What Climate Scientists Want from Satellite Data Providers
Climate technology investment is booming. Scientists have more access than ever to data acquired by space-based, remote-sensing technologies to help monitor and mitigate the effects of climate change, but need to access it faster, and make sure it can be translated into actionable policy.
September 27, 2022
The expansion of satellite data from new providers and increased remote sensing technologies is helping climate scientists fill in critical gaps in government-based coverage, and helping to bring the big picture of climate change to life.
Climate technology investment is booming, with companies raising over $50 billion from 2019-2020, according to a 2021 Space Capital report. Earth Science continues to be best monitored from space, the report says, and that investment is fueling the emergence of remote sensing companies that are focused on climate efforts such as emissions monitoring, or providing their existing services for climate change tracking and mitigation efforts.
“With a growing satellite data infrastructure, entrepreneurs can focus on building highly specialized applications without having to develop their own hardware,” the report said. “As advancements in space-based technology make access to climate-relevant data more widely available, early warning systems will help to detect and manage risk.”
The ways in which government agencies employ satellite-derived data for climate change-related programs are manifold. Scientists from the U.S. Agency for International Development (USAID), for example, use satellite imagery to map potential sea level rise in Senegal, and combine data on precipitation and flooding in Malawi with household survey information to track which populations will be most affected by flash floods. Meanwhile, the State Department is using space-gathered data to study soil content in the Horn of Africa – where scientists may not have field access – to help guide climate policy together with its partners in the region.
The combination of government and commercially derived remote sensing data is helping scientists fill in coverage gaps, develop new climate mitigation techniques, and more deeply monitor methane and greenhouse gas emissions. But those in the field say there’s more to be done: The vast swaths of raw data they now have access to need to be translated into actionable information at a faster pace and in a digestible manner, so policymakers can create more effective policies to combat climate change.
How Government Agencies Use Satellite Data
Since the 1960s, NASA, the U.S. Geological Survey, and the Weather Bureau – now called the National Oceanic and Atmospheric Administration (NOAA) – have worked together to conduct Earth science research, with NASA developing observational systems like the Landsat constellation to acquire space-based land remote sensing data. As the Landsat constellation remains operational to this day, it has provided the world’s longest continually acquired collection of space-based data, per the Space Capital report.
Since the Landsat 1 launch in 1972, climate scientists have been able to build a picture of climate change using a variety of remote sensing data like electro-optical (EO) imagery and synthetic aperture radar (SAR). NOAA’s National Environmental Satellite, Data, and Information Service (NESDIS) has been developing pilot programs to study how radio occultation (RO) techniques can help monitor climate changes. Radio occultation measures properties like atmospheric density from space, for example by calculating the angle of a GPS radio signal bending as it travels through the Earth’s atmosphere.
The Commercial Data Program currently acquires RO technology from providers Spire Global and GeoOptics for weather and climate applications, says Mitchell Goldberg, NESDIS chief scientist. The resulting RO data are used by operational weather forecast centers across the world and for monitoring climate change. Currently, the administration uses a combination of government- and commercial-provided RO measurements to better study atmospheric changes, and the future mix of government and commercial RO is being discussed, Goldberg says.
Traditionally, NOAA has used data acquired via other U.S. government space systems, like NASA, but also satellites operated by allies such as the European Space Agency (ESA), or the Japanese Aerospace Exploration Agency (JAXA), Goldberg says. “Now, the next group are the commercial vendors, which have very attractive, very capable systems,” he says.
Goldberg says he’s looking for more ways that NOAA entities can better harness satellite data. He chairs the administration’s Satellite User Needs Working Group, which conducts periodic surveys that help uncover data gaps in current programs of record, and recently launched a study assessing all available commercial satellite data providers and how their services could support NOAA line offices.
“The commercial sector is really exploding, and so we’re going through the process within NOAA, to determine what type of data we really need for our stakeholders,” Goldberg says. Broadly, the working group sees commercial providers supplementing government-built data collection, to furnish higher tempo resolution and retrieve the data more quickly, he says.
