ABOVE: Observatories on the summit of Mauna Kea depend on accurate predictions from the Mauna Kea Weather Center (MKWC) to make the most of their telescope time.
I'm milling around in a courtyard at the University of Hawaii at Manoa, waiting for a forecast meeting. I'm more excited than anyone should be about a National Weather Service weekly forecast update. This is my first step in understanding how to forecast weather around Hawaii and specifically at one of the most interesting and extreme places in the world: the summit of Mauna Kea.
Starting in the 1980s, telescopes were built atop the tallest mountain in the world (measured from seafloor to summit; from sea level Mauna Kea is 13,796 feet). But it isn't just the mountain's height that makes it optimal for astronomy. Being in the middle of the Pacific, on the most remote archipelago in the world, Hawaii is far from light pollution. But also importantly, the weather at Mauna Kea's summit is ideal for stargazing about three hundred days of the year. The other sixty-five can be downright nasty, with winds strong enough to blow doors off of cars and snow too deep for the rare Hawaii snowplows. So, predicting which days will be ideal is critical information. With hundreds of millions invested in building and operating these observatories, few weather predictions are as valuable as those of the Mauna Kea Weather Center. It's up to MKWC to provide the highest possible resolution forecast so that precious telescope time can be used efficiently and safely.
Technicians check one of the Submillimeter Array's receivers.
MKWC director Steven Businger acclimates to the altitude at Hale Pohaku.
A false-color map illustrates winds and water vapor, both important factors for "good seeing."
The Hawaii regional office of the National Oceanic and Atmospheric Organization's National Weather Service (NWS) resembles a well-run military command post—NOAA is a uniformed service, after all—with multiple displays at each workstation. (One major difference between NOAA and, say, a Navy command post is that it's perfectly acceptable to have a surf cam feed on a workstation display.) This is where NOAA forecasters consult "the Global," that is, the Global Forecast System (GFS), which is shared—and revered—internationally; the idea is that everyone in the world shares data and forecasts because everyone needs a heads-up about bad weather. On a day like today, when the weather is calm, there's plenty of room, but when something like a hurricane menaces the Hawaiian Islands, the workforce and workload might triple.
The satellite image of the Pacific displayed on the main screen is a wash of false colors, with barely visible dots representing the Hawaiian Archipelago in the center, the Marshall Islands to the left and hints of the US West Coast on the far right. The NWS model is built from data from ground stations, weather balloons and satellite data from just about every country in the world, then compiled and crunched by a supercomputer on the continent. This big Pacific view is one of many outputs, albeit one more valuable to transpacific sailors than to people planning a weekend beach campout.
The NWS focuses on predicting big weather events, things that impact commerce or threaten the safety of the region. From the big Pacific-wide view, the displays zoom in on Hawaii. A small-craft advisory is in effect for most of Hawaii (pretty common during the winter) and "trade winds forever" is the forecast. The GFS is pretty low-res; not one city is mentioned by name, and specific islands are noted only in passing. It's the job of local weather forecasters to pore over the GFS' many charts to predict potential weather impacts on their areas. Here, it might be raining deep in Manoa Valley, while it's sunny a mile downhill in Waikiki. Even on dry days it's usually raining in Hawaii, somewhere.
This is all very interesting, but nothing about Mauna Kea is being mentioned. But, the "air desk" chimes in with a report that an airliner traveling at thirty thousand feet encountered unexpected turbulence. Pilots take great care to avoid turbulence, and the GFS is especially useful for flight planning. At that altitude the airliner was well into the jet stream, one of the high-altitude rivers of two hundred-plus mph winds that circle the globe and drive big changes in regional weather. There, on the false color satellite map tortured with squiggly lines, is the source of that unexpected turbulence, a "little" whorl dipping south a hundred miles from what should be a smooth-flowing jet stream. And that information could very well impact observations from Mauna Kea.
Operator Angelu Ramos monitors data from the Submillimeter Array remotely from Hilo. Radio telescopes like SMA can observe in daylight and are less affected by weather conditions than optical telescopes, which require clear, dark skies and calm wind for optimal seeing.
The MKWC, conveniently stationed directly above the NWS center, zooms in much more finely than the NWS. (Also unlike NWS, MKWC is not a government agency; it's funded by the observatories themselves, even though its data is available to the general public.)
MKWC director and principal investigator Steve Businger is taking a break from grading his grad students' labs, and lead modeler/research meteorologist Tiziana Cherubini is poring over the MKWC's own model. Unlike the broad regional forecast from the NWS downstairs, Businger and Cherubini are focused on a few square kilometers of Mauna Kea's summit. A supercomputer in Wyoming crunches inputs from the GFS, along with data from an array of local sensors and four daily weather balloon reports, to inform MKWC's high-resolution model.
The result is ... more charts and squiggly lines. Tiziana explains the importance of an "inversion layer" detected at or below ten thousand feet—a critical data point insofar as summit astronomy is concerned. "This inversion layer keeps tradewind moisture from climbing any higher than the peak of Saddle Road," she says, pointing to a horizontal divot in a smooth line on a chart. For the next three days, the forecast for Mauna Kea astronomy is looking good, but after that, she says, conditions degrade. Why?
Tiziana switches to an image of the winds at the jet stream level from the GFS. The jet stream has dipped south, as it does in most winters, but a tributary has formed and will pass right over the Islands in three days. "That's bad for seeing," Businger says, explaining that the fast-flowing winds of the jet stream create turbulence, which distorts starlight. Optical telescopes love very calm air; the "optical turbulence" from the jet stream, although fifteen thousand feet above the summit, compromises seeing (and that is what astronomers call it: "seeing").
