1998 was an extreme year for wildfire activity throughout the North American and Russian boreal forests. More than 11 million hectares (110,000 square km) burned that year (Kasischke et al 1999). Examination of NOAA Advanced Very High Resolution Radiometer (AVHRR) and Landsat satellite images over the world’s high northern latitude forests reveals a heavy peppering of burn scars across the boreal landscape. Closer examination of those image data reveals the boreal forest canopy to be a patchy mosaic of splotches where there are various stages of plant regrowth in the wake of earlier fires.
 

Coming Soon:
· The Migrating Boreal Forest

Although most people regard fire as a destructive force that should be fought and quickly extinguished, the fact is the boreal forest evolved in the presence of fire and adapted to it. Forrest Hall says it’s not a question of if a given region of the boreal forest will burn, it’s a question of when. Hall, a physicist at NASA’s Goddard Space Flight Center, explains that wildfire is an integral part of the boreal ecosystem. Indeed, the high northern latitude forests would be quite different were it not for frequent fires (Hall 1999).
 

"Fire is the mechanism by which the forest is continually regenerated," states Hall. Fires consume dead, decaying vegetation accumulating on the forest floor, thereby clearing the way for new growth. Some species, such as the jack pine, even rely on fire to spread their seeds. The jack pine produces "seratonous" (resin-filled) cones that are very durable. The cones remain dormant until a fire occurs and melts the resin. Then the cones pop open and the seeds fall or blow out.

"In the Canadian boreal forest, aspen and jack pine are the most important ‘pioneer species’," observes Hall. "They are usually the first to grow back in a region that has been affected by fire. A few years after the fire, you typically see either dog-hair thick aspen stands or knee-high young jack pine trees sprouting all over the place."

Then comes secondary growth of the other tree species common to the boreal region—spruce or fir trees. Once spruce trees reach the same height as the other species, they become competitively superior to the other trees. Spruce have better access to sunlight while shading the other trees with their more dense canopies. Consequently, the jack pine and aspen begin dying out in regions where spruce trees are plentiful. But, should another fire come and disturb this cycle, the more fire-tolerant jack pine and aspen regain the advantage and again proliferate. (At far northern latitudes and in most of the Siberian boreal forest—depending upon such variables as soil type and wetness, topography, and local climate—species such as jack pine don't fair well and the regeneration species are mostly shrubs and black spruce.)

The amount of "standing biomass" (plants and trees) continually increases for 140 to 200 years after a fire; then the amount of standing biomass stabilizes and decreases as mature trees begin to die off. As plants die, fall to the ground and decay, the amount of carbon stored in the soil and ground moss rises over centuries to millennia. In this way, over geologic time, huge amounts of carbon have accumulated in the world’s boreal forest floor—about 37 percent of the carbon stored on land, and about 15 percent of the world’s total carbon content (Kasischke et al 1995).

A Burning Issue
In the Russian boreal forest, most fires are ignited by lightning strikes hitting trees or the ground. In North America, about 58 percent of the wildfires are caused by humans, while the rest are caused by lightning (Turner 1999). In relatively dry summer seasons, like 1998, thousands of fires are started by lightning, consuming millions of acres of boreal forest lands. Most of these fires not only consume trees, but also burn the top layers of soil, thereby releasing the carbon stored there too.

Of concern to Earth scientists is the impact widespread boreal fires have on climate. Fires release huge amounts of smoke, carbon dioxide, and methane into the atmosphere. These are greenhouse gases that act like "insulation" in the atmosphere to help trap and retain heat emitted from the surface. Some scientists are concerned that extreme fire seasons in the boreal forest, like last year’s, may contribute to global warming. Others point out that when a region is disturbed by fire, there is vigorous regrowth that could offset global warming. The rationale is that while the younger plants are growing back they are absorbing carbon dioxide back out of the atmosphere and using the carbon to build plant structures.

Over the long term, does fire render the boreal ecosystem a "source" or "sink" of carbon?

The data used in this study are available in one or more of NASA's Earth Science Data Centers.

In some areas of the Canadian boreal forest, jack pine is the first species to appear. They are dependent on fire to open their tightly-sealed pinecones. Aspen, mixed with the jack pine in this photo, are another pioneer species.

With their taller, thicker canopies, black spruce gain a competitive advantage over the shorter aspen and jack pine. Deprived of sunlight in the shade of mature black spruce, these latter trees begin to die off.

