News Story

May 25, 2017

Earth Science’s Big Picture


By Alvin Powell

“Wow,” Elsie Sunderland recalls thinking, “that’s what I need to be doing.”

Sunderland’s imagination had been caught by a satellite that trained its instruments not on the stars, but back on Earth. She had been working at the Environmental Protection Agency in Washington, D.C. and listening to an MIT scientist talk about atmospheric chemistry and the broader insights gained by coupling his detailed, well-understood local measurements with the regional and global data streaming back from Earth-observing satellites. 

That years-ago flash of inspiration is bearing fruit in Sunderland’s current career as an academic scientist. Now the Cabot Associate Professor of Environmental Science and Engineering at Harvard, Sunderland recently found that Canada’s plans to develop more hydropower, however well-intentioned and climate-friendly, could wind up poisoning native populations with mercury. 

“Satellite data is essential to making any kind of spatial interpretation,” Sunderland said. “Typically, our approach is to obtain field measurements, which allow you to say something about the area you studied … then use satellite data to extend that over larger spatial scales. Our approach using this method is not perfect, but [it shows] when we should be concerned about human health impacts.”

Sunderland isn’t alone in her inspiration. All around the world, scientists like Sunderland rely on a steady stream of data from an array of large-scale Earth-observing systems. Highest profile may be the suite of orbiting satellites, but measurements are also being taken from ships at sea, by floating and moored buoys, through instruments mounted on buildings, atop towers, or buried in the earth, even floating on balloons or soaring aboard planes and drones.

These large-scale Earth-observing systems are expensive and logistically challenging to build and deploy. More than a century ago, the British Royal Navy led early efforts to understand the ocean, providing the foundation upon which modern oceanographers understand the seas today. And, though technology has come a long way, the logistical and financial challenges of such large-scale measurements mean the world’s governments remain important sponsors of such networks, and the U.S. is the largest among them. 

A 2014 report by the U.S. Office of Science and Technology Policy identified 325 different observing systems spanning 11 federal departments. It defined 145 of those systems as “high impact” and identified benefits to society in everything from agriculture to transportation, weather forecasting to climate change, and human health to the environment.

The programs cost $2.5 billion for satellite-based observations and another $1 billion for Earth-observing systems on land, sea and air. But that expenditure pays for itself many times over, the report said, estimating that the programs add $30 billion to the U.S. economy annually and are “vital national infrastructure.”

“The satellite observations have become such a mainstay, and we rely so much on satellite measurements in understanding the Earth system, it’s essential,” said Frank Keutsch, Stonington Professor of Engineering and Atmospheric Science, who uses satellite observations in his work on atmospheric chemistry. “People think it’s just chemistry and climate, and that’s not true. Think about weather and weather predictions, we need a variety of information for that. Satellite instruments focused on the oceans are also a growing need.”

John Holdren, the Heinz Professor of Environmental Policy at the Harvard Kennedy School and Professor of Environmental Science and Policy in the Department of Earth and Planetary Sciences, served until January as Assistant to President Obama for Science and Technology and as Director of the Office of Science and Technology Policy. While in the administration, Holdren worked to boost the priority of such Earth-observing systems in the NASA budget, and helped untangle conflicting priorities between the National Oceanic and Atmospheric Administration (NOAA)—responsible for weather forecasting—and the Defense Department around weather satellites in polar orbit.

Weather forecasting is an obvious priority for Earth-observing satellites, Holdren said, and the ability to forecast weather more than 48 hours in advance depends on a small group of satellites in polar orbit, maintained and operated by NOAA. Short-term weather forecasts, convenient in deciding whether to grab a coat or umbrella during ordinary times, are potentially life-saving when the weather turns severe, while medium-range forecasts are essential for farmers considering what and when to plant. Longer-range trends—an area that needs improvement—are important in understanding climate change.

In describing the scope of Earth observations today, Holdren rapidly reeled off a long list, including vegetation and land use change, deforestation and reforestation, fisheries and shellfish production, agricultural production, natural disasters, and the Earth’s water cycle. 

“We’re using sophisticated gravitational instruments to understand aquifers,” Holdren said. “Some people think water is the next energy in terms of potential for crisis regionally.”

This Earth-observing network is at once robust and providing more data than ever before, but also fragile, reliant on aging platforms, the advocacy of influential individuals and groups, and on fickle, cash-strapped governments to continue their operation and finance replacements. 

