Can NASA Get Its Satellite Data into the Real World? - Eos

Less than two months ago, the data spigot for NASA’s Ice, Cloud, and Land Elevation Satellite–2 (ICESat-2) finally turned on. It was a long-awaited moment for scientists hoping that the state-of-the-art satellite, which sends out laser pulses and detects returning photons, would provide exquisitely precise measurements of the elevation of Earth’s ice sheets, forests, and land.

Earth scientists won’t be the only ones keeping a close eye on the satellite’s performance. A group of NASA program managers will be scrutinizing not the data themselves, but whether they make their way beyond the scientific literature into practical, public-facing applications such as weather and climate forecasts, disaster prediction and response efforts, and environmental conservation plans.

ICESat-2 will provide a major test of a potentially revolutionary program that NASA is readying for prime time: All missions recommended in the latest decadal survey, which largely sets the agency’s Earth science agenda for the next 10 years, have been directed to be codesigned with nonresearch, practical users at the table from the beginning. That decision, says Stephanie Uz, NASA’s applied sciences manager at its Goddard Space Flight Center (GSFC) in Greenbelt, Md., is “a big paradigm shift for NASA.”

With the Trump administration perpetually threatening NASA’s Earth-observing fleet and budget, including an 8% proposed cut for fiscal year 2020, NASA officials hope the new approach can bolster support for the agency’s nearly $2 billion a year Earth science program.

“We can use demonstrations [of applications] to convince Congress, the administration, and fundamentally the taxpayers to continue to invest in the sorts of endeavors we’re engaged in,” Michael Freilich, director of NASA’s Earth Science Division until February 2019, told Eos.

The effort requires reshaping an agency where scientists and engineers have long led the design of missions and applications have been an afterthought. Codesigning satellites with outside users “is not normal for NASA,” said Vanessa Escobar, who oversees mission design for the applied sciences program at NASA headquarters in Washington, D.C. “This is not the culture.”

The effort has seen both successes and setbacks. Although it has catalyzed some new uses of satellite data, other hoped-for applications have not materialized. Applied science program managers have struggled to get mission designers to make satellites more user friendly or to reach potential stakeholders beyond academia and other federal research agencies.

And it is far from clear whether the increased emphasis on applications will ultimately transform NASA’s culture. Budget constraints and the enthusiasm of the Earth science director who will replace Freilich will play major roles, program managers and outside experts say.

Scientists in the Lead

Like the rest of the space agency, the NASA Earth Science Division has traditionally chosen its missions to answer questions that research scientists identify as important. Since the 1970s, “we always considered applications after the mission was designed for research purposes,” says Lawrence Friedl, director of NASA’s Applied Sciences Program.

Indeed, it’s almost a mantra in the halls of the agency: NASA is a research institution. The laser focus on researchers’ needs has sometimes left potentially valuable applications behind, said Molly Brown, a University of Maryland geographer and former research scientist at GSFC who helped develop the agency’s approach to increasing satellite data applications.

For example, in the 1990s, NASA designed a constellation of Earth-observing satellites called the A-Train, four of which are still active. Their data have been a boon for geoscientists, who have used them to publish thousands of research papers. But the A-Train satellites’ data were initially compiled and released every 8 days, which made the constellation less useful for weather forecasters, who operate on a 10-day cycle, and for everyone on the standard 7-day week, said Brown. It took several years for NASA to begin providing the data more quickly and in more user-friendly formats.

“They designed the perfect system” for improving scientists’ understanding of the planet’s land-ocean-atmosphere interactions, Brown said. But mission designers didn’t consider nonscientists’ needs and constraints. “No thought was given to the use of this data in the real world.”

Two Pushes from Outside

Brown and her applications-focused colleagues have had some help from the outside. In the National Aeronautics and Space Administration Authorization Act of 2005, Congress directed the agency to “research, develop, and demonstrate prototype earth science applications to enhance federal, state, local, and tribal governments’ use of government and commercial remote sensing data, technologies, and other sources of geospatial information for improved decision support to address their needs.”

