The Global Carbon Cycle
USGCRP Fiscal Years 2004-2005 Accomplishments

USGCRP
Program Elements

Atmospheric Composition

Ecosystems

Global Carbon Cycle

Decision-Support Resources Development and Related Research on Human Contributions and Responses

Climate Variability and Change

The Global
Water Cycle

Observing and Monitoring the Climate System

Communications

International Research and Cooperation

The following are selected highlights of recent research supported by CCSP participating agencies (as reported in the fiscal year 2006 edition of the annual report, Our Changing Planet).

AmeriFlux Network Measures Terrestrial Carbon Sinks and Sources and Identifies Biological Controls [7, 10, 18]

The AmeriFlux network of research sites measures terrestrial carbon sinks and sources and biological processes that regulate the net exchange of CO₂ between terrestrial ecosystems and the atmosphere [referred to as "net ecosystem exchange" (NEE)]. The measurements show that mature forests are important sinks for atmospheric CO₂ (Hollinger et al ., 2004). Disturbances that replace or remove forests can result in the land being a net source of CO₂ for a few years in mild climates or up to a decade in harsh climates while the forests are recovering (Law et al ., 2004), after which they may gradually revert to being a sink until the next disturbance. Moist tropical forests are not necessarily strong sinks because some of them have experienced recent mortality events that leave large amounts of decaying plant material that can release more CO₂ to the atmosphere than the forests remove [net source of about 1 tonne per hectare per year (Saleska et al ., 2003). Thus, the range of observed annual NEE of CO₂ ranges from a source of about 1 tonne per hectare to a net sink of 2 to 4 tonnes per hectare for forests and about 1 tonne or less per hectare for agricultural crops and grasslands. These results are expected to dramatically improve process depictions in carbon and climate models.

AmeriFlux Measurements Reveal Direct Influence of Volcanic Eruptions on Terrestrial Carbon Cycle. [1, 5]

The growth rate of atmospheric CO₂ concentration decreased after the explosive eruption of Mt. Pinatubo in the Philippines in June 1991. Scientists have been debating the cause of this atmospheric CO₂ decrease for many years. In a recent study by Gu et al . (2003), a group of AmeriFlux scientists found that aerosols (tiny particles suspended in the air) formed in the aftermath of the eruption altered the quantity and quality of solar radiation for vegetation, and this alteration significantly enhanced photosynthesis under cloudless conditions in a deciduous forest in the United States. These measurements indicate that increased CO₂ uptake by terrestrial vegetation caused by the effects of aerosols from the Pinatubo eruption on incident solar radiation were at least partly responsible for the sudden decline in the growth rate of atmospheric CO₂ concentration after the eruption. The study also indicates that anthropogenic aerosols, which can similarly change the quantity and quality of solar radiation for vegetation, may affect carbon cycling dynamics. A modeling study by Angert et al . (2004) suggests an alternate mechanism for the enhanced CO₂ sink in 1992-1993, due to a unique combination of an enhanced ocean CO₂ sink, reduced respiration driven by cooling and drying of the upper layers of the soil thereby reducing heterotrophic respiration, and reduced biomass burning. These studies illustrate the complexity of the global carbon cycle and underscore the crucial need for continued observations to resolve carbon cycle uncertainties.

Impact of Fires on Interannual Variability in Global Atmospheric Trace Gases [20]

Year-to-year changes in the atmospheric concentration of CO₂ and CH₄ are linked to fire activity associated with the El Niño/La Niña cycle. Information on vegetation, precipitation, surface air temperatures, and changes in Earth's radiation, derived from the Terra, Aqua, Tropical Rainfall Measuring Mission, and Active Cavity Radiometer Irradiance Monitor satellites, was used in modeling analyses of fire effects for 1997 to 2001. Researchers quantified the amount of CO₂ and CH₄ emitted by fires by combining the satellite data with in situ measurements of atmospheric gases. Emissions of greenhouse gases from fires increased across multiple continental regions in the El Niño year (1997-1998), including Southeast Asia (60% of the global increase), Central and South America (30%), and boreal forests of North America and Eurasia (10%). Vast areas of the tropics dry out and become vulnerable to fire during El Niño events, thus enabling humans to use fire more effectively as a tool for clearing land. Increases in fires and associated emissions of greenhouse gases are expected if El Niño events increase in frequency and/or intensity in the future.