All that has scientists casting worried looks at Washington, D.C. The politicization of science in general—and climate science in particular—has created the prospect of government becoming hostile not only to economically costly solutions, but also to the very information that underlies the science. In March, Mick Mulvaney, the budget director for President Donald Trump—who’s called climate change “a hoax”—said simply of climate change: “We’re not spending money on that anymore.” 

The president’s proposed 2018 NASA budget, which would go into effect on Oct. 1, would cut $108 million for Earth science and end four missions in development, including OCO-3, which would make measurements of atmospheric carbon dioxide, and CLARREO Pathfinder, which would test technologies to measure the Earth’s energy absorption and radiation for use in climate science.  

Those cancellations still need Congressional approval to take effect, and Congress sent a hopeful signal in late April, when it reached a bipartisan budget agreement to keep the government operating for the remainder of the 2017 fiscal year, which ends Sept. 30. 

That agreement, which President Trump is expected to sign, would spare key parts of the Earth-observing budget. It would boost NASA’s budget by 1.9 percent over 2016 and continue funding for NASA Earth science programs at last year’s levels. And, while the NOAA budget would decrease overall by 1 percent, its office that supports climate change research around the country would see a 3.5 percent increase, to $478 million.

Even given the administration’s opposition, Steven Wofsy, Rotch Professor of Atmospheric and Environmental Science, said the fragile state of Earth-observing systems today can’t be laid solely at the door of the current administration. Earth science has always had to scratch and claw for its piece of the budgetary pie, a task made more difficult for satellite-mounted instruments, Wofsy said, because NASA is viewed by many as an exploration agency, defined more by moon landings and missions to Mars than measurements of the planet that we live on and are, in many ways, intimately familiar with.

Daniel Schrag, Director of the Harvard University Center for the Environment, the Hooper Professor of Geology, and Professor of Environmental Science and Engineering, said that tightening NASA’s focus in a way that excluded Earth observing would be devastating to efforts to understand the planet in a time of change.

“There’s some concern now that NASA’s mission is going to be narrowed,” Schrag said. “That would be catastrophic.”

Even if the budgets for major Earth-observing programs survive in the new administration, the initiatives can be harmed by a lack of direction from the top, said Schrag, who served on the President’s Council of Advisors on Science and Technology under President Obama. Obama, Schrag said, made it clear that Earth-observing was a priority, which sent a signal to the many agencies involved that their cooperation and coordination was important.

More, not less

These doubts about the future of Earth-observing systems come at a time when those studying the Earth and human impacts on it say the need for information is greater than ever before. Though we know more about the Earth than at any other time in human history, there are also more humans than ever before, crowding the globe and appropriating more and more of the Earth’s resources. To better manage human impacts, to devise strategies for conservation of natural resources and of biodiversity, and to create systems of all sorts that are environmentally sustainable, we need more measurements, more observations, better algorithms, more computing power, and more understanding.

“We have to [keep] what we have, and we need more,” Sunderland said. “We really do have big gaps in observations that we’re making. What we want to do is build on what we’ve done successfully and [cover] the rest of the world…. You just wouldn’t be able to begin unless you had these satellite data products.”

To Holdren, the need is not only to expand observations to places where they’re not, but to get better observations of places at which we’re already looking. 

“Coverage is important, geographically. The density of observations is important, the continuity of observations is important, resolution is important when talking about satellite imagery, depending on application,” said Holdren.

That means that we need to not only fund these programs, but seek partnerships with other governments to help fill the holes that remain, Holdren said.

“All of that means international collaboration is important because if you’re interested in Earth observations you need ground truthing: ocean-based, land-based, atmosphere-based measurements, in addition to satellite measurements. In order to get the coverage … you need international collaboration,” Holdren said. “Even in terms of space-based [observations], international collaboration is important. The budget for Earth observations—instruments and satellites—is limited and the capabilities of other countries complement ours.”

Should such Earth-focused data gathering be interrupted—whether by a hostile political climate or more mundane federal budget austerity—researchers say the loss of continuity in measurements actually represents a dual loss. Not only is the data left uncollected lost, but also lost is the continuous trend over time, which allows better interpretation of the data at hand. 

“Without continuity, if you just came and went every once in a while for a short time period, if you saw a change, you can’t tell if it’s a fluke or a trend,” said J. William Munger, Senior Research Fellow in Atmospheric Chemistry and principal investigator for a long-running project to measure exchanges of gases, such as climate-warming carbon dioxide, between the atmosphere and Petersham-based Harvard Forest. “Because there’s so much variability in climate and the forests’ response to it, without long records you don’t have the trend.”

