A carbon sink is a natural or artificial reservoir that accumulates and stores some carbon-containing chemical compound for an indefinite period. The process by which carbon sinks remove carbon dioxide (CO2) from the atmosphere is known as carbon sequestration. Public awareness of the significance of CO2 sinks has grown since passage of the Kyoto Protocol, which promotes their use as a form of carbon offset.
The main natural sinks are:
Natural sinks are typically much larger than artificial sinks. The main artificial sinks are:
Carbon sources include:
Contents |
Because growing vegetation absorbs carbon dioxide, the Kyoto Protocol allows Annex I countries with large areas of growing forests to issue Removal Units to recognise the sequestration of carbon. The additional units make it easier for them to achieve their target emission levels.
Some countries seek to trade emission rights in carbon emission markets, purchasing the unused carbon emission allowances of other countries. If overall limits on greenhouse gas emission are put into place, cap and trade market mechanisms are purported to find cost-effective ways to reduce emissions.[1] There is as yet no carbon audit regime for all such markets globally, and none is specified in the Kyoto Protocol. National carbon emissions are self-declared.
In the Clean Development Mechanism, only afforestation and reforestation are eligible to produce certified emission reductions (CERs) in the first commitment period of the Kyoto Protocol (2008–2012). Forest conservation activities or activities avoiding deforestation, which would result in emission reduction through the conservation of existing carbon stocks, are not eligible at this time.[2] Also, agricultural carbon sequestration is not possible yet.[3]
Soils represent a short to long-term carbon storage medium, and contain more carbon than all terrestrial vegetation and the atmosphere combined.[4][5] Plant litter and other biomass accumulates as organic matter in soils, and is degraded by chemical weathering and biological degradation. More recalcitrant organic carbon polymers such as cellulose, hemi-cellulose, lignin, aliphatic compounds, waxes and terpenoids are collectively retained as humus.[6] Organic matter tends to accumulate in litter and soils of colder regions such as the boreal forests of North America and the Taiga of Russia. Leaf litter and humus are rapidly oxidized and poorly retained in sub-tropical and tropical climate conditions due to high temperatures and extensive leaching by rainfall. Areas where shifting cultivation or slash and burn agriculture are practiced are generally only fertile for 2–3 years before they are abandoned. These tropical jungles are similar to coral reefs in that they are highly efficient at conserving and circulating necessary nutrients, which explains their lushness in a nutrient desert. Much organic carbon retained in many agricultural areas worldwide has been severely depleted due to intensive farming practices.
Grasslands contribute to soil organic matter, stored mainly in their extensive fibrous root mats. Due in part to the climactic conditions of these regions (e.g. cooler temperatures and semi-arid to arid conditions), these soils can accumulate significant quantities of organic matter. This can vary based on rainfall, the length of the winter season, and the frequency of naturally occurring lightning-induced grass-fires. While these fires release carbon dioxide, they improve the quality of the grasslands overall, in turn increasing the amount of carbon retained in the retained humic material. They also deposit carbon directly to the soil in the form of char that does not significantly degrade back to carbon dioxide.
Forest fires release absorbed carbon back into the atmosphere, as does deforestation due to rapidly increased oxidation of soil organic matter.
Organic matter in peat bogs undergoes slow anaerobic decomposition below the surface. This process is slow enough that in many cases the bog grows rapidly and fixes more carbon from the atmosphere than is released. Over time, the peat grows deeper. Peat bogs inter approximately one-quarter of the carbon stored in land plants and soils.[7]
Under some conditions, forests and peat bogs may become sources of CO2, such as when a forest is flooded by the construction of a hydroelectric dam. Unless the forests and peat are harvested before flooding, the rotting vegetation is a source of CO2 and methane comparable in magnitude to the amount of carbon released by a fossil-fuel powered plant of equivalent power.[8]
Current agricultural practices lead to carbon loss from soils. It has been suggested that improved farming practices could return the soils to being a carbon sink. Present worldwide practises of overgrazing are substantially reducing many grasslands' performance as carbon sinks.[9] The Rodale Institute says that Regenerative agriculture, if practiced on the planet’s 3.5 billion tillable acres, could sequester up to 40% of current CO2 emissions.[10][11] They claim that agricultural carbon sequestration has the potential to mitigate global warming. When using biologically based regenerative practices, this dramatic benefit can be accomplished with no decrease in yields or farmer profits. Organically managed soils can convert carbon dioxide from a greenhouse gas into a food-producing asset.