Regionalization of Methane Emissions from the Amazon Basin [13]

The largest single source of CH₄ emissions to the atmosphere is natural wetlands, which account for between 20 and 40% of annual emissions; 60% of this is estimated to come from tropical wetlands, although there is large uncertainty regarding the magnitude and variability of that source. The Amazon Basin contains one of the largest areas of seasonally inundated tropical wetlands. Microwave remote-sensing techniques were used to measure wetland extent and seasonal variability, and the resulting data were combined with results from existing field studies of CH₄ flux in a model to extrapolate CH₄ flux across the entire Amazon Basin. Extrapolation to the Central Amazon Basin (1.77 million km² area) produces an estimated 6.8 TgC yr -1 , and extrapolation to the full Amazon River Basin below 500-m elevation (5.2 million km² , of which 17% is wetland) produces an estimated 22 TgC yr -1 from CH₄ emissions. This estimate is lower than those previously reported for the region. The satellite data analyses indicate that past studies overestimated the length of time seasonal wetlands were flooded and therefore overestimated annual methane emissions.

Variability in Low Productivity Ocean Gyres [12]

The mid-ocean gyres are huge circulation cells within the major ocean basins. They are vast expanses of very low primary production and biomass. Their low productivity results from the physical dynamics of the circulation within the gyre, which depresses the depths at which significant nutrient concentrations (e.g., nitrate, phosphate, and silicate) required for photosynthesis occur. Global surface chlorophyll concentrations measured over seven years by the U.S. Sea-viewing Wide Field-of-view Sensor (SeaWiFS) and Japanese Ocean Color and Thermal Scanner (OCTS) satellite sensors were used to estimate temporal changes in the size of each gyre. The analysis shows that these areas have a distinct seasonal cycle and that the overall area of the gyres in the North Pacific and North Atlantic expanded during 1996 to 2003, while there was little change in the area of the South Pacific, South Atlantic, and southern Indian Ocean gyres. The study did not attempt to explain why the Northern Hemisphere gyres are expanding and was too short to determine whether the trends will continue. This first-ever quantification of year-to-year variability across the global oceans has implications for carbon dynamics, and raises a concern about whether continued expansion of the low-productivity gyres would reduce the size of the ocean carbon sink. Continuing systematic satellite observations will be available to monitor future trends and address this concern.

Changing Carbon Dynamics in the Oceans [4, 17]

The Repeat Hydrography CO₂ / Tracer Program is a systematic and global re-measurement of select cross-sections of the ocean to quantify changes in storage and transport of heat, freshwater, CO₂, chlorofluorocarbon tracers, and related properties. For cruises in the North Pacific, difference plots for the 2004 Repeat Hydrography compared with the 1994 World Ocean Circulation Experiment quantitatively document significant changes. Increases in dissolved inorganic carbon (DIC) of up to 35 µmol kg -1 were observed in surface waters and in intermediate depths ranging from 200 to 1000 m. On average, mixed layer DIC increases of 1.5 ± 0.2 µmol kg -1 yr -1 were observed in the subtropical waters of the North Pacific, indicating that over the past decade the oceanic uptake of CO₂ in this part of the global ocean has been faster than the rate of growth of CO₂ in the atmosphere. These results indicate that the rate of uptake of CO₂ may differ significantly in different regions of the global ocean.

Synthesis of Forest FACE Experiment Data [8, 21]

Common data sets were assembled from four Free-Air CO₂ Enrichment (FACE) experiments being conducted in forest ecosystems. The ongoing, multi-year experiments were in established stands in North Carolina (loblolly pine) and Tennessee (sweetgum), and in young stands in Wisconsin (aspen-birch-maple) and Italy (poplar). Soil respiration increased in response to CO₂ enrichment, and the relative response was larger in the young stands. The temperature sensitivity of soil respiration was unaffected by CO₂ enrichment. No effects on soil nitrogen pools or processes were observed in the three forest FACE experiments in the United States, indicating that nitrogen limitations are unlikely to constrain increases in forest productivity at similar sites in the initial years following CO₂ enrichment. Continued study of these effects over longer time periods will be critical for our ability to predict long-term changes in soil nitrogen availability and the potential for sustained increases in productivity in a CO₂-enriched atmosphere.