Carl Wunsch, Visiting Professor of Physical Oceanography and Climate in Harvard’s Department of Earth and Planetary Sciences and Emeritus Professor of Physical Oceanography at MIT, said that continuity is nearly as important as the data itself. 

“Gaps are a real problem, we have a lot of gapped records. They’re extremely hard to use because we don’t know what happened in between,” Wunsch said. “What if there was a big ENSO (El Niño) in there and we see a lot of rainfall, and then it’s gapped again because a mooring broke or a satellite failed. These are hard scientific problems that require a long-range view, and we don’t have the infrastructure for coping with them. Understanding rainfall variations without a 50-year record means that you are mainly dealing with weather noise. If a theory is delayed by five years, it’s still an appropriate theory, but if nobody measures the deep Indian Ocean in the 2010s and in 2030 somebody wants to know what was going on in the deep Indian Ocean in 2017, it’s gone forever.” 

In an ideal world, not only would there be no gaps in data, new instruments would overlap in time with old ones, Wofsy said. That would allow researchers to compare new and old measurements to understand how data gathered from different instruments should be adjusted so they are truly comparable and researchers don’t inadvertently compare apples to oranges. 

The fight to ensure that data gathering is continuous is an old one, Wunsch said, and involves some of the best known research of the modern era. Charles Keeling, whose observations of atmospheric carbon dioxide from the top of a Hawaiian volcano allowed him to draw the now famous Keeling Curve illustrating rising atmospheric carbon dioxide concentrations, regularly faced funding problems. Keeling, Wunsch said, was able to overcome them with a personality that Wunsch described as a “tour-de-force … obnoxious enough to keep it going.” While climate science benefitted from Keeling’s don’t-take-no-for-an-answer attitude, Wunsch said that not all those studying important questions can replicate his doggedness.  

“It’s basic science, which you should do even though you have no idea what the outcome could be,” Wunsch said. “The first geologist couldn’t have predicted the basic science would result in the mining industry as we know it. There are endless examples where basic science, unplanned, turned into something extremely valuable.”

Making the models

One important use of Earth observations is to build better computer models of complex physical processes. 

Kaighin McColl, a Ziff Environmental Fellow at the Harvard University Center for the Environment, is working with McKay Professor of Atmospheric and Environmental Science Zhiming Khuang to better understand interactions between the water cycle on land and in the atmosphere.

McColl is relying on soil moisture measurements from NASA’s SMAP (Soil Moisture Active Passive) satellite to help with this work. Because the interactions between moisture in the air and on land are poorly understood, many models split the two apart and deal with them separately, McColl said. Surface hydrologists oversimplify how water behaves in the atmosphere, and those studying the atmosphere oversimplify the water cycle on land. That’s a problem, McColl said, because it has become clear that feedbacks exist between the air and land that affect the behavior of each.

“The way the land and atmosphere behave when coupled to each other is different from how they behave alone,” McColl said. “In models, you can get these feedbacks between land and atmosphere that won’t happen if they’re modelled separately.”

Getting the coupled model right is important, McColl said, because soil moisture can have an impact on rainfall. All else being equal, solar radiation first goes to work evaporating water in the soil. If the soil is wet, moisture evaporates into the atmosphere, cooling the air and forming clouds to produce rain. Dry soil, by contrast, results in the air heating up more quickly, which further dries the land, increasing the chance a drought will persist.

“You get a feedback loop that can alter the persistence of droughts, floods, etc. That’s what we can see that we won’t if we don’t study it as a coupled system,” McColl said.

The oversimplification of the land-atmosphere feedback is present in climate models as well, McColl said, and, because this feedback operates most strongly on a regional scale, it’s thought to be a factor in the poor performance of climate models as they zoom in. 

“Climate models are very good at looking at global mean properties, but if you want to know how temperature in Boston will vary over time, climate models struggle with regional estimates,” McColl said. 

The trouble is partly due to the complexity and messiness of the land surface, with mountains and buildings and forests and fields, McColl said. Each of those affect surface fluxes differently and add enough complexity that it has only been recently that enough computing power has become available to model it, McColl said. 