The final stage of succession is a mature forest of black spruce. Over time dropped needles, dead branches, and fallen trees will build up on the forest floor, decaying only slowly because of the cold climate. When enough fuel accumulates, lightning will ignite another fire, resetting the cycle. (Photographs courtesy Lou Steyaert, NASA Goddard Space Flight Center/United States Geological Survey)

From 1993-97, teams of scientists from all over the world participated in the NASA-sponsored Boreal Ecosystem-Atmosphere Study (BOREAS) to examine the physical and chemical interactions that occur between the boreal forest and the lower atmosphere. Among the primary objectives of the experiment were: (1) to improve our understanding of the mechanisms by which carbon is exchanged, and (2) to map the types and geographical distribution of plant species across the Canadian boreal landscape. Ultimately, participating scientists hope the new BOREAS data will help them improve their computer models of the boreal ecosystem so that they can better predict how climate change is likely to affect the northern forests in the future; and how changes in the forest may in turn impact climate.

According to Eric Kasischke, a fire scientist at the Environmental Research Institute of Michigan, satellite remote sensing systems are the key to understanding the role of fire in the boreal forest. He says satellites can tell scientists how fire transforms the land surface as well as enable them to measure how the transformation alters the ecosystem’s biophysical processes—the mechanisms and rates at which energy and trace gases are exchanged with the atmosphere.

Lou Steyaert, a remote sensing scientist for the U.S. Geological Survey, participated in BOREAS from the earliest planning stages. One of his assignments was to use multi-temporal satellite image data to help the teams decide where to go make measurements. ("Multi-temporal" means multiple data acquisitions were made over time. The BOREAS team collected satellite images over the Canadian boreal region throughout the growing seasons over four consecutive years.)

"The BOREAS modelers told us what forest types they were looking for so if given a regional land cover map they would know the forest composition and distribution, hence they would know where to go make the measurements they considered important to run their models," Steyaert recalls. "Using AVHRR and Landsat images offered the quickest, most cost-effective way of surveying a million square kilometers to understand and map the boreal forest composition in the BOREAS study areas."

Steyaert completed his initial analysis of the 1-km-resolution AVHRR data in 1993. As he and his colleagues reviewed the satellite images, they found the boreal forest canopy to contain a complex mosaic of landscape patches of various tree types at widely ranging stages of growth. "The first time I looked at the multi-temporal AVHRR satellite analysis over the BOREAS study areas, I had no idea what it [the patchy appearance of the canopy] meant," he said. "The importance of fire in the boreal region was well known going in. However, what had never been done before 1990 was obtaining a synoptic (large-scale) multi-spectral view from satellite remote sensors of not just recent burns, but historical burns as well."

The BOREAS team didn't succeed in mapping the forest (using 1-km AVHRR data) and validating its map until 1996. They could not be sure their satellite map was correct until after they collected ground observations with Global Positioning System (GPS) along several thousands of kilometers of roads, and conducted more than five sets of low-level aircraft flights to make measurements in remote areas. Then the team used high-resolution Landsat data to scale up from the BOREAS study areas to the entire North American boreal region.
 

Because they make measurements at multiple wavelengths of the electromagnetic spectrum, satellite sensors can help scientists identify various stages of regrowth in the forest. The pink area in this image is a recent burn scar in the Canadian boreal forest.

The light green patch shown above is an area of regrowth surrounded by older (dark green) forest.

Roughly 30 years after a fire, forest cover appears uniform in satellite imagery. The variation here is due to topography and microclimate, not fire.

The above series of images was taken by taken by the Landsat Thematic Mapper, and is a composite of visible and near-infrared wavelengths. full image (Images courtesy Dave Knapp, NASA Goddard Space Flight Center)

The BOREAS team found that land cover across the boreal landscape was more widely-varied than previously thought. Using multi-spectral and multi-temporal remote sensing data proved to be a great way to visualize and map out the regional differences. "Our study shows the extensive heterogeneity in the land cover types as a result of natural wildfire, and the regeneration of vegetation that is a function of the date of the burn," Steyaert explained. "We then deduced that the age of the vegetation—hence biophysical characteristics—is probably a very important factor in determining the land surface fluxes of water, energy and carbon for process and modeling studies."

According to Steyaert, the satellite data show that about 30 percent of the BOREAS study areas burned within the last 35 years. (After that amount of time, the newer vegetation starts to mature and blend with the surrounding land cover so it becomes more difficult to spot burned areas from satellites.) The fires were most likely started by lightning, and were typically 20-25 km in diameter within the Canadian Shield Zone.

