The Mystery of the Missing Carbon
by David Herring and Robert Kannenberg
April 28, 1999

  Page 2

Boreal Ecosystem Series
· Introduction to BOREAS
· The Mystery of the Missing Carbon
· Should We Talk About the Weather?
· Evolving in the Presence of Fire
· The Migrating Boreal Forest

Scientists estimate that between 1 and 2 billion metric tons of carbon per year are "missing" from the global carbon budget. Or, more precisely, they cannot account for the location of between 15 and 30 percent of the carbon released into the atmosphere each year from fossil fuel burning (Sellers et al. 1997). Worldwide, humans annually release about 7 billion tons of carbon. Of that amount, 3 billion tons remain in the atmosphere, 2 billion tons are absorbed into the ocean, and…the rest? Scientists assume land vegetation absorbs the rest, but they don’t know exactly where or how much.

  boreal forest


The main motivation for studying Earth’s global carbon cycle is to enable scientists to predict future levels of carbon dioxide in the atmosphere. According to Steven C. Wofsy, an environmental scientist at Harvard University, the ability to predict carbon dioxide levels is important if Earth scientists are to answer fundamental questions like how much will global temperatures rise over time, and how will this affect other aspects of Earth’s climate?

"Are those 2 billion tons of carbon missing permanently, or temporarily?" Wofsy asks. "You’re at a loss to predict if you don’t know why the carbon is disappearing and if it will stay gone."


Ecologist Joe Berry checking instruments at the top of a tower in Saskatchewan, Canada. He is surrounded by 120-year-old black spruce, one of the predominant trees in the boreal forest. (Photograph courtesy BOREAS project)

boreal forest map In a concerted effort to solve the mystery of the missing carbon, NASA led an interdisciplinary Boreal Ecosystem-Atmosphere Study (BOREAS) from 1994-97 that spanned two Canadian provinces. Wofsy, along with members of 85 other science teams from five nations, participated in the investigation. Their prime suspect was the boreal forest. Named after Boreas, Greek god of the north wind, boreal refers to the mostly evergreen forest that encircles the Earth at high northern latitudes–between 43°N-65°N–occupying between 16 to 20 million square kilometers of the Earth’s land surface. Could this cold, mostly coniferous ecosystem be the culprit?

next Some Important Clues

The circumpolar range of the boreal forest. From Hare and Ritchie (1972) (Map courtesy BOREAS project)

  The Mystery of the Missing Carbon   Page 1 Page 3

Some Important Clues
For the past 10 years, scientists, using instruments aboard ships that are stationed at different latitudes around the world, measured carbon dioxide variations through the changing seasons. Computer modelers entered these data every month into "tracer models"–general circulation models that track the movement of pockets of air over time. By matching the results of these tracer models to actual ground-based measurements, scientists concluded that most of the atmospheric carbon dioxide is being absorbed somewhere above 40°N latitude in the Northern Hemisphere, which is about the latitude of New York City (Hall 1999).

But, if these tracer models are accurate, then shouldn’t we see obvious places above 40°N latitude where all this carbon is accumulating? Scientists estimate that while the boreal forest occupies about 21 percent of the Earth’s forested land surface and contains 13 percent of all the carbon stored in "biomass" (or living matter), it holds about 43 percent of all of the world’s carbon that is stored in soil (Sellers et al. 1997).

boardwalk"For the past 7000 years, the boreal forest floor has been accumulating carbon at a rate of about 30 grams (or roughly 1 ounce) per square meter per year," observes NASA’s Forrest G. Hall, physicist and BOREAS project scientist. "When you walk through the Canadian boreal forest, you can literally go from ankle deep to in over your head in carbon litter. In some places where there were once holes or ravines, we have measured peat several meters deep."

Thirty grams doesn’t seem like much–it is roughly equal to the amount of carbon contained in a paper napkin. But when you multiply that 30 grams per square meter by the roughly 16 to 20 million square kilometers that comprise the total area of the boreal forest, you find that the boreal ecosystem has been taking in an average of 0.6 billion metric tons of carbon per year for roughly the last 7000 years. But carbon uptake varies from year-to-year. Scientists speculate that the boreal forest could have increased its rate of carbon intake in recent decades.

"The amount of uptake depends upon the type of forest, its location, and its age," Hall explains. "During the BOREAS experiment we saw some kinds of forest (aspen) taking up 200 grams of carbon per square meter, per year."