“The atmosphere that we live in—the bottom 100 meters—is highly turbulent. You can think of the air as a fluid with lots of whirling motions of different sizes mixing it up a lot,” McColl said. “The soil moisture changes the heat pumped into the atmosphere, which changes the turbulence too. It’s hard to model that.”

With the help of SMAP, which can measure moisture in the top five centimeters of soil over 40 square kilometers, McColl is zeroing in on convection over land, the relatively small-scale interaction of air, land and moisture that is the driving force behind summer thunderstorms. 

“I think they’re essential,” McColl said of the satellite observations. “We’ve been stonewalled for a long time, struggling to make progress without them.” 

Think globally, research locally

Since Sunderland left the EPA for Harvard, she’s been investigating how pollutants travel through the environment and wind up in the human body. She’s studying PCBs, which persist in the environment despite being banned, selenium, and an organic form of mercury, called methylmercury, that can impair the development of the brain in fetuses and has been traced to cardiovascular problems in adults. 

She’s also studying highly florinated compounds, or PFASs, whose properties as a surfactant have made them widely used since the 1950s. Today, they’re found in everything from rain gear to dental floss to food packaging. More recently, they’ve also been found concentrated in the flesh of Arctic polar bears, thousands of miles from the nearest factory; in the drinking water of 6 million Americans at levels exceeding federal guidelines; and at detectable levels in 98 percent of the rest of us. 

PFASs have a potent impact on the immune system, according to a Faroe Islands study, Sunderland said, with each doubling of their concentration in islanders correlating to a 50 percent decline in immune function.

Her 2016 study of methylmercury in Canadian hydropower development shows how Earth-observing data can be coupled with intensive, localized study to draw broader conclusions. Methylmercury forms naturally through the action of microbes when soil containing mercury and methane is flooded. The new compound is transported through the food web, where it is concentrated as it moves from prey to predator to larger predator. This poses a potential hazard to Inuit communities that eat a lot of fish and live near the hydropower sites.

Sunderland and colleagues examined native communities and their environment near a facility that is scheduled to be completed in 2017. They measured individuals’ current mercury exposure and modeled their exposure to methylmercury once the project is complete, finding that mean exposure could double, and that concentrations in women of childbearing age and in young children would exceed U.S. EPA guidelines. Using Earth-observing data of 21 other proposed hydropower sites, they then forecast Inuit exposures, finding that methylmercury concentrations at 11 would be comparable or greater than the original site. 

“We used satellite data on soil carbon reservoirs to extend that to planned expansion sites, across the country, to understand the potential human health impacts of hydroelectric power development in those different regions,” Sunderland said.  

A similar approach has resulted in a better understanding of the exchange of climate-warming carbon dioxide and other gases between the world’s forests and the atmosphere. 

In 1991, Harvard’s Wofsy erected a thin metal tower at Harvard Forest. The tower was laden with instruments to measure how much carbon dioxide was taken in and given off by the forest. The work turned into the first long-term such project in the world, and has helped fill in the picture of how trees, shrubs, soil and other components of the forest contribute to the makeup of the atmosphere.

“The original goal was to be the first ones to actually measure, at the scale of a whole forest, the net exchange of CO2 up and down,” Wofsy said. “Nobody had done these measurements before. We had a device at the top of the tower that measures wind speed 10 times a second, and that measures CO2 concentrations at about the same [interval].”

The project was launched as awareness and concern about climate change was rising, and—because living trees lock up carbon in their wood while dead ones give it off as they decompose—there were questions about what role mid-latitude forests were playing in the broader climate change picture.

In 1993, Wofsy reported that Harvard Forest, though it appeared mature, was still growing, absorbing two tons of carbon dioxide per hectare each year. Munger, who is principal investigator for the project today, said the work has thrown into question whether such forests ever reach a steady, “mature” state of development.

“We thought the forest at some point would become carbon neutral, and we changed our ideas about when a forest gets to that point,” Munger said. “When I look at data I don’t see any signs it’s approaching a steady state. It may be that it will keep on growing and accumulate carbon in trees and deadwood and keep doing it until some disturbance comes.”

In the years since, other researchers have erected similar “eddy flux towers” around the world. But as many sites as there are, Wofsy said their coverage isn’t broad enough to be able to make general statements about the world’s forests without large-scale Earth observations.

“You can’t characterize the world’s forests with eddy flux sites,” Wofsy said. “You can try to get as representative a set of measurements as you can, then combine them with remote sensing that covers the whole globe and really understand what’s going on.”