The "Shield Zone" refers to a geographic boundary above 52°N latitude where the land surface was more noticeably impacted by receding glaciers in the wake of the Earth’s last ice age (producing more small lakes and rock outcrops). In general, Canadian firefighters more aggressively monitor and fight fires south of this boundary; particularly in regions near urban populations. In multispectral AVHRR and Landsat images, you can clearly see the boundary of the Canadian Shield, north of which the landscape is obviously more varied. The BOREAS team found that at latitudes above the Shield, the forest is about 80 percent coniferous and 20 percent deciduous; below the Shield the reverse is true—20 percent coniferous and 80 percent deciduous. Moreover, above the Shield, more fires are simply allowed to run their course, therefore there are more recent burn scars that span larger areas, so there is much more new vegetation growth.

"All these extensive new growth patches from old fires in the northern side of the shield are probably a good sink for carbon," Steyaert surmised. But what about south of the Shield—is it a sink or source of carbon in the presence of fire? And, perhaps more importantly, what about the Eurasian boreal ecosystem? Roughly two-thirds of the world’s boreal forest spans from Europe to Siberia, where fire fighting is known to be a much bigger problem. Historically, the data regarding the total area burned in the Russian boreal ecosystem are very poor.

This map—derived from NOAA Advanced Very High Resolution Radiometer (AVHRR) data from one year—shows the distribution of different landcover types in the Canadian boreal forest. The redish areas show how widespread fire is in this ecosystem. This image shows clearly the Canadian Shield boundary passing from upper left toward the lower righthand corner of this image. Notice that above the Shield there are more recent burn scars over larger regions of the forest.full image (Image courtesy Lou Steyaert, NASA Goddard Space Flight Center)

Boreal forests, or northern evergreen forests that encircle the Earth above 48°N latitude, are second in areal extent only to the world’s tropical forests and they occupy about 21 percent of the world’s forested land surface (Sellers et al 1997). Together, the North American and Eurasian boreal ecosystems span about 14.3 million square km (Kasischke et al 1995). Hence, fighting fires across these vast landscapes is a big problem, particularly for the Canadian and Russian governments. It simply isn’t economically feasible to fight them all, and most fires are allowed to burn themselves out. Quite often, firefighters in those countries aren’t even aware of wildfires in remote regions until after they have been burning for days. Given their size and remoteness, satellite remote sensors are the only cost effective way of monitoring fires in the boreal regions.

Firebombers
Canada is highly dependent upon the forest for both recreation and commerce. One in every 17 Canadians works in the forestry industry, earning an annual $8 billion in wages (Turner 1999). Today, Canada is the world’s largest exporter of lumber and paper products (Turner 1999). Consequently, millions of dollars are spent each year on managing and protecting the trees there (Turner 1999).
 

In the 1940s, Canadian firefighters began using airplanes—called "fire bombers"—to combat wildfires. Initially, they used latex-lined bags to make large "water bombs" that firefighters would drop through holes in the cabin onto their targets. By 1950s, the firebombers had been modified to allow them to scoop up water from nearby lakes so they could fly as many sorties over a wildfire as fuel permitted. While the use of aircraft greatly improved the speed and efficiency with which firefighters could extinguish wildfires, ironically, they helped create a new problem in some relatively small regions near parks and urban centers. Fire suppression allowed unusually large amounts of fuel (dead vegetation) to accumulate on the ground in those regions, thus creating a greater fire danger.

Paradoxically, over the last 25 years, forest resource managers have begun to fight fire with fire. Increasingly, they are using fire as a tool to prevent wildfire—a technique called "prescribed burning." These small-scale fires both help clear away dead or dying vegetation for new growth as well as reduce the risk of larger, uncontrolled outbreaks later. In 1998, Canadian fire officials burned about 16,000 hectares (160 square km) in prescribed fires (Turner 1999). Additionally, Canadian officials allow some wildfires to burn in a controlled manner in areas where there is heavy fuel accumulation.

Some Burning Issues
Due to cost and manpower constraints, Canadian forestry managers must decide which fires to fight and which to let burn. An average of 9,000 fires burn each year, consuming roughly 3 million hectares (300,000 square km) annually. While only 2 to 3 percent of these wildfires grow larger than 200 hectares (2 square km) in size, those 2 to 3 percent of fires account for about 97 percent of the total burned in Canada.

In Russia, however, the boreal wildfire problem is both more severe and poorly documented. The Russian boreal region is larger than its Canadian counterpart, thus there is a greater challenge of fighting fires across a wider frontier. Consequently, the Russian government only actively suppresses fires in roughly two-thirds of its boreal region, typically leaving the rest to burn (Kasischke et al 1999). Moreover, scientists who monitor fires on a global scale suspect that the Russians greatly underestimate the total area burned on an annual basis due to the financial incentives that the government provides to those regions that report good success at fire suppression (Kasischke et al 1999).
 