Mystery solved, right? Not so fast. Although there is plenty of circumstantial evidence, scientists need more data before they can make any firm conclusions. Moreover, they must collect these data over a large region through multiple years if they are to succeed in their mission to understand the role of the boreal forest in the global carbon cycle. Technically, we know that the boreal ecosystem is both a sink ("storage area") and a source of carbon, as it both absorbs and releases carbon dioxide. However, what scientists really want to determine is whether the boreal forest is a net sink or a net source for carbon; and how this determination depends on climate. In other words, after all the additions and subtractions of carbon dioxide, does the boreal forest as a whole retain more carbon than it releases, or vice versa?

"We won't know until we add up the contributions from all the kinds of forest that make up the boreal ecosystem," Hall states. "That will require satellite-derived maps of the entire circumpolar boreal region that enable us to sort out the different types of forests and assess their productivity."

Boardwalks allowed Dave Landis and other scientists to move between the research stations without sinking into the soft, wet soil. For most of the year waterlogged soil and cool temperatures prevent forest detritus — fallen needles, leaves, and branches — from decomposing. (Photograph courtesy BOREAS project)

The BOREAS Measurement Strategy
In planning the BOREAS experiment, researchers had two main objectives: (1) improve their understanding of how the boreal ecosystem exchanges radiant energy (sunlight and heat), water, and carbon with the atmosphere; and (2) develop better computer simulation models that in the future will enable scientists to measure and even predict changes in the boreal ecosystem using satellite and other forms of meteorological and environmental data. In order to meet their objectives, BOREAS scientists selected two large regions, about 500 kilometers apart, in the Canadian provinces of Manitoba and Saskatchewan. Over a period of four years (1994-97), and in each season of the year, they made local-scale measurements on the ground and in towers that stand high above the forest canopy. The BOREAS team complemented these leaf-scale and local-scale measurements with regional- and large-scale measurements made by remote sensors aboard aircraft and satellites. They continually cross-compared data from these various instruments to make sure they agreed.

"I think that our strategy was remarkably successful," Wofsy observes. "The people who organized the BOREAS experiment did a wonderful job." Piers J. Sellers and Forrest Hall, of NASA’s Goddard Space Flight Center, were primarily responsible for the design and oversight of the BOREAS operations.

next Findings from One BOREAS Study Site
back The Mystery of the Missing Carbon

varying measurement scales
(Diagram courtesy BOREAS project)

  The Mystery of the Missing Carbon   Page 2 Page 4
Findings from One BOREAS Study Site
instrument towerIn a stand of 120-year-old black spruce trees in Manitoba, the BOREAS team constructed a tower on which to mount their instruments. They selected this site, as it is typical of much of the North American boreal forest. There are tall (10 meters), densely-packed black spruce trees in the slightly higher, better-drained areas while the spruce trees are shorter (1-to-6 meters) and more sparse in lower and wetter locations. The ground is wet and covered mostly by moss. In order to gain a clear understanding of the forest dynamics there, Wofsy’s team made 22,000 hours of intensive measurements–in the soil, on the ground, in the forest canopy, and even high above the forest–from March 1994 to October 1997.

They found that the growing season begins and ends abruptly with the temperature transitions that mark the onset of spring (usually in May) and fall (usually in October). As the surface layers of soil begin to warm and thaw, the above-ground vegetation begins to take in carbon dioxide during photosynthesis–which is the process by which plants use sunlight as energy to "fuse" carbon dioxide and water molecules into larger, more complex molecules called carbohydrates (plants’ basic building blocks). Wofsy’s team found that from late May through July the old black spruce forest "inhaled" 1-to-1.5 grams of carbon per square meter per day (Goulden et al. 1998). Yet in August and September, the hottest, driest period of the Canadian summer, the rate of carbon dioxide intake fell to almost zero. Then, in October, the forest began to "exhale" carbon back into the atmosphere at a rate of 0.6-to-0.8 grams per square meter per day, which tapered off to 0.2-to-0.3 grams of carbon per square meter per day from December to April (Goulden et al. 1998). Averaged over the course of a year, from October 1994 to October 1995, for that region (within a 500-meter radius of the tower), Wofsy’s team found that the old black spruce stand lost 70 grams of carbon per square meter (Goulden et al. 1998)! The following year it lost 20 grams per square meter.

