In early November of 2018, my colleagues Alison Coil and Gene Leung and I were beginning our third observing run with a new instrument on the Keck Observatory, the Keck Cosmic Web Imager (or KCWI). We had proposed to the telescope time allocation committee that we should use KCWI to study the true size of outflows in nearby galaxies with actively accreting supermassive black holes. In fact, we spent most of the night doing just that. However, we were also interested in another sample of star-forming galaxies that we know hosts fast outflows, but for which we had failed to get observing time through the competitive proposal process. So, like good scientists, we did a little exploration: we devoted a small portion of our time that night as a test run on this other sample, just to see if anything interesting showed up. Our target was selected because it has a broad emission line from oxygen, which is often a signature of a fast galactic wind. The initial computer processing of the data during the night showed some interesting structure in the emission lines. When we compared with an optical image of the galaxy's stars, which reveals strong tidal tails resulting from an earlier merger between two galaxies, we assumed the emission line structure was from these tidal tails and moved on.
Several weeks later, Alison emailed to see if I had done a more careful computer processing, or reduction, of the data; she and her collaborators had looked again at the stellar image and she was starting to wonder if a tidal interpretation was really correct. Maybe it was an outflow? I said I hadn't gotten around to it, but that I would look into it. It wasn't until a month later, when I was again observing with KCWI on New Year's Eve 2018, that I found time to do a proper data reduction. I aligned the resulting emission line image from ionized oxygen, which traces diffuse gas, and the image of the galaxy's stars. This time, I compared the scale of the two images carefully. The result seemed different than our casual inspection from two months earlier, because the gas was now on much larger scales than the starlight. "OK, I must have calculated it wrong," I thought to myself. "Let's do this again." When I recalculated the image scales and again found the gas nebula was enormous compared to the starlight, it was the "aha" moment of discovery---whatever it was I was seeing was not what we expected. Shortly thereafter I sent Alison an email that included a number of superlatives and the phrase, "We need to write a paper!!!" Her response was equally effusive.
Besides the thrill of a eureka moment, why might this have been so exciting? Galactic winds have been studied for decades, and a lingering question has been how far they extend from their host galaxies. This ultimately determines how winds drive the interaction between galaxies and the gas that lives around and between them. It has long been thought that winds are in important source of circumgalactic and intergalactic gas, but few winds have been found conclusively far from their hosts. This is where new instrumentation like KCWI plays a key role. KCWI was designed specifically to be very sensitive to faint gas far from galaxies, using highly reflective image slicers (seen in the image at right).
The discovery of such an enormous galactic wind brought me into a new collaboration with a group that had been working on these galaxies for over a decade, starting in 2007 with a publication by Christy Tremonti and co-authors. This group had amassed a large, multi-wavelength dataset on their sample of galaxies, including the galaxy we observed with KCWI. As a result, there was immediately a large amount of data and analysis to help validate and further interpret our new result. The KCWI result prompted one group member, Jim Geach, to dive into the data on molecular gas from the Atacama Large Millimeter Array, ALMA. He immediately saw that our galaxy had a a molecular outflow, as well, and that its morphology correlated with that of the ionized gas outflow we had found. Furthermore, modelling of the stellar population that had been completed years earlier had shown two episodes of star formation that lined up surprisingly well with two distinct structures in the outflow. Ultimately, this paper became a true collaborative effort, effectively harnessing the skills and experiences of a large group of expert astronomers. It was a case study in how scientists should work together.
The image above shows the ionized oxygen emission in colour from "our" galaxy, heretofore un-named. The small dark patch at center is the galaxy's starlight, which is highly concentrated within the tiny green circle. Jim Geach was the first to suggest we name our galaxy, and he proposed "Makani." Naming in astronomy is typically a dry task, usually involving some sort of catalog label. Some subfields of astronomy have more fun with this than others (there are lots of asteroids named after people, and of course the planets!). Astronomers who study galaxies---let's just say names are not done, mostly. But this galaxy and its nebula seemed special, at least to us. So we thought, why not? At worst, we're the only ones who end up calling it Makani. And at best, it has a moniker which reflects why it's scientifically important and also speaks to the origin of the data itself. Makani is a native Hawai'ian word which means "wind." Astronomers have for decades made amazing discoveries using telescopes atop Mauna Kea, which is a site that is sacred for many Hawai'ians. Mauna Kea has been in the news recently because of conflict over use of the mountain. I can only speak for myself and say how fortunate we are to conduct observations from this mountain. Many native Hawai'ians, conversely, feel disrespected on their own land, and they are speaking up for what they believe is right. There is plenty of history on their side. Perhaps we can at least find common ground for a brief moment in appreciating the wonder of this newly minted galactic wind. The first Hawai'ian explorers to find the way to their homeland on the winds of the ocean might be proud.