Wofsy uses several satellites in his work, including OCO-2, which measures total column carbon dioxide in the atmosphere and solar-induced florescence in trees’ leaves, an indication of the activity going on inside the plant. Others include Landsat, which measures land use change, and data from the MODIS instrument, flown aboard two satellites, which measures greenness at high temporal and spatial resolution, Wofsy said.

“They synergize,” Wofsy said of the tower and satellite measurements. “The whole is greater than the sum of the parts no matter which you’re looking at.”

A research continuum

To Keutsch, Stonington Professor of Engineering and Atmospheric Sciences, large-scale Earth observations are an irreplaceable part of a research continuum that extends from a laboratory’s tightly controlled experiments, through local instrument observations, to atmospheric satellite measurements. 

Keutsch’s work on atmospheric chemistry has revealed something of a bright spot in efforts to lessen humankind’s impact on the environment. It shows that over the U.S. Southeast, the atmosphere has returned to pre-industrial conditions in at least one respect.

His work explores different atmospheric “regimes” that favor the conversion of methane to either formaldehyde or hydroperoxide. Formaldehyde regimes, he said, are found in industrial air and promoted by the release of nitrogen oxides in fossil fuel burning. His research group is exploring the tipping points from one regime to another and the human influences that push the atmosphere toward those tipping points.

A change in air over the U.S. has occurred over the last 20 years, Keutsch said. Before that, the air was clearly in the formaldehyde regime, but parts of the country have been getting cleaner, in particular the U.S. Southeast, Keutsch said. Because the conversion of methane to formaldehyde occurs on a much shorter time scale than carbon dioxide’s cycling, which takes millennia to clear from the atmosphere, the change is observable today. 

“It really is remarkable,” Keutsch said, “and a direct impact of the reduction in nitrogen oxides from power-generating industries.”

Keutsch’s research occurs over a range of scales. He conducts tightly controlled laboratory experiments that allow researchers to zoom in on chemical reactions they believe are occurring in the atmosphere. They also take localized measurements from places like Michigan, the Amazon, and the Antarctic, where a South Korean icebreaker is sailing with a formaldehyde detector aboard.

They also use large-scale measurements from aircraft and satellites to see what’s going on in the atmosphere and broaden the geographic range. 

“The question is how well does the model predict aircraft measurements and then satellite measurements,” Keutsch said. “We do lab studies on one component in great detail, and we can understand everything very well because we can control it well. We take the mechanism and put it in the model, then compare it to the real atmosphere and see if we understand the atmosphere. You have to have these field measurements of this system.

“Regional variability we can only get by satellite…The critical thing in understanding the system is we have observations from very local to satellite observations. These satellite observations are very important for this because the idea that we can make these point measurements everywhere, that is just not feasible.”  

Water and earth

Not all of the large, Earth-observing networks look down from space. Some of the oldest examine the seas, according to Carl Wunsch. And, though our understanding of the ocean as a large, turbulent body may seem obvious today, it wasn’t always so. The shift in our thinking about the ocean is an illustration of the power of Earth observations to transform our understanding. 

As recently as the late 1970s and early 1980s, Wunsch said, detailed knowledge of the oceans was lacking. Unlike the atmosphere, the ocean is hard to see through, so its depths were still largely unknown. 

Oceanic surveys began in the 1870s, and scientists aboard oceangoing vessels were able to discern deep layers formed by variations in temperature and salinity, Wunsch said.

Oceanographers understood that this static picture was incomplete, but their work was limited by the difficulty and expense of their primary tool: ship-board surveys. Additional observations gathered as commercial mariners traversed the world’s oceans helped, but Wunsch pointed out they tend to avoid some places of obvious scientific interest, like hurricanes and the high latitudes in winter. 

“People began to understand that [the ocean] is changing, but didn’t know what to do with that insight. To some extent it was ignored. A way had to be found to leave instruments at sea,” Wunsch said. “It awaited the integrated circuit.”

In the 1970s, technology permitted the development of self-contained instruments that could be set afloat. Understanding began to grow of the ocean as a swirling turbulent system “much like the atmosphere but without weather,” Wunsch said.

Despite that early work, climate models still lacked a realistic depiction of the oceans.

“Meteorologists were making climate models in which the ocean was treated as a swamp—immobile and doesn’t do anything—because that’s what they’d been taught,” Wunsch said. “So you start to think about how the devil do you measure a global fluid, because climate is global and we knew from these ancient charts that it was interconnected. What happened in the North Atlantic has consequences for the deep Pacific Ocean decades or centuries later.”  