In 2006, U.S. carbon dioxide emissions from fossil fuel combustion were estimated at nearly 6.5 billion tons. If a 2,000 (lb/ac)/year sequestration rate was achieved on all 434,000,000 acres (1,760,000 km2) of cropland in the United States, nearly 1.6 billion tons of carbon dioxide would be sequestered per year, mitigating close to one quarter of the country's total fossil fuel emissions.
Oceans are at present CO2 sinks, and represent the largest active carbon sink on Earth, absorbing more than a quarter of the carbon dioxide that humans put into the air.[12] On longer timescales they may be both sources and sinks - during ice ages CO2 levels decrease to ~180 ppmv, and much of this is believed to be stored in the oceans. As ice ages end, CO2 is released from the oceans and CO2 levels during previous interglacials have been around ~280 ppmv. This role as a sink for CO2 is driven by two processes, the solubility pump and the biological pump.[13] The former is primarily a function of differential CO2 solubility in seawater and the thermohaline circulation, while the latter is the sum of a series of biological processes that transport carbon (in organic and inorganic forms) from the surface euphotic zone to the ocean's interior. A small fraction of the organic carbon transported by the biological pump to the seafloor is buried in anoxic conditions under sediments and ultimately forms fossil fuels such as oil and natural gas.
At the present time, approximately one third[14] of human generated emissions are estimated to be entering the ocean. The solubility pump is the primary mechanism driving this, with the biological pump playing a negligible role. This stems from the limitation of the biological pump by ambient light and nutrients required by the phytoplankton that ultimately drive it. Total inorganic carbon is not believed to limit primary production in the oceans, so its increasing availability in the ocean does not directly affect production (the situation on land is different, since enhanced atmospheric levels of CO2 essentially "fertilize" land plant growth). However, ocean acidification by invading anthropogenic CO2 may affect the biological pump by negatively impacting calcifying organisms such as coccolithophores, foraminiferans and pteropods. Climate change may also affect the biological pump in the future by warming and stratifying the surface ocean, thus reducing the supply of limiting nutrients to surface waters.
In January 2009, the Monterey Bay Aquarium Research Institute and the National Oceanic and Atmospheric Administration announced a joint study to determine whether the ocean off the California coast was serving as a carbon source or a carbon sink. Principal instrumentation for the study will be self-contained CO2 monitors placed on buoys in the ocean. They will measure the partial pressure of CO2 in the ocean and the atmosphere just above the water surface.[15]
In February 2009, Science Daily reported that the Southern Indian Ocean is becoming less effective at absorbing carbon dioxide due to changes to the regions climate which include higher wind speeds.[16]
Forests are carbon stores, and they are carbon dioxide sinks when they are increasing in density or area. In Canada's boreal forests as much as 80% of the total carbon is stored in the soils as dead organic matter.[17] A 40-year study of African, Asian, and South American tropical forests by the University of Leeds, shows tropical forests absorb about 18% of all carbon dioxide added by fossil fuels.[18] Tropical reforestation can mitigate global warming until all available land has been reforested with mature forests. However, the global cooling effect of carbon sequestration by forests is partially counterbalanced in that reforestation can decrease the reflection of sunlight (albedo). Mid-to-high latitude forests have a much lower albedo during snow seasons than flat ground, thus contributing to warming. Modeling that compares the effects of albedo differences between forests and grasslands suggests that expanding the land area of forests in temperate zones offers only a temporary cooling benefit.[19][20][21][22]