Then, in its 2007 decadal survey—the first for NASA’s Earth science program—the National Research Council chastised the agency for how little data from its expensive Earth-observing satellites were being applied to real-world problems. Essentially, the committee concluded, NASA was failing to return the full value of its work to the taxpayer.

“We were trying to make the point that application is the twin to the intellectual merit of Earth observations from space,” said Richard Anthes, emeritus president of the University Corporation for Atmospheric Research in Boulder, Colo., who cochaired the report committee.

In turn, the agency told satellite design teams to find potential users for their data and figure out how to better serve them.

Successes and Setbacks with SMAP

The first satellite to sign up was the awkwardly named Soil Moisture Active Passive (SMAP). Launched in early 2015, SMAP was designed to measure Earth’s soil moisture at an unprecedented spatial resolution to examine how more precise information on soil moisture could help monitor drought, predict floods, assist crop productivity, and forecast weather. The mission’s tools and technologies were tailored to help answer those questions, from the methods of collecting data to how fast the data would be cleaned, processed, analyzed, and delivered.

After the design was set, Escobar was tasked with finding “early adopters”—people or teams willing to devote their own time and money to studying how they could integrate SMAP data into their workflows. She met with foundations, water managers, farmers, and even wine growers.

“I lived out of a suitcase for the first two and a half years,” Escobar said. “There wasn’t a meeting I did not attend, to find out, were there people interested in using soil moisture data from a satellite with a resolution of 3 kilometers?”

Escobar initially identified 55 potential practical users. These included the U.S. Department of Agriculture (USDA) for crop forecasts, the National Oceanic and Atmospheric Administration (NOAA) and Environment Canada for improving weather models, researchers from Ohio State University and a power company who thought they might use the data to predict where trees would fall and cause power outages during severe storms, and the insurance company Willis Re for predicting flood risk to insured properties.

But Escobar and her colleagues also hit a major snag: The mission was designed to deliver data 3 weeks after they were gathered. For researchers looking to publish scientific papers, such a delay, or latency, is almost inconsequential. But for people who make decisions based on conditions on the ground, delays of days or even weeks can make the information nearly useless.

Escobar’s team eventually persuaded the mission to deliver faster; the latency is down to 12 hours for some data. The team also pushed for data delivery in simpler formats that users requested and created a listserv through which about 1,000 potential users get regular updates about mission data.

Another disaster hit when, shortly after launch, SMAP’s radar failed, leaving it with only passive instruments. That failure dropped its measurement resolution from 3 to 36 kilometers, said Brown—far too coarse to provide useful information for individual farm fields. Some adopters dropped out.

Still, the program has had successes. In August 2017, NOAA began incorporating SMAP data in its Soil Moisture Operational Products System, which estimates soil moisture globally. Such a quick adoption, less than 3 years after launch, “doesn’t happen in government,” said Escobar.

In 2018, USDA’s Foreign Agricultural Service started using SMAP data, along with other satellite data, to predict crop yields around the globe, a key metric for predicting crop prices as well as hunger and social unrest. “The fact that [the SMAP data] was integrated into the Foreign Ag Service,” said Brown, “is a giant success.” Without the early-adopter program, she says, it might never have happened.

But of the 53 other early adopters, only 1 is known to have put the data into operational use, Chalita Forgotson, who has taken over from Escobar as the mission’s applications lead, confirmed in an email. (The third is a University of New Hampshire–led project to better predict snowmelt flooding using soil moisture conditions prior to freeze-up.) The other 52 either are not using the data operationally or are using it but have not informed NASA.

ICESat-2

The next satellite in line to test the early-adopter program is ICESat-2. As with SMAP, the ICESat-2 design team nailed down its essential features long before applications-focused personnel were brought on board.

Partly as a result, ICESat-2 has saddled its potential users with a 45-day latency. That “really narrowed the field for the kind of users we could have,” said Escobar.

Sabrina Delgado Arias, the NASA program manager tasked with building the ICESat-2 applications program, ultimately recruited 24 prospective adopters.