Oceanography verged on irrelevancy, Wunsch said. That spurred planning for a major international effort, the World Ocean Circulation Experiment, in which Wunsch played a leading role. Years in the making, it ran from 1990 to 2002 and employed everything from traditional ship-based observations to instruments drifting on floats, moored on buoys, or borne aloft on satellites. What emerged from that effort was a new understanding of the ocean as a dynamic part of the Earth-ocean-atmosphere system.

Today, Wunsch said, our knowledge of the oceans is on par with that of the atmosphere, and scientists have a better understanding of the challenges they face in making additional observations. 

“This is an extraordinarily complicated observational problem and an extraordinarily complicated theoretical problem to understand it. But without the observations, you’re sort of dead,” Wunsch said. 

The latest generation of ocean observing floats, called Argo, represent an evolution of the technology used in the World Ocean Circulation Experiment. Some 3,800 advanced, self-contained Argo floats have been deployed in the world’s oceans, Wunsch said. They descend to a “parking level” 1,000 or 2,000 meters down, take measurements as they drift and then rise every 10 days and send their location and other data to a satellite overhead. 

“At any given time, we have measurements of 3,000 plus of these things, all over the ocean,” Wunsch said. “In some sense it’s been revolutionary, because we’ve finally broken away from the ship as the major instrument of in situ measurements.”

Wunsch said the Argo program allows expensive ship-based observations to be deployed where they’re most needed. Instead of being the primary way to gather data, they can fill in where the Argo floats cannot, like collecting data from the deep ocean beyond the depths Argo floats are designed to go.

“If you’re interested in climate, you must know what the deep ocean is doing and has been doing. Developments are underway to take Argo to the bottom,” Wunsch said. 

Schrag and graduate student Lauren Kuntz are among those very much interested in climate. Argo data from the upper ocean is helping them formulate a new theory that explains the seeming periodicity in global temperature rise. 

“You need to know more, and observations are really the only way you can get that level of detail and get that level of understanding,” Kuntz said. “I think everyone would agree that we want to know where we’ve been and where we’re going, in terms of climate.”

Measurements of global temperatures have marked several periods of relatively rapid warming, separated by periods of relative stability. Temperatures remained fairly stable from the mid-1940s to the mid-1970s, then rose sharply through the late 1990s, when they stabilized again until a couple of years ago. 

“The question is why, that is what we’re trying to understand,” Schrag said.

The dominant theory, Schrag said, is that atmospheric particles called aerosols, generated in volcanic explosions and industrial emissions, reflected enough solar radiation to interrupt warming during those periods. But Schrag and Kuntz aren’t so sure. They’re looking at an entirely different process.

When Schrag arrived at Harvard in 1998, he wrote a paper about the possibility that the ocean waters in the tropical Pacific played a role in the oscillations in global temperature change but, he says today, the data back then were so sparse, it was little more than speculation.

Now, using data from the Argo floats, Schrag and Kuntz are tracking fluctuations in the thickness of layers in the tropical Pacific made up of water that is a constant density—called “isopycnal layers” that they believe are linked to the “fits and starts” of global temperature rise.

“We had no idea it existed there, but what we’ve observed is that the upper layers of the ocean are changing their thickness … in a pattern that’s led us to new insights,” Schrag said. “The theory was required to explain the observations that we made from Argo. It was Argo that alerted us to this going on.”

Another scientist eagerly examining Argo data is Sunderland, who wants to see how it can benefit her work on pollutants moving through the environment. She’s seeking clues about how pollutants move from the Atlantic to the Arctic Oceans and whether that is changing as seas warm and currents shift.

“I think this is going to be very exciting,” Sunderland said. “We’re borrowing the approach from Earth sciences. We’re using the same tools, assimilated Earth observations, to drive our understanding of how the compounds distribute in the global environment and then how they accumulate in food webs.”

Open science, open data

In addition to the steady stream of new observations, these observing networks also archive data that can be accessed by researchers who are exploring similar questions or historical trends. Marine Denolle, Assistant Professor of Earth and Planetary Sciences at Harvard, is exploring the nature of earthquakes and how seismic waves travel through the ground. To conduct her work, she deploys her own networks of seismometers, but also accesses data from past research held by the National Science Foundation (NSF). 

