brought astronomy professionals from Maunakea and across the nation into Hawai’i island classrooms, visit- ing thousands of students — one classroom at a time (Figures 6-11). The diverse group of astronomers, scientists, engineers, and in- formal educators provided an authentic and Figure 6. Hilo-Waiākea and Ka’ū-Kea’au-Pāhoa Complex Area personal window into the process of scientific Superintendent Chad Farias speaks about the success of Journey in the community and its future. discovery and the splendors of our Universe. Credit (all Journey photos): Joy Pollard During Journey “week,” which began on March 2nd, 80 astronomy educators shared their pas- sion for science with approximately 8,000 stu- dents. Journey as a year-round program also includes StarLab Portable Planetarium shows for grades K-1, career panel presentations for high schoolers, astronomy educator work- shops, Lunar and Meteorite Sample Certifica- tion workshops hosted by NASA’s Solar System Exploration Research Virtual Institute team, Family Science Night, and a public presenta- tion on recent discoveries from the telescopes on Maunakea.

Updates on what Journey is accomplishing in Figure 7. Two students learn about robotics provided by the Hawai'i the community can be viewed here. Science and Technology Museum at Journey's Family Science Night.

Alyssa Grace is an Outreach Assistant at Gemini Figure 8. Science Operations Specialist, Jocelyn Ferrara (far left) uses an North. She can be reached at: 8-meter tarp in the classroom to model the size of the primary mirror in the [email protected] twin Gemini telescopes.

January 2020 / 2019 Year in Review GeminiFocus 71 Figure 9. Digital Architect, Jason Kalawe (standing) discusses careers and career diversity at the Maunakea observatories with local high schoolers.

Figure 10. Science Fellow, Matt Taylor (fourth from the right), uses a large scale interactive model to show students the relationship between actual distances to stars and perspective.

Figure 11. Assistant Astronomer, Trent Dupuy (left), helps students understand the scale of star sizes by making paper models.

72 GeminiFocus January 2020 / 2019 Year in Review Alison Peck October 2019 Papa ‘Ōlelo Hawai‘i Kilohōkū* Figure 1. Ka‘iu Kimura, Executive Director of Astronomers in Hawai‘i have long embraced Hawaiian culture ‘Imiloa, speaking to 2,000 astronomers and traditions, including finding ways to include them in the and students at the naming of astronomical discoveries. Now, through a new exciting January 2019 American Astronomical Society program, hosted in part at Gemini Observatory, Observatory staff meeting in Seattle, are on their way to better understanding the history and culture Washington. Credit: ‘Imiloa that shape the communities in which they live. Astronomy Center

One of the many wonderful aspects of living in Hawai‘i is the strong sense of history and culture that makes these is- lands unique. Learning more about this culture and how it has shaped the com- munities we live in is an important goal for most observatory staff, whether they grew up here, have become long- term residents, or are making the most of a short-term position, like an intern- ship or postdoctoral fellowship. In addi- tion to everyday life in the community, we can see the ‘ano nui (importance) of Hawaiian culture through novel astron- omy programs such as A Hua He Inoa, a Hawaiian phrase that refers to the prac- tice of calling forth a name. This collab- orative naming project, led by the ‘Imiloa Astronomy Center in Hilo, Hawai‘i, includes ex- perts in Hawaiian culture, language, and astronomy and aims to weave traditional culture and practices into the process of officially naming astronomical discoveries. In January 2019, Ka‘iu Kimura, Executive Director of ‘Imiloa, was invited to give a lecture about the program at the January 2019 meeting of the American Astronomical Society in Seattle,