Based upon analysis of AVHRR data, fire scientists estimate that roughly 12 million hectares (120,000 square km) of Eurasian boreal forest burned in 1987; and most of that was in areas actively monitored by Russian foresters (Kasischke et al 1999). This figure compares dramatically with the estimate of 1.27 million hectares burned reported by Russian officials (Kasischke et al 1999).

Another point of contention between fire scientists is the type of fires that occur in the Russian boreal forests (Kasischke et al 1999). The Russians report that more than 95 percent of all fires burn at the surface; whereas North American foresters report that more than 90 percent of the fires in Alaska and Canada are "crown fires," or fires that burn up to the top of the forest canopy (Kasischke et al 1999). Fire scientists suspect that the Russian reports are inaccurate and that the relative number of crown fires is similar to the proportion they see in North America. Why is this significant? Because the type of fire is suggestive of its intensity and the efficiency with which it consumes biomass and releases emission products (smoke and gases).

Surface fires generally consume less fuel—8 to 12 tons of biomass per hectare burned—and release lower amounts of smoke particles and greenhouse gases into the air (Kasischke et al 1999). Crown fires burn hotter, consume much more fuel—30 to 40 tons of biomass per hectare burned—and tend to release emission products much higher into the atmosphere where they are usually spread over much wider distances (Kasischke et al 1999). Emission products injected at higher levels in the atmosphere tend to remain in the air longer and have a greater effect on air quality in the surrounding region.

Which fire estimates for the Russian boreal region are most accurate? Today, most fire scientists believe that the numbers derived from AVHRR and Landsat images are more accurate because when they compare images of fires in North America with images of fires in Russia, the spectral signatures (thermal infrared radiant energy) match closely (Kasischke et al 1999). Moreover, scientists point out, if the dense forest canopy was undamaged, then it would effectively mask the thermal energy from fires confined to the surface, so the satellites wouldn’t see them (Kasischke et al 1999).

According to Forrest Hall, boreal fires can affect climate in two broad ways: (1) by changing the carbon balance, and (2) by changing Earth’s radiant energy balance. The boreal carbon cycle is regulated by four processes: (1) the rate of plant growth, which determines how much and how fast plants absorb carbon dioxide from the air during photosynthesis; (2) the rate of decomposition of dead biomass, which releases carbon dioxide back into the atmosphere; (3) the rate of formation of frozen soil (called "permafrost"), which prevents the organic matter in the soil from decomposing; and (4) the frequency and intensity of fires, which release carbon, methane, and aerosol particles into the atmosphere. (Kasischke et al 1995)

Tipping the Carbon Balance
Over geologic time, as tree litter fell to the forest floor, boreal soils became a carbon-rich sink. An estimated 231 petagrams (231 billion metric tons, or 43 percent of the world’s total) of carbon is stored in boreal soils, while another 58 petagrams (58 billion metric tons, or 13 percent of the world’s total) of carbon is stored in the live vegetation (Kasischke et al 1995). (For more details, please see "The Mystery of the Missing Carbon.")

"For the past 7,000 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 Hall. "When you walk through the boreal forest, you can literally go from ankle deep to in over your head in carbon litter."
 

Hall shares the concern of many Earth scientists that if the frequency and areal extent of wildfire in the boreal forests should increase, then that ecosystem could shift from being a net sink to a net source of carbon. In turn, this new net source of carbon released into the atmosphere could significantly contribute to global warming. Of particular concern to scientists are the fires that get so hot they burn deeply into the soil, releasing the carbon stored there. According to Hall, it only takes a change in the storage of roughly 50 grams of carbon per square meter per year to release a billion tons of carbon.

"All long-term carbon is stored in the ground moss, duff (decayed organic matter in the soil), and root production," Hall explains. "When the moss and duff are wet, they act as a surface fire retardant. But when they are dry, they catch fire easily; the moss acting much like a mattress fire, smoking and smoldering as it spreads and then flaming up when it hits a tree. In a drought year, the boreal forest floor dries quickly—within a few weeks—down to the water table (about 1 meter deep). When the moss and duff burns, it produces this unbearably thick, dingy smoke."

In addition to fire’s direct effects of releasing carbon dioxide, Kasischke points to another, longer term effect fire has on the carbon cycle. "Organic soils and mosses conduct heat five to ten times less efficiently than mineral soils," he explains. "When those layers are consumed by fire it’s like removing an insulating blanket, so the efficiency of heat transfer improves tremendously."

In the wake of fire there is no longer a forest canopy to shade and cool the surface. Uninsulated by moss and organic layers of topsoil, the boreal forest floor warms and dries more readily in the arid summer air, thus accelerating the rate of decomposition. The boreal soil is converted into a longer-term (beyond the immediate effects of the fire) source of carbon until, decades later, the forest regenerates there and begins to revert back into a carbon sink.