“The NSF funds a lot of data collection. There’s an incredible amount of data already collected that is terribly underused,” Denolle said. “There’s terabytes and terabytes of seismometer data unused.” 

Denolle’s research seeks to understand quake-induced ground motion in major cities, with the aim of one day predicting how the ground might shift in a quake. Understanding that movement would allow scientists to predict damage to buildings and, ultimately, inform changes to building codes to make buildings safer. 

Denolle is also examining the interaction of earthquakes and groundwater to understand how water affects the movement of seismic waves. It’s already apparent, she said, that seismic waves travel at different speeds through dry ground than through saturated ground and aquifers. That means depleting aquifers, as occurred during California’s recent five-year drought, alters how an earthquake might be felt and could mean quake predictions need to be updated.

“You’ll find wave speed variation highly correlates with variation of groundwater level or with precipitation level,” Denolle said. 

Observing the behavior of seismic waves could also be used to answer questions about aquifers and their ability to recharge when the rains return. In California’s heavily agricultural Central Valley, so much groundwater has been withdrawn that the ground has sunk as a consequence. 

“The question is how elastic aquifers are, that’s really important for water regulations,” Denolle said. 

Looking ahead

With climate change one of the world’s most pressing scientific concerns, it’s perhaps not surprising that Holdren listed climate change-related measurements—the concentrations of all atmospheric greenhouse gases, as well as temperature measurements in the air, on land, and in the oceans—among those most important to continue.

Observations focused on the Arctic are also important, Holdren said, because climate change is occurring two to four times faster there than elsewhere. In addition, the permafrost holds twice as much carbon as the atmosphere, releasing it as it thaws. Carbon is also released from the shallow water sediments in the increasingly ice-free Arctic Ocean, he said. 

“This is potentially a powerful feedback that could accelerate climate change,” Holdren said. “That has serious consequences for everybody.” 

Another important priority, Holdren said, is improving local and regional climate forecasting, which would let people, businesses and communities plan how to mitigate coming climate impacts. 

“We still need to know more,” Holdren said. “In particular, we need better observations to get more accurate regional disaggregation of ways that the climate is changing and ways it’s likely to change going forward so communities, farmers, fisherman can plan and make investment in preparedness, resilience and adaptation. Climate is a global system, but all of its impacts are local.”

Beyond pressing, short-term concerns lie broader issues related to the mismatch between the time scales of scientific research and the processes that we’re trying to understand, Wunsch said. Many key processes involved in climate science occur over decades, centuries, even millennia, while human scientists’ funding is measured in years. 

“In a system [such as the oceans] that we know is responding to events thousands of years ago, the instrumental record is decades old, at best,” Wunsch said. “I date it to 1992—call it 30 years. Anything older than that, we don’t know whether you’re seeing [a response due to] something from long ago or due to modern atmospheric climate. There just aren’t enough observations. You look at the problem of climate and say the need is for records that are long enough to make sense of them, and not just in the ocean.”  

The scientific establishment, Wunsch said, needs to think hard about the best way to study such long-term phenomena, which he termed intergenerational scientific problems.

“If I wrote a proposal to NSF saying I need 25 years of funding to get longer records to really say something, I’d be laughed out of the room,” Wunsch said. “Nobody funds that sort of thing. That’s a real problem.” 

One positive trend, Holdren said, is the development of a new generation of small, inexpensive satellites. These “small sats,” are enabled by the abundance of off-the-shelf parts that are driving costs down and make it increasingly possible for other players to step in—like wealthy individuals, foundations and even states.

“We’re better and better at miniaturization now,” Holdren said. “Companies make small sats using cellphone parts and other commercially available components. It’s possible, with the help of lowered costs, you’ll see some states getting involved—Governor Brown said California will put up its own—as Earth observing gets cheaper.”

Though we don’t understand precisely how the world is changing, the fact that we know it is and that some of those changes will impact people, perhaps dramatically, argues that Earth observing should continue, Wunsch said.

“We don’t have so many years [of data that] we can afford to give anything up,” Wunsch said. “At the end of the day, the basic science of the Earth is too important to let it lapse for one or 10 years.”

This article originally appeared in Environment@Harvard: Volume 9, Issue 1

Harvard University
Center for the Environment

Address: 26 Oxford Street, 4th Floor, Cambridge
Phone: (617) 495-0368

Connect with us

Follow HUCE to stay updated on energy and the environment at Harvard and beyond.

Subscribe to our mailing list