* Hawaiian language lessons for astronomers

January 2020 / 2019 Year in Review GeminiFocus 73 Washington, which attracted an audience of and the UH Institute for Astronomy (both in more than 2,000 astronomers and students. Hilo and Mānoa on O‘ahu) registering, and attendance and enthusiasm remaining just But not only astronomers work at the obser- as high throughout the semester. vatories, of course. There are engineers, tech- nicians, librarians, accountants, educators In Hilo, the class met in the Lecture Hall at and more, many of whom were born here in the Gemini North Base Facility, which is op- the islands. Many observatory staff have the timized for sound quality and ease of class opportunity to hear ‘ōlelo Hawai‘i (Hawaiian participation. The class was streamed in re- language) and oli (chants) through their chil- altime to sites in Waimea on the Big Island, dren, who learn about important traditions Mānoa, and even one participant in Iowa, Figure 2. and mo‘olelo (stories) in school, but gaining a using videoconferencing technology that Maunakea observatory more in-depth knowledge and understand- the observatories have in place to enable staff preparing for the ing requires a more concerted effort. That’s scientific collaboration. Although this under- Merrie Monarch parade in April. The Merrie why the ‘Imiloa Astronomy Center recently taking was technically challenging at first, Monarch is a week-long joined forces with the University of Hawai‘i after a few learning experiences on the part festival that honors the at Hilo’s (UHH) Ka Haka ‘Ula O Ke‘elikōlani of the organizers, the class was transmitted legacy of King David College of Hawaiian Language to provide a smoothly to all sites. Kalākaua, who inspired weekly class on Hawaiian language and cul- the perpetuation of The organizers also recorded each class and ture to staff from all observatories on Mau- Hawaiian traditions, made them available to all participants, so nakea. The observatories paid the tuition native language and that they would not miss anything if they for the 12 week course, and the participants arts. could not attend. This was all made possible purchased their own textbooks, which they Credit: East Asian by the outstanding skill of kumu (teacher) kept after the classes finished. Observatory

The first class was a bit experimental, as it Kamalani Johnson (UHH), and his willing- was difficult to gauge how many people ness to embrace not only the challenges would be able to attend the class every Fri- of distance learning, but also an unusual day lunchtime, and how many would be set of haumāna (students) from all over able to make time to watch the recordings the globe and all types of jobs, from scien- and practice the lessons on their own if they tific research and education, to engineering, were traveling or on a night shift. Never- computer support, and administration. With theless, participation was outstanding with participants from diverse backgrounds, all over 100 staff from Maunakea Observatories levels of proficiency in ‘ōlelo Hawai‘i, and

74 GeminiFocus January 2020 / 2019 Year in Review Figure 3. Haumāna Alyssa Grace (left) and Jocelyn Ferrara (middle) present kumu Kamalani Johnson with a photo of the Maunakea observatories signed by several class participants. Credit: Alison Peck (Gemini)

extremely varied personal and professional interests, one would expect that holding the attention of everyone in the class for 12 weeks would be a challenge. But Kamalani handled it with ease, dividing the class time between stories (legends, place names, tra- ditions, hula) and grammar, vocabulary, and sentence structure.

Participants in the class said they looked for- ward to Friday lunchtime every week, and were quite sad when the course ended. We at Gemini Observatory are extremely grate- ful for kumu Kamalani Johnson and to Ka‘iu Kimura for their eagerness to lead this ini- tiative, and especially for their willingness to work with the observatories on plans to continue providing these courses for obser- vatory staff who are so grateful to have the opportunity to pursue their careers while also becoming more knowledgeable about the history and culture that shape the com- munities in which they are privileged to live and work.

Alison Peck is an Instrument Program Scientist at Gemini North. She can be reached at: [email protected]. January 2020 / 2019 Year in Review GeminiFocus 75 Gemini staff engage local Honolulu students in an infrared imaging activity at the January meeting of the American Astronomical Society at the Hawai‘i Convention Center. Credit: Joy Pollard/Gemini Observatory

The Gemini Observatory is operated by the Association of Universities for Research in Astronomy, Inc., under a cooperative agreement with the National Science Foundation on behalf of the Gemini Partnership. Canada Chile Brazil Argentina Korea Gemini Observatory 670 N. A‘ohoku Place Hilo, HI 96720, USA