Some scientists counter that although increased fire activity would release more carbon into the atmosphere, the affect on climate would only be a short-term one. They argue that the presence of fire stimulates vigorous new plant growth and, therefore, in the longer term the boreal forest would become a more efficient carbon sink than it is now. (This debate is further addressed on the next page.)
 

Unbalancing the Radiation Budget
In addition to affecting the carbon cycle, wildfire also impacts the boreal region’s radiant energy budget. Clearing away the forest canopy dramatically changes the albedo (amount of reflected sunlight) for a region, particularly in the fall, winter, and spring when there is snow on the ground. (For more details, please see "Should We Talk About the Weather?") The BOREAS team demonstrated that changing albedo over such a large region has a noticeable effect on weather.

Snow on the ground reflects about 80 percent of the sun’s light and absorbs 20 percent (Betts 1999). The more sunlight that is reflected back up into the atmosphere, the cooler the surface temperatures. In contrast, conifer trees (e.g. spruce and pine) reflect only about 10 percent of the sun’s light and absorb the rest, which warms the surface and, in turn, the lower atmosphere (Betts 1999). These changes also influence wind currents, the formation of clouds, and precipitation patterns across the boreal region.

Moreover, due to the smoke particles released, fire has an added cooling effect. Aerosol particles tend to cool the area beneath them by reflecting and scattering incoming sunlight, as well as by promoting increased cloud formation. Overall, however, because greenhouse gases remain in the atmosphere much longer (years to decades) than aerosols (days to weeks), scientists think that the warming effect of fires will have a greater effect over the long term.

The Intergovernmental Panel on Climate Change (IPCC) concluded recently that "the observed increase in global mean temperature over the last century (0.3-0.6°C) is unlikely to be entirely due to natural causes, and that a pattern of climate response to human activities is identifiable in the climatological record" (IPCC 1995). The warming trend over the last 100 years has been most dramatic at high northern latitudes, resulting in temperature increases of 2-3°C in the world’s boreal regions (Stocks et al 1998). Moreover, many General Circulation Models predict that the rate of global temperature increase will accelerate over the next century, by anywhere from 0.8 to 3.5°C warmer than today (Stocks et al 1998). Such a rate of change would be unprecedented in our planet’s recent history, going back millennia.
 

"One speculation is that as it gets warmer, the boreal forest will get drier," stated Hall. "If this happens, you could see an increase in fire frequency."

Another hypothesis suggests that as the atmosphere warms, it will hold more moisture and precipitation will increase, thereby reducing fire frequency. But to investigate this alternative hypothesis, scientists need better models of the interactions between the land surface and atmosphere. But it seems unlikely that the increase in precipitation would be enough to eliminate the problem. Kasischke points out that, despite appearances, the boreal ecosystem is actually a very dry environment—BOREAS investigators dubbed it the "green desert." He says the Alaskan boreal forest interior only receives about 25 cm (10 inches) of precipitation per year.

Some fire scientists use computer models to help them understand and predict how climate change is likely to affect the frequency and extent of wildfires in the next century. (For background information, please see "Modeling the Land Biosphere.") They constructed a hypothetical scenario in which the levels of carbon dioxide in the atmosphere doubled pre-industrial levels of the greenhouse gas over 100 years ago. The models produced some alarming results. In that scenario, the danger of wildfire outbreak across the North American boreal forest would increase by 50 percent, the length of the fire season would increase by 30 days, and the frequency of lightning strikes—one of the foremost causes of boreal fires—would increase significantly (Stocks et al 1998).

According to AVHRR data, 1987 and 1998 were severe fire years in North America. Not coincidentally, those were also severe drought years. Scientists estimate that in the 1970s, roughly 1.5 million hectares of boreal forest burned annually in North America, while today an annual average of 3.2 million hectares burns. But in years like 1998, upwards of 15 to 20 million hectares (150,000 to 200,000 square km) of boreal forest may burn (Kasischke et al 1999). These statistics, together with the fact that the areal extent of burned regions has been steadily increasing in recent decades, paint an ominous picture for a future that could include global warming.

"When we compared AVHRR satellite images of Russian and North American boreal fires," Kasischke states, "we noticed they have an 'episodic nature'. They are not uniformly distributed (over time or space). Most of the burning tends to occur in severe years, like we saw last year.

"Looking at fire activity in the North American boreal forest over the last 30 years, most of the burning occurred during seven (of those) years. During the most severe years, 5.5 million hectares burned, while only 1 million hectares burned during the other 23 years."