The BOREAS team used instruments positioned at various heights on towers like this one (located in Manitoba, Canada) to measure the movement of carbon dioxide between the forest and the atmosphere. The towers also measured the transfer of heat and moisture, and housed digital cameras to study cloud coverage and type. They were erected among pine, aspen, and fir trees, and above marshy fens. (Photograph courtesy BOREAS project)

Carbon Uptake


If the boreal forest is a carbon sink, how can it be losing carbon? While it is not possible to extrapolate the measured dynamics of this one old black spruce stand to the entire boreal forest, the BOREAS team’s findings raise some troubling questions.

According to Wofsy, most of the carbon in the boreal ecosystem lies deep in the soil, some 40-to-80 centimeters below the surface–below the depth to which even large forest fires penetrate. The soil at this depth remains frozen most of the year so it cannot decompose. The decomposers–bacteria–are simply dormant. Then, when the soil does thaw toward mid-summer, it is so water-saturated from melting snow that there is not enough oxygen present for the bacteria to thrive. So, when the soil thaws and then begins to dry in late summer, the bacteria wake up and begin furiously decomposing the carbon, thus releasing carbon dioxide.

"We found that what’s going on in the soils is more important than what is going on in the trees," Wofsy explains. "We found that the boreal peat is drying and thawing out. This year, the peat never froze all the way down. If that trend continues, the ‘permafrost’ [or permanently fozen layer of soil from 75-to-100 centimeters beneath the surface] will eventually melt and…the soils will release carbon dioxide faster than the trees can absorb it."

Ironically, Wofsy found that the old black spruce site was a carbon sink in 1998–the warmest year ever on record, a year you would expect the site to be a strong source of carbon. Wofsy explains what happened: "Usually spring comes at once and dries the soil quickly. But in 1998, spring came early and lasted longer than usual."

Wofsy and his colleagues observed that the soils thawed, but remained cool throughout the spring, thus allowing the trees to "inhale" carbon dioxide for a longer than usual period of time. The net result was that from October 1997-98, the old black spruce site took in 40 grams of carbon per square meter (Wofsy 1999).

"This finding speaks to the importance of understanding the hydrology (water processes) of the boreal system," Wofsy notes.


The above chart shows Wofsy's measurements of carbon exchange between the atmosphere and the boreal forest. Positive numbers represent an accumulation of carbon in the forest, while negative numbers show carbon released into the atmosphere. The measurements are totalled over the course of each year, and set to zero each January first.

Although this site loses a modest amount of carbon in average years (1996 & 1997), 1998's long growing season and cool, wet summer resulted in a net gain of carbon. In contrast, 1995 had a hot and dry summer, and more than three times the usual amount of carbon was released. Over the winter the forest loses carbon at a steady rate year after year, which is shown in the graph by the similar slope of each line from October—April. (Graph by Robert Simmon, based on data from Wofsy)

carbon probeHall agrees, stating that if the boreal soils were to dry, this might increase soil carbon decomposition, but the carbon loss might be offset by the intake of carbon by trees as they grow. "This is why we need [better computer] models that incorporate all of these complicated tradeoffs–to accurately predict the effects of climate change on the boreal ecosystem. Precipitation is actually increasing at these latitudes as the climate warms; thus the future may bode a warmer, wetter, and more tropical environment for the boreal ecosystem."

At the old black spruce site in Manitoba, are we already seeing the first sign of the boreal ecosystem’s transition from a net sink to a net source of carbon? In a paper submitted in January 1997 to Science, Wofsy and his colleagues reported that the boreal carbon budget is roughly in balance, but that the boreal ecosystem demonstrates an extreme sensitivity to climate. The net carbon balance at a site is the difference over a year in the amount of carbon dioxide taken up by plant gowth and the carbon dioxide lost through bacterial decomposition of dead limbs, leaves and soil organic matter. For instance, the old black spruce site consistently takes up 800 grams of carbon per square meter per year. But in 1995 it exhaled 890 grams of carbon–a net source of 90 grams per square meter–whereas in 1997 it exhaled 790 grams of carbon–a net source of 10 grams per square meter (Goulden et al. 1998). In short, subtle changes in climate can be the difference between whether the boreal forest is a carbon source or sink; and this difference can show up on a global scale when you consider annual atmospheric carbon dioxide levels.

"If you look at the rate of increase in global atmospheric carbon dioxide, it’s not constant at all," Wofsy observes. "It sometimes varies [from one year to the next] by a factor of 3. Presumably, that gives us a message about the underlying mechanisms for putting away those missing 2 billion tons [of carbon]."