NASA is also exploring the potential of radio occultation data. The agency started looking at this “interesting new data set” several years ago, says Will McCarty, program scientist at NASA’s science mission directorate.
McCarty became the initial project scientist on the Commercial Smallsat Data Acquisition (CSDA) program, which formally launched in 2020 after a successful pilot, and now explores the viability of new satellite data types, and to create new procedures for NASA to work with commercial data vendors. The CSDA effort is exploring a wide swath of remote sensing data options, from EO/IR, to SAR, to RO. At the moment, the program office is working with companies Teledyne Brown Engineering, Maxar Technologies, Planet, and Spire Global, and has ongoing or planned evaluations with Airbus, BlackSky, GHGSat, GeoOptics, Iceye US, and Capella Space, according to McCarty.
The idea is not for commercial satellite systems to replace NASA’s flagship birds, which can provide more accurate measurements than smaller satellites, but to fill in gaps, he notes.
“What these small satellites and constellations provide us that the big satellites can't, is the ability to measure more frequently,” he says. That helps climate and weather scientists observe short-term impacts of climate change, such as landslides or floods, on top of long-term evolutions that are monitored by government satellites.
“The supplementary data is providing extra information; it’s not replacing the information content that our existing satellites are already targeting,” says McCarty.
Increased access to satellite data will help scientists study both climate monitoring as well as mitigation of the effects of climate change, notes Goldberg. Amid the abnormal heat waves in Europe, flash floods and landslides in Pakistan, and a historic drought in China, more Earth observation is needed not only to track the effects of climate change, but to provide more early warning and allow populations to prepare adequately.
“That’s where the commercial industry can help us quite a bit,” says Goldberg.
The public is expecting more rapid responses to extreme weather events, and for that to happen, NOAA needs more predictable, faster weather forecasts.
“If there's a heavy precipitation event taking place, you want to know with high precision and higher accuracy, to be able to forecast from minutes well into days [or] one week, two weeks,” he adds. That could require access to a new satellite constellation that can provide more frequent observations.
For climate monitoring, NESDIS will need less frequent observations, but higher quality and longer continuity constellations, “so you can track over decades, into the future,” Goldberg says.
As NESDIS continues its survey of available satellite data providers, several opportunities stand out to Goldberg. Constellations that could detect fires from Low-Earth Orbit (LEO) would be very interesting, he notes. “If you imagine LEO constellations providing 30-minute tempo resolution at 20-meter resolution, so you can really see fire perimeter lines, we would definitely be very interested in that.” Data stemming from SAR sensors would also be helpful to better monitor sea ice levels and flooding situations, and scientists can also derive ocean winds from that technology, he notes.
But the degree to which NESDIS can harness that commercially derived data hinges on its budget, Mitchell adds. The Commercial Data Program to acquire RO data is funded, but for any new commercial data opportunities, “we will have to do that analysis and then we'll go through the process of submitting budget requests,” he says.
The Promise of Satellite Data for Methane Tracking
Methane tracking is one area that has proven more elusive for climate scientists in the past, but new technologies and processes are now putting it within reach.
Hyperspectral imagery has many scientists excited about new ways to track methane, as well as greenhouse gas emissions. Earth imaging company Planet has teamed up with NASA’s Jet Propulsion Laboratory to fly hyperspectral imaging sensors in a new constellation called Tanager to map emissions from methane, carbon dioxide, and other environmental indicators. Open data pools provided by public-private venture Carbon Mapper and the California Air Resources Board will make the evidence freely available, per Planet.
Meanwhile, companies like Montreal-based GHGSat are providing new data to climate scientists with its ability to measure high-resolution methane from space. Scientists could previously measure methane on the spatial scales of kilometers; GHGSat can measure it on the order of 30 meters, NASA’s McCarty says.