Scientists are concerned that a continued warming trend would increase the frequency of severe boreal wildfire years and, due to the accelerating rate carbon of dioxide emissions, a "positive feedback loop" could occur in which each trend drives the other. But shouldn’t both the fires and the warmer climate stimulate increased plant growth, thereby enhancing the boreal forest’s ability to act as a carbon sink?

Yes, says Hall and his colleagues, but the rate of carbon storage is likely to be slower than the rate of its release. To date, both experimental and model data suggest that there is no way the plants can absorb carbon as quickly as fire and decomposition can release it from the soils where it has been stored for thousands of years.
 

Fires are already widespread throughout the boreal forests. Will warming temperatures increase the number and severity of these fires, potentially accelerating the rate of change? Researchers at NASA and elsewhere are using satellite data, computer models, and field work to find out. (Photograph Courtesy Forrest Hall, NASA Goddard Space Flight Center/University of Maryland)

Several times more carbon is stored in the organic soils of the boreal forest than in living plants. More fires will likely reduce the amount of carbon being stored by the boreal forests, especially if they are intense enough to burn the ground layer. This graph shows the amount of carbon stored in a typical region of the Alaskan forest relative to its age. (Graph by Kasischke, Christensen, and Stocks)

Moreover, in our lifetimes we might witness the transformation of the boreal ecosystem from a sink of 1-2 billion metric tons of carbon per year—a role it has played for millennia—to a significant source of carbon. Models indicate that a global warming scenario, combined with the increased presence of fire, would yield a net transfer of 27 to 52 petagrams (27 to 52 billion metric tons) of carbon from the boreal ecosystem to the atmosphere over the next century (Kasischke et al 1995).

The BOREAS scientists are eagerly awaiting the launch of NASA’s Terra spacecraft. They say that, in particular, the Moderate-resolution Imaging Spectroradiometer (MODIS) instrument provides them the multi-spectral and multi-temporal resolution they need to monitor the boreal ecosystem over the next decade. The new sensor is expected to not only dramatically improve scientists’ ability to measure the extent of fires globally, but it also help them quantify the effects fires have on the Earth’s carbon and radiation budgets. With MODIS data, they hope to begin answering some of the burning questions raised by an increased presence of wildfires in the boreal forest.

References

Betts, Alan K., Pedro Viterbo, Anton Beljaars, Hua-Lu Pan, Song-You Hong, Mike Goulden, and Steve Wofsy, 1998: "Evaluation of land-surface interaction in ECMWF and NCEP/NCAR reanalysis models over grassland (FIFE) and boreal forest (BOREAS)." Journal of Geophysical Research, 103, pp. 23,079-85.

Hall, Forrest G., David E. Knapp, & Karl F. Huemmrich, 1997: "Physically based classification and satellite mapping of biophysical characteristics in the southern boreal forest." Journal of Geophysical Research, 102, pp. 29,567-80.

Intergovernmental Panel on Climate Change, 1995: "Impacts, Adaptations and Mitigation of Climate Change: Scientific-Technical Analyses." Eds. Robert T. Watson, et al., Cambridge University Press.

Kasischke, Eric S., N.L. Christensen, Jr., & Brian J. Stocks, 1995: "Fire, Global Warming, and the Carbon Balance of Boreal Forests." Ecological Applications, 5, pp. 437-51.

Kasischke, Eric S., Kathleen Bergen, R. Fennimore, F. Sotelo, G. Stephens, Anthony Janetos, & H. Hank Shugart, 1999: "Satellite Imagery Gives Clear Picture of Russia's Boreal Forest Fires." Eos Transactions, 80, pp. 141 and 147.

Li, Zhanqing, Josef Cihlar, Louis Moreau, Fengting Huang, & Bryan Lee, 1997: "Monitoring fire activities in the boreal ecosystem." Journal of Geophysical Research, 102, pp. 29,611-24.

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.

Steyaert, Louis T., Forrest G. Hall, & T.R. Loveland, 1997: "Land cover mapping, fire regeneration, and scaling studies in the Canadian boreal forest with 1 km AVHRR and Landsat TM data." Journal of Geophysical Research, 102, pp. 29,581-98.

Stocks, Brian J., M.A. Fosberg, T.J. Lynham, L. Mearns, B.M. Wotton, Q. Yang, J-Z. Jin, K. Lawrence, G.R. Hartley, J.A. Mason, & D.W. McKenney, 1998: "Climate Change and Forest Fire Potential in Russian and Canadian Boreal Forests." Climate Change, 38, pp. 1-13.

