On 7 September, the microwave sensor on the Sumi National Polar-orbiting Partnership satellite captured heavy rainfall rates within hurricanes Irma (left) and Jose (right). 


Fragile polar weather satellite system could be bolstered by microwave-sensing CubeSats

The polar weather satellite system is safe, for now. After years of rising costs and delays, the $1.6 billion Joint Polar Satellite System-1 (JPSS-1) rocketed into orbit on 18 November from Vandenberg Air Force Base in California. The launch will still fears that a failure of the JPSS-1's aging predecessor would cripple the armada of polar satellites that provides 85% of the input data for weather forecast models. "This data is so important for forecasting that you can't take those chances," says Mitch Goldberg, program scientist for the JPSS at the National Oceanic and Atmospheric Administration (NOAA) in Greenbelt, Maryland.

Hitchhiking on board the same rocket that carried the 4-meter-tall JPSS into orbit was a strikingly smaller and cheaper probe that is a harbinger of a yet more resilient weather satellite system. The $3 million Microwave Radiometer Technology Acceleration (MiRaTA)—classed as a CubeSat because its components are stuffed into a stack of three 10-centimeter cubes—carries a cloud-penetrating microwave sensor that rivals one on the much bigger and costlier JPSS-1. There is a "kind of David and Goliath thing," says Kerri Cahoy, MiRaTA's principal investigator and an engineer at the Massachusetts Institute of Technology (MIT) in Cambridge. "There will be a lot of interesting opportunities for comparison."

Though MiRaTA will not come close to replacing the JPSS-1, which carries four other instruments besides a microwave sensor, Cahoy says it foreshadows a day when weather data come from constellations of small, cheap, and replaceable satellites rather than from big battleships like the JPSS-1. "If these missions are successful then people will say, ‘Hmm, this is a different way to do this,’" she says. "Maybe it's supplemental for a while, and maybe they meet in the middle."

Weather satellites come in two main flavors. Geostationary satellites are parked 36,000 kilometers out, where their orbital speed matches Earth's spin, so that they can stare down at one region and watch weather unfold. Polar satellites orbit at altitudes of only hundreds of kilometers and build a global view by tracing swath after swath along lines of longitude, as Earth spins. With the launch of the JPSS-1, the National Weather Service will be ingesting data from nine major polar satellites that cover every spot on Earth several times a day.

From their closer vantages, polar satellites can measure the faint infrared and microwave emissions from oxygen and water vapor molecules in the atmosphere, indicating temperature and humidity at different altitudes. The infrared detectors have much better vertical resolution. But microwaves have their own advantage: They pass through clouds. That's especially valuable for looking into storms at sea, where no ground-based radars exist.

The JPSS-1's microwave instrument is a technological step up from those on older polar satellites. It will help scientists map polar sea ice and measure intense bands of rain within hurricanes. Its temperature and humidity data will also provide a stronger baseline for measuring climate change. "We'll have another time record that we'll have more confidence in that will tell us how temperature has changed," Goldberg says.

Through the clouds

On the $1.6 billion Joint Polar Satellite System-1 (JPSS-1), the Advanced Technology Microwave Sounder (ATMS) peers beneath clouds to measure temperature and humidity data crucial for weather forecasts. The $3 million Microwave Radiometer Technology Acceleration (MiRaTA) aims to gather similar data in a smaller package.


That confidence did not come cheap. At one point, NOAA, NASA, and the Department of Defense each weighed in on designs for what would become the JPSS, tacking on requirements. "Everything in the kitchen sink was thrown into the solution," says Dave Powner, an official at the U.S. Government Accountability Office in Denver who has led audits of the program since 2004. The total budget for the program—including the JPSS-1; its precursor, the Suomi National Polar-orbiting Partnership (NPP), launched in 2011; and the JPSS-2, a clone due for launch in 2021—ballooned to $11.3 billion. Poor interagency communication slowed development, Powner says. Despite delays, the JPSS-1 was set to slide into orbit this week while Suomi NPP is still working.

MiRaTA is an attempt to show that the JPSS's microwave measurements can be done with a smaller package. It's the brainchild of Bill Blackwell, an engineer at MIT's Lincoln Laboratory in Lexington, who came up with the idea on a cocktail napkin about a decade ago during a lunch at MIT's faculty club. He realized that a 10-centimeter microwave antenna—the maximum possible size on a CubeSat—would provide enough resolution to see into the eyes of hurricanes, and that the data could be transmitted to Earth with a low-bandwidth CubeSat radio. A colleague tweaked the design of a microwave receiver used for astronomy, and Blackwell bore down on miniaturizing the electronics needed to process the signals.

There have been teething pains. An earlier version, launched in 2014, made it to orbit, but its communication radio antenna got jammed, dooming the mission. But the low cost of a CubeSat means that failure is a chance to learn and try again. Blackwell plans to launch two more CubeSats in the coming months. Unlike MiRaTA, which keeps its antenna fixed on the planet, these CubeSats will alternately point their antennas at Earth and space as the head of the satellite spins 30 times a minute. That will allow the sensor to be calibrated against the background cold of the cosmos—a closer match to how the JPSS-1's microwave sensor scans Earth.

The agencies that built the JPSS-1 are starting to take small satellites seriously. In 2016, NASA selected Blackwell to lead a $30 million mission that would fly six of his CubeSats in equatorial orbits as early as 2020. That would increase the revisit rate for observing quickly evolving tropical storms from several times a day with the polar constellation to once every 40 to 50 minutes, says Scott Braun, the project scientist for the mission at NASA's Goddard Space Flight Center in Greenbelt. "You can get that high temporal resolution to know what's happening in between the looks from the bigger satellites."

Eventually, small weather satellites could replace the behemoths altogether. NOAA is considering missions called Earth Observation Nanosatellites—bulked-up CubeSats that would observe in the microwave and infrared and would be designed to match the JPSS data as well as possible. Christian Kummerow, an atmospheric scientist at Colorado State University in Fort Collins and an expert in microwave sounding, says the field is at a "crossroads." "We could see a future where we change the paradigm from large, core observatories to constellations of smaller ones," he says. "NOAA is clearly looking at that with an open mind."