According to Hall, this underscores the importance of the length of the boreal growing season, which also varies from one year to the next. "Climate variation from year-to-year affects the carbon uptake differently than carbon loss," he explains. "Years with warmer springs and falls are better for plant uptake of carbon, but result in increased rates of carbon loss." It is during these latter years that the boreal forest becomes a net source rather than a sink.

Some climatologists argue that a warming trend would simply trigger longer growing seasons and more vigorous plant growth, thereby offsetting the warming trend by enabling plants to absorb more carbon dioxide. Not necessarily so, says Wofsy. "We found no evidence of high rates of carbon accumulation associated with elevated carbon dioxide or climatic warming."

But, Hall counters, Wofsy's black spruce site is old, nutrient poor, and is located near the northern edge of the boreal forest. There may be other regions where the trees are increasingly accumulating carbon due to the lengthening growing season.

next Implications for Global Climate
back Some Important Clues

BOREAS scientists used probes like this one to measure the amount of carbon released from the forest floor back into the atmosphere. They also monitored soil temperature and moisture to gain an understanding of the factors that influence the rate of carbon exchange. Once the processes are understood, computer models can be developed that can predict future change. Note the carpet of moss that insulates the soil from the atmosphere and prevents the soil from drying out. (Photograph courtesy BOREAS project)

  The Mystery of the Missing Carbon   Page 3

Implications for Global Climate
Compared to other ecosystems, the boreal forest responds relatively quickly to changes in climate. Moreover, analysis of meteorological trends from 1961-90 shows that temperature increases are occurring most rapidly over the interiors of continents at high latitudes (43°N to 65°N)–precisely the location of the boreal forest. In the last 40 years, average temperatures have increased over these regions by as much as 1.25°C per decade (Sellers et al. 1997). If the world’s missing carbon is in fact being stored within the boreal forest, and if the present warming trend is turning the boreal forest into a source of carbon, then there could be a significant increase in the rate of carbon build-up in the atmosphere. Scientists fear that by the year 2100 the amount of carbon dioxide in the atmosphere might actually double pre-industrial levels (Wofsy 1999)–to more than 560 parts per million (the 1990 level was 353 parts per million).

In their Science article, Wofsy and his colleagues note that the global mean temperature is predicted to increase by about 2°C by the year 2100. They state that warming of this magnitude would likely completely thaw the deep layers of frozen boreal soil at the old black spruce site and, as they dry, significantly increase the decomposition of the carbon there. If this trend occurs on a large scale across the boreal ecosystem, the decomposing boreal soils could significantly accelerate the rate of rise of carbon dioxide levels in the atmosphere.

Today, Wofsy, Hall, and their BOREAS colleagues can only speculate what will happen to the boreal ecosystem over the next century. But their hope is that as they continue to collect more data and refine their models, they can one day solve the mystery of the missing carbon and accurately predict what future levels of carbon dioxide will be. "How can we manage the forests for economic return and still keep carbon out of the atmosphere?" Wofsy asks rhetorically. "If we are to have any hope of managing the world’s ecosystems more efficiently, we need to understand the system better."

  • References
  • Sellers, P.J., F.G. Hall, R.D. Kelly, A. Black, D. Baldocchi, J. Berry, M. Ryan, K.J. Ranson, P.M. Crill, D.P. Lettenmaier, H. Margolis, J. Cihlar, J. Newcomer, D. Fitzjarrald, P.G. Jarvis, S.T. Gower, D. Halliwell, D. Williams, B. Goodison, D.E. Wickland, and F.E. Guertin, 1997: BOREAS in 1997: Experiment Overview, Scientific Results and Future Directions, Journal of Geophysical Research, 102, pp. 28731-28770.
  • Hall, Forrest G. Personal Interview, 1999.
  • Goulden, M.L., S.C. Wofsy, J.W. Harden, S.E. Trumbore, P.M. Crill, S.T. Gower, T. Fries, B.C. Daube, S.-M. Fan, D.J. Sutton, A. Bazzaz, and J.W. Munger, 1998: Sensitivity of Boreal Forest Carbon Balance to Soil Thaw, Science, 279, pp. 214-217.
  • Wofsy, Steven C. Personal Interview, 1999.

back Findings from One BOREAS Study Site

BOREAS science team
Work on the ground is a necessary component of the scientific process, even at NASA. These researchers are collecting samples of vegetation for the BOREAS archive. (Photograph courtesy BOREAS project)