One of those adopters is the duo of Philip Graea and Mikkel Rasmussen, remote sensing experts at the Danish consulting company DHI GRAS. They plan to use ICESat-2 data from forests to help energy companies site wind turbines. Right now, wind companies use only simple land cover maps to determine whether or not an area is forested, Rasmussen said. ICESat-2 will provide much more detailed information on how trees’ size and structure could affect wind speeds.

“We joined the project because it’s the only source of this information,” he noted.

Other adopters have studied ways to monitor water levels in reservoirs and to assess the amount of burnable fuel on a landscape. Because ICESat-2’s laser operates very differently from prior missions, early adopters say the opportunity to study simulated data has helped them prepare for using real data as soon as it is available.

“Instead of starting from photon zero, we’ll start with a little bit more education and probably advance a little more quickly in terms of being able to use it efficiently,” says Nancy Glenn, a geoscientist at Boise State University in Idaho who is studying the satellite’s ability to map vegetation, with the goal of helping the Department of Defense prioritize environmental conservation dollars on lands it owns in the West.

The program’s true value will not be known, however, until Glenn, Rasmussen, and other early adopters receive the data, integrate them into their planned applications, and report back—a process that could take years, said Delgado Arias.

New Directive, New Directions

Even before it’s clear whether the ICESat-2 early-adopter program has yielded benefits, NASA is giving applications even more prominence in future missions. In 2016, Freilich signed a directive encouraging all future Earth science missions recommended in the then-in-process 2017 decadal survey to be codesigned with nonresearch users at the table from the beginning. The directive affects satellites that will, among other things, measure variables related to air quality, ecosystems, geological hazards, and sea level rise.

Every mission is supposed to start with a “community assessment” so that weather and climate forecasters, land managers, and others can tell mission designers what kinds of data they need, how quickly they need them, and what formats and time intervals they require, Escobar said.

The directive was the culmination of years of discussions and meetings. Because of internal resistance, NASA stopped short of requiring missions to hire applications teams. But NASA is “pushing harder” for their inclusion, Freilich said. “You pretty much have to make the case for why you wouldn’t have one.”

Dan Sarewitz of Arizona State University’s Consortium for Science, Policy and Outcomes in Washington, D.C., a member of the advisory board for NASA’s applied science program, said he views the directive as “an incredible example of impact.” But it is not enough to transform the agency’s culture, Sarewitz added, because applied sciences still get only a tiny fraction of NASA’s overall Earth science budget. “They look at applied sciences as this relative budgetary backwater,” he said.

The directive does not affect missions whose design began before the 2017 decadal survey, some of which are struggling to put together an applications program. Plankton, Aerosol, Cloud, Ocean Ecosystem (PACE), the next free-flying Earth-observing satellite NASA is set to launch, has been unable to hire an applications coordinator because of budget constraints, said applied science manager Uz. That could finally happen later this year, Woody Turner, a program manager at NASA headquarters, wrote in an email.

Future missions will also face the challenge of finding potential users beyond the “usual suspects” within government agencies and academia. NASA’s applied science team held a town hall at AGU’s Fall Meeting 2018 that illustrated this struggle. Escobar, who led the discussion, asked the 50 or so attendees whether any of them were from the private sector. Only one was, and he said he was not a potential end user, but rather a representative of a satellite company hoping to learn what NASA was doing. Around 90% were from academia or the federal government.

Getting user input “has been uphill both ways,” said Brown. “You have to go out and talk to people you do not know, who are not scientists.”

Still, there is evidence that applications’ star is rising. Last August, NASA held its first official community assessment to understand what users might want out of the planned Surface Biology and Geology mission, likely to be the first of the new decadal missions to be designed and launched.

But will the overhauled design process transform the satellites that end up being launched?

“There’s skepticism,” said Uz. “How much this will impact the engineering is yet to be determined. And I think it’s unlikely.”

Brown is more optimistic. “Instead of doing it backwards, we’re going to do it fundamentally correctly from the beginning,” she said. But for all her enthusiasm, she too is hedging her bets. “I don’t know if we’re going to succeed.”

—Gabriel Popkin (@GabrielPopkin), Science Writer

23 July 2019: This story has been updated to clarify that ICESat-2 is successfully delivering data

Let's block ads! (Why?)