Turner, Carla, 1999: "Canada's Trees." CBC News Online. http://cbc.ca/news/indepth/
fightingfires/canadastrees.html

1998 was an extreme year for wildfire activity throughout the North American and Russian boreal forests. More than 11 million hectares (110,000 square km) burned that year (Kasischke et al 1999). Examination of NOAA Advanced Very High Resolution Radiometer (AVHRR) and Landsat satellite images over the world’s high northern latitude forests reveals a heavy peppering of burn scars across the boreal landscape. Closer examination of those image data reveals the boreal forest canopy to be a patchy mosaic of splotches where there are various stages of plant regrowth in the wake of earlier fires.
 

Although most people regard fire as a destructive force that should be fought and quickly extinguished, the fact is the boreal forest evolved in the presence of fire and adapted to it. Forrest Hall says it’s not a question of if a given region of the boreal forest will burn, it’s a question of when. Hall, a physicist at NASA’s Goddard Space Flight Center, explains that wildfire is an integral part of the boreal ecosystem. Indeed, the high northern latitude forests would be quite different were it not for frequent fires (Hall 1999).
 

"Fire is the mechanism by which the forest is continually regenerated," states Hall. Fires consume dead, decaying vegetation accumulating on the forest floor, thereby clearing the way for new growth. Some species, such as the jack pine, even rely on fire to spread their seeds. The jack pine produces "seratonous" (resin-filled) cones that are very durable. The cones remain dormant until a fire occurs and melts the resin. Then the cones pop open and the seeds fall or blow out.

"In the Canadian boreal forest, aspen and jack pine are the most important ‘pioneer species’," observes Hall. "They are usually the first to grow back in a region that has been affected by fire. A few years after the fire, you typically see either dog-hair thick aspen stands or knee-high young jack pine trees sprouting all over the place."

Then comes secondary growth of the other tree species common to the boreal region—spruce or fir trees. Once spruce trees reach the same height as the other species, they become competitively superior to the other trees. Spruce have better access to sunlight while shading the other trees with their more dense canopies. Consequently, the jack pine and aspen begin dying out in regions where spruce trees are plentiful. But, should another fire come and disturb this cycle, the more fire-tolerant jack pine and aspen regain the advantage and again proliferate. (At far northern latitudes and in most of the Siberian boreal forest—depending upon such variables as soil type and wetness, topography, and local climate—species such as jack pine don't fair well and the regeneration species are mostly shrubs and black spruce.)

The amount of "standing biomass" (plants and trees) continually increases for 140 to 200 years after a fire; then the amount of standing biomass stabilizes and decreases as mature trees begin to die off. As plants die, fall to the ground and decay, the amount of carbon stored in the soil and ground moss rises over centuries to millennia. In this way, over geologic time, huge amounts of carbon have accumulated in the world’s boreal forest floor—about 37 percent of the carbon stored on land, and about 15 percent of the world’s total carbon content (Kasischke et al 1995).

A Burning Issue
In the Russian boreal forest, most fires are ignited by lightning strikes hitting trees or the ground. In North America, about 58 percent of the wildfires are caused by humans, while the rest are caused by lightning (Turner 1999). In relatively dry summer seasons, like 1998, thousands of fires are started by lightning, consuming millions of acres of boreal forest lands. Most of these fires not only consume trees, but also burn the top layers of soil, thereby releasing the carbon stored there too.

Of concern to Earth scientists is the impact widespread boreal fires have on climate. Fires release huge amounts of smoke, carbon dioxide, and methane into the atmosphere. These are greenhouse gases that act like "insulation" in the atmosphere to help trap and retain heat emitted from the surface. Some scientists are concerned that extreme fire seasons in the boreal forest, like last year’s, may contribute to global warming. Others point out that when a region is disturbed by fire, there is vigorous regrowth that could offset global warming. The rationale is that while the younger plants are growing back they are absorbing carbon dioxide back out of the atmosphere and using the carbon to build plant structures.

Over the long term, does fire render the boreal ecosystem a "source" or "sink" of carbon?

In some areas of the Canadian boreal forest, jack pine is the first species to appear. They are dependent on fire to open their tightly-sealed pinecones. Aspen, mixed with the jack pine in this photo, are another pioneer species.

With their taller, thicker canopies, black spruce gain a competitive advantage over the shorter aspen and jack pine. Deprived of sunlight in the shade of mature black spruce, these latter trees begin to die off.

The final stage of succession is a mature forest of black spruce. Over time dropped needles, dead branches, and fallen trees will build up on the forest floor, decaying only slowly because of the cold climate. When enough fuel accumulates, lightning will ignite another fire, resetting the cycle. (Photographs courtesy Lou Steyaert, NASA Goddard Space Flight Center/United States Geological Survey)

1998 was an extreme year for wildfire activity throughout the North American and Russian boreal forests. More than 11 million hectares (110,000 square km) burned that year (Kasischke et al 1999). Examination of NOAA Advanced Very High Resolution Radiometer (AVHRR) and Landsat satellite images over the world’s high northern latitude forests reveals a heavy peppering of burn scars across the boreal landscape. Closer examination of those image data reveals the boreal forest canopy to be a patchy mosaic of splotches where there are various stages of plant regrowth in the wake of earlier fires.
 