“They're actually able to see, potentially, individual plumes from individual sources, while our larger satellites can really only see the broad global coverage of what the methane fields look like,” he adds.
NOAA plans to work with the commercial airline industry to put new sensors on aircraft that can track atmospheric data as planes ascend and descend, Ariel Stein, acting director of the administration’s Global Monitoring Laboratory and director of the Air Resources Laboratory, recently told U.S. lawmakers. The U.S. government needs both global measurements and modeling, but also local measurements to properly track methane and other greenhouse gasses, Stein said. In situ data from satellites and information gathered from airborne platforms will be critical. Each environment – from an industrial area, to an urban city, to a forest – will also require a different sensor or solution set, he said.
While federal agencies have made strides in greenhouse gas measurement, there are still some knowledge gaps to overcome, said Bryan Hubbel, national program director for air, climate, and energy at the Office of Research and Development at the U.S. Environmental Protection Agency (EPA).
“This work includes resolving uncertainties in various airborne and satellite systems, establishing approaches to capture information about source locations and operational status, and identifying locations and events that can account for disproportionate amounts of greenhouse gas emissions,” Hubbel told the House subcommittees.
Making Satellite Data Actionable
After decades of building bespoke space systems to gather imagery, now, “we are being chased by data,” said Robert Cardillo, chief strategist and board chairman at Planet Federal, and former director of the National Geospatial-Intelligence Agency (NGA).
But the explosion of information gathered by remote sensing technologies brings with it the increased likelihood of false data being introduced to the stream – whether by accident or for nefarious reasons, he shared during a recent Planet-sponsored webinar. Geospatial assurance is now a critical element to keep in mind, especially when the data sources are not known or validated, he added.
Climate scientists will take all the data they can get, but in order to move from observing raw images to developing policy, they need a way to translate the information into understandable facts, while also taking bias into account.
“It’s about bringing together the data and models that we have, but then also creating the interface that makes that easy to access,” said Karen St. Germain, Earth Science division director at NASA’s Science Mission Directorate. The goal is to make the data “easy to access and understand by users at all levels, to inform the decisions they have before them,” she told the House subcommittees.
Pilot projects are one way to help stakeholders understand the information derived from satellite data. Government officials cited the way the U.S. and its allies and partners have harnessed commercially derived satellite data to assess the ongoing war in Ukraine, and how that data has helped inform issues surrounding food and water insecurity in other nations.
The U.S. State Department has developed a pilot project along those lines, said the department’s climate change specialist Chelsea Cervantes de Blois. She called it “an excellent opportunity” to develop toolkits that harness commercially derived satellite data that supports global climate change efforts. “It's quite exciting to see that we are using such tools and such imagery, to improve and design toolkits that we haven't had before,” she said during the Planet webinar.
The ubiquitous use of satellite data is a big bonus for climate scientists. As more remote sensing providers come online, their data may have scientific use even if that’s not their primary purpose. “Something that makes one person's data noisy or is a source of error, may actually be a very useful piece of information to someone else,” says McCarty.
But data volume remains a challenge, and the climate science community needs to converge and exploit both the traditional Earth science element – including chemistry, physics, and meteorology – and data science. New technologies like hyperspectral imagery have a lot of promise, but come with their own challenges – hyperspectral imagery comprises 10 to 100 times more raw data than electro-optical imagery, says McCarty.
“How do you handle that?” he says. “And then how do you expect an Earth system scientist, who's not a computer scientist, to handle that data as optimally as a computer scientist – who could probably handle the bits and bytes, but not understand the physics behind it?”
Not every climate scientist is adept at using data processing tools, and not every data scientist can interpret the information they receive into meaningful scientific findings. “We’re going to have so much data that doing things by hand physically is a challenge, but the data science tools will be a challenge to adapt to,” says McCarty. VS
Vivienne Machi is an award- winning reporter based in Stuttgart, Germany.