I point excitedly to the little whorl that was responsible for the airliner's encounter with turbulence. They both give me a gracious nod that would make a meteorology student smile. I've correctly surmised that the whorl portends that our observation-friendly inversion layer will dissipate by Friday. Indeed, it does.
Technicians walk among three of SMA's eight robotic receivers, which can be positioned in specific patterns to optimize observing cold, gaseous bodies and black holes; SMA was one of many radio telescopes that provided data for the first images of a black hole in 2018.
But that doesn't mean I should quit my day job just yet. The weather over the next two weeks—late February to early March of 2024—goes bananas, which is, I think, the official meteorological term. The whorl turns into a whirlwind. Local news weather forecasters are flummoxed to the point of issuing on-air apologies. Sunny one day and Island-wide rain the next. MKWC's forecast page is a string of red and blue bars, indicating snow and turbulent air at the summit. In such conditions the astronomers and even the maintenance personnel of Mauna Kea take a snow day.
"The warm pool of air that moved through was very alarming to me. ... I haven't seen that before over the last twenty-plus years of forecasting for Mauna Kea," says Ryan Lyman. Lyman, a forecast meteorologist, is the third member of the MKWC team, responsible for maintaining the websites, data links and a handful of weather instruments on the summit. Most of the observatories have their own sensors and cameras, and MKWC combines all available metrics for the summit as a whole to fulfill its mandate of ensuring the safety of observatory operations personnel on their trek above the snow line.
For the past twenty years, Lyman has been writing the few paragraphs of prognostication that appear on the MKWC website first thing in the morning and at 5 p, five days a week. The narratives are long, prophetic and an enjoyable read for a lay meteorologist like myself, filled with discussions of cloud layers, convection, ridges and troughs, organization and instability, and boundary layer turbulence. But to put MKWC's forecast to the grindstone, you have to get high.
Businger (left) and lead modeler Tiziana Cherubini, two of the three meteorologists comprising the MKWC, in their main office on the University of Hawaii at Manoa campus.
A technician inspects the controllers on one of the SMA's receivers.
If you head up to the Mauna Kea summit, your first stop is Hale Pohaku (stone house) at nine thousand feet, the mandatory altitude adjustment station for visitors and Mauna Kea veterans alike. The cafeteria is lively with observatory personnel chatting and acclimating over breakfast. At the entrance a huge flatscreen TV displays vital information: webcams, road conditions and the MKWC forecast.
For the past two weeks, the summit was mostly closed due to heavy snow and ice, according to three ice detectors that Lyman maintains. Now, with a break forecast and despite the persistent, pesky jet stream, Mauna Kea personnel are heading up to clear snow from the domes and perform the innumerable tasks associated with keeping the observatories running. With these high winds, the optical telescopes have little hope for good seeing, but the radio telescopes can study the cosmos despite these conditions.
Radio telescopes have sensitive receivers, like oversize satellite dishes, that detect electromagnetic whispers from comets, nebulae and even faint emissions from black holes. One such telescope, the Submillimeter Array, attempts astronomy on every marginal day and night that MKWC forecasts. Instead of a giant parabolic dish, the SMA consists of eight synchronized robotic receivers. Like most of Mauna Kea's observatories, SMA has a backlog of four requests for every available observation period, so the moment the snow is dusted off SMA's silvery receivers, observation begins.
Operator Angelu Ramos grew up just downhill from Mauna Kea. She has studied and conducted radio astronomy since graduating from Kohala High School in 2018 and during college. During fair weather, observers like Ramos prefer to operate the array from the summit, but remote observing from SMA's Hilo or Boston headquarters is the better option during questionable weather. I'm granted access to watch an astronomy session with the SMA crew without having to climb the summit at night. For once, I'm excited to join a Zoom call.
The WM Keck Observatory fires twin lasers to create "artificial stars" in space to help pinpoint the telescope. For optical telescopes like those at Keck, MKWC provides critical information about cloud cover and high-altitude winds that distort observations even on clear nights.
With an average of three hundred clear nights annually, Mauna Kea is one of the premier locations in the world for astronomy. Background (left to right): Canada France Hawaii Telescope, Gemini North, United Kingdom Infrared Telescope, UH 0.6m telescope. Foreground (left to right): NASA Infrared Telescope Facility, WM Keck I & II, Subaru, James Clerk Maxwell Telescope and a few of SMA's receivers.
I log in at dusk to find that Ramos has already started her shift, narrating her work for the astronomers with commentary. "Adjusting phase ... we are getting good readings, but the weather is slowly turning." I'm brimming with questions about their target for the night, knowing that there is a sophisticated queuing system in place that selects targets according to their importance and position in the sky. Tonight's agenda is a big-sky survey, hunting for filaments of gas crashing into protostars. But then suddenly, it's a wrap.
"We have indications of snow on the summit," Ramos reports. "Humidity is over 90 percent." She knows this due to the real-time reporting from MKWC webcams and weather stations. The arrays have heated surfaces, but there's a limit to how much snow they can melt. She initiates the shutdown sequence. This is not an "L" for MKWC; it was a college try on a marginal evening.
The next week, I log in again because the MKWC forecast is for dry weather with high- speed jet streams above. If anyone can see in those conditions, it's the SMA. Ramos is again on watch. All is well: The array is returning to the aborted observations from the prior week, scanning a wide patch of sky to catch hints of stars being born. I stay up until 4 a.m. to check in with her again. She answers, a little groggy. "The winds gusted to thirty-five mph, but we are still observing," she reports. I ask if we are looking at a rare visitor to our side of the solar system: 12P/Pons-Brooks (a.k.a. the Devil Comet), a long-period comet similar to Halley's Comet. Thanks to a spot-on MKWC forecasted break amid a fortnight of bad weather, Ramos is indeed watching the comet when no one else on Mauna Kea can.