Coming Soon:
· The Migrating Boreal Forest

Although most people regard fire as a destructive force that should be fought and quickly extinguished, the fact is the boreal forest evolved in the presence of fire and adapted to it. Forrest Hall says it’s not a question of if a given region of the boreal forest will burn, it’s a question of when. Hall, a physicist at NASA’s Goddard Space Flight Center, explains that wildfire is an integral part of the boreal ecosystem. Indeed, the high northern latitude forests would be quite different were it not for frequent fires (Hall 1999).
 

"Fire is the mechanism by which the forest is continually regenerated," states Hall. Fires consume dead, decaying vegetation accumulating on the forest floor, thereby clearing the way for new growth. Some species, such as the jack pine, even rely on fire to spread their seeds. The jack pine produces "seratonous" (resin-filled) cones that are very durable. The cones remain dormant until a fire occurs and melts the resin. Then the cones pop open and the seeds fall or blow out.

"In the Canadian boreal forest, aspen and jack pine are the most important ‘pioneer species’," observes Hall. "They are usually the first to grow back in a region that has been affected by fire. A few years after the fire, you typically see either dog-hair thick aspen stands or knee-high young jack pine trees sprouting all over the place."

Then comes secondary growth of the other tree species common to the boreal region—spruce or fir trees. Once spruce trees reach the same height as the other species, they become competitively superior to the other trees. Spruce have better access to sunlight while shading the other trees with their more dense canopies. Consequently, the jack pine and aspen begin dying out in regions where spruce trees are plentiful. But, should another fire come and disturb this cycle, the more fire-tolerant jack pine and aspen regain the advantage and again proliferate. (At far northern latitudes and in most of the Siberian boreal forest—depending upon such variables as soil type and wetness, topography, and local climate—species such as jack pine don't fair well and the regeneration species are mostly shrubs and black spruce.)

The amount of "standing biomass" (plants and trees) continually increases for 140 to 200 years after a fire; then the amount of standing biomass stabilizes and decreases as mature trees begin to die off. As plants die, fall to the ground and decay, the amount of carbon stored in the soil and ground moss rises over centuries to millennia. In this way, over geologic time, huge amounts of carbon have accumulated in the world’s boreal forest floor—about 37 percent of the carbon stored on land, and about 15 percent of the world’s total carbon content (Kasischke et al 1995).

A Burning Issue
In the Russian boreal forest, most fires are ignited by lightning strikes hitting trees or the ground. In North America, about 58 percent of the wildfires are caused by humans, while the rest are caused by lightning (Turner 1999). In relatively dry summer seasons, like 1998, thousands of fires are started by lightning, consuming millions of acres of boreal forest lands. Most of these fires not only consume trees, but also burn the top layers of soil, thereby releasing the carbon stored there too.

Of concern to Earth scientists is the impact widespread boreal fires have on climate. Fires release huge amounts of smoke, carbon dioxide, and methane into the atmosphere. These are greenhouse gases that act like "insulation" in the atmosphere to help trap and retain heat emitted from the surface. Some scientists are concerned that extreme fire seasons in the boreal forest, like last year’s, may contribute to global warming. Others point out that when a region is disturbed by fire, there is vigorous regrowth that could offset global warming. The rationale is that while the younger plants are growing back they are absorbing carbon dioxide back out of the atmosphere and using the carbon to build plant structures.

Over the long term, does fire render the boreal ecosystem a "source" or "sink" of carbon?

The data used in this study are available in one or more of NASA's Earth Science Data Centers.

In some areas of the Canadian boreal forest, jack pine is the first species to appear. They are dependent on fire to open their tightly-sealed pinecones. Aspen, mixed with the jack pine in this photo, are another pioneer species.

With their taller, thicker canopies, black spruce gain a competitive advantage over the shorter aspen and jack pine. Deprived of sunlight in the shade of mature black spruce, these latter trees begin to die off.

The final stage of succession is a mature forest of black spruce. Over time dropped needles, dead branches, and fallen trees will build up on the forest floor, decaying only slowly because of the cold climate. When enough fuel accumulates, lightning will ignite another fire, resetting the cycle. (Photographs courtesy Lou Steyaert, NASA Goddard Space Flight Center/United States Geological Survey)