Root Dynamics Control Forest Response to Atmospheric Carbon Dioxide [11, 14]

Analysis of a nearly continuous 6-year record of fine root production and mortality in an experimental sweetgum forest revealed a large increase in root production and a significant change in the depth distribution of roots in forest plots exposed to a CO₂-enriched atmosphere. These responses, which were measured in a FACE experiment, have important implications for carbon sequestration and nitrogen and water uptake in this and other forest ecosystems. Allocation of carbon to fine roots reduces the potential for carbon sequestration in plant biomass, but increases the potential for carbon storage in soil. Comparison of the responses of the deciduous forest with those in a similar experiment in a pine forest suggests that root system dynamics can explain differences among ecosystems in their response to elevated atmospheric CO₂. These results indicate that accurate assessments of carbon flux and storage in forests must account for the responses of root systems.

Long-Term Forest Management Studies Inform Carbon Management Options [15]

The Long-Term Soil Productivity (LTSP) study seeks to understand how anthropogenic disturbances affect the land's capacity to store carbon through the conduct of a series of 62 long-term field experiments in major forest types of the United States and Canada. This unique study has been in place for more than a decade and is the largest and most extensive experiment of its type in the world. Results show differences in carbon and nutrient dynamics by soil type and climate regime and provide critical information for developing management systems that maintain and enhance productivity and carbon sequestration options. Findings from studies of alternative fuel treatments for reducing fire risk indicate no significant differences in CO₂ efflux following treatment and provide a basis for local communities to reduce fire risk and improve the health and vigor of forests without affecting greenhouse gas emissions.

Land Resource Management and Reclamation Activities have Significant Potential to Sequester Carbon [ 2, 3, 9, 16, 19]

As noted previously, changes in soil carbon can potentially change the concentrations of greenhouse gases in the atmosphere. If CO₂ from the atmosphere, captured through photosynthesis, is ultimately stored in the soil to a greater degree, the resulting soil carbon sequestration may help slow the rise of atmospheric CO₂.   Studies of soil carbon dynamics and storage in a variety of managed systems in the United States and elsewhere have identified and quantified significant carbon sequestration potentials:

• In situ measurements and models have determined soil organic carbon (SOC) sequestration potentials for croplands, pastures and rangelands, disturbed lands such as mining areas, natural areas such as wetlands, land in the Conservation Reserve Program (CRP), and urban areas such as lawns, parks, and golf courses. With improved management practices, the potential average rates of SOC sequestration are approximately 70, 40, 60, and 13 Tg of organic carbon annually for cropland, grazing land, forests, and CRP soils, respectively (Lal et al ., 2003; USDA, 2004).
• Conversion of cropland to perennial systems in central Iowa increased the SOC content by an average of 11 tonnes per hectare in the upper 35 cm of soil (Cambardella et al ., 2004).
• For agroforestry systems in Costa Rica, increased biodiversity resulted in greater soil nutrient supply and carbon storage; plant roots were primarily responsible for soil carbon accrual in these systems. Root organic matter quality, and not the amount of root inputs, best explained effects of species diversity on soil carbon sequestration (Russell et al ., 2003).
• Restoration of semi-permanent wetlands in the prairie pothole region of the northern Great Plains increased carbon content in surface sediments (Euliss et al ., 2003).

These results are examples of findings that contribute new scientific knowledge about relationships of terrestrial ecosystem processes and management practices on carbon sequestration. Land and resource managers, agricultural consultants, and environmental organizations are using this new information to foster the development of agricultural and land management systems that mitigate greenhouse gas emissions, enhance soil fertility, and improve soil and water quality.

Climate-Induced Thawing of Permafrost and Implications for Soil Carbon Stocks [6]

The high-latitude regions of North America, including interior Alaska, are critical areas for research because thawing of permafrost within the soil zone can cause a release of carbon to the atmosphere. In the past 40 years, boreal North America has warmed by at least 2°C and has experienced pervasive drying of lakes and water tables. Landscapes with near-surface permafrost store greater than 60% of their carbon stocks in organic soils. While long-term records indicate net carbon storage on land, fires in recent decades have burned these landscapes in unexpected proportions. Trace gas studies at the Bonanza Creek Long-Term Ecological Research Station have revealed complex responses of permafrost landscapes to fire, with post-burn hydrological conditions and plant ecological processes determining the net balance between CH₄ and CO₂ emissions and carbon sequestration. Drought and fire induce complex responses (e.g., water table fluctuation, revegetation patterns) due in part to the discontinuous nature of the permafrost. These findings emphasize the importance of surface and subsurface hydrology for understanding the carbon dynamics of boreal and arctic regions and emphasize the need for baseline studies of carbon cycling on a variety of temporal and spatial scales.