In the United States in 2004 (the most recent year for which EPA statistics[23] are available), forests sequestered 10.6% (637 teragrams[24]) of the carbon dioxide released in the United States by the combustion of fossil fuels (coal, oil and natural gas; 5657 teragrams[25]). Urban trees sequestered another 1.5% (88 teragrams[24]). To further reduce U.S. carbon dioxide emissions by 7%, as stipulated by the Kyoto Protocol, would require the planting of "an area the size of Texas [8% of the area of Brazil] every 30 years".[26] Carbon offset programs are planting millions of fast-growing trees per year to reforest tropical lands, for as little as $0.10 per tree; over their typical 40-year lifetime, one million of these trees will fix 0.9 teragrams of carbon dioxide.[27] In Canada, reducing timber harvesting would have very little impact on carbon dioxide emissions because of the combination of harvest and stored carbon in manufactured wood products along with the regrowth of the harvested forests. Additionally, the amount of carbon released from harvesting is small compared to the amount of carbon lost each year to forest fires and other natural disturbances.[17]
The Intergovernmental Panel on Climate Change concluded that "a sustainable forest management strategy aimed at maintaining or increasing forest carbon stocks, while producing an annual sustained yield of timber fibre or energy from the forest, will generate the largest sustained mitigation benefit".[28] Sustainable management practices keep forests growing at a higher rate over a potentially longer period of time, thus providing net sequestration benefits in addition to those of unmanaged forests.[29]
Life expectancy of forests varies throughout the world, influenced by tree species, site conditions and natural disturbance patterns. In some forests carbon may be stored for centuries, while in other forests carbon is released with frequent stand replacing fires. Forests that are harvested prior to stand replacing events allow for the retention of carbon in manufactured forest products such as lumber.[30] However, only a portion of the carbon removed from logged forests ends up as durable goods and buildings. The remainder ends up as sawmill by-products such as pulp, paper and pallets, which often end with incineration (resulting in carbon release into the atmosphere) at the end of their lifecycle. For instance, of the 1,692 teragrams of carbon harvested from forests in Oregon and Washington (U.S) from 1900 to 1992, only 23% is in long-term storage in forest products.[31]
One way to increase the carbon sequestration efficiency of the oceans is to add micrometre-sized iron particles in the form of either hematite (iron oxide) or melanterite (iron sulfate) to certain regions of the ocean. This has the effect of stimulating growth of plankton. Iron is an important nutrient for phytoplankton, usually made available via upwelling along the continental shelves, inflows from rivers and streams, as well as deposition of dust suspended in the atmosphere. Natural sources of ocean iron have been declining in recent decades, contributing to an overall decline in ocean productivity (NASA, 2003). Yet in the presence of iron nutrients plankton populations quickly grow, or 'bloom', expanding the base of biomass productivity throughout the region and removing significant quantities of CO2 from the atmosphere via photosynthesis. A test in 2002 in the Southern Ocean around Antarctica suggests that between 10,000 and 100,000 carbon atoms are sunk for each iron atom added to the water. More recent work in Germany (2005) suggests that any biomass carbon in the oceans, whether exported to depth or recycled in the euphotic zone, represents long-term storage of carbon. This means that application of iron nutrients in select parts of the oceans, at appropriate scales, could have the combined effect of restoring ocean productivity while at the same time mitigating the effects of human caused emissions of carbon dioxide to the atmosphere.
Because the effect of periodic small scale phytoplankton blooms on ocean ecosystems is unclear, more studies would be helpful. Phytoplankton have a complex effect on cloud formation via the release of substances such as dimethyl sulfide (DMS) that are converted to sulfate aerosols in the atmosphere, providing cloud condensation nuclei, or CCN. But the effect of small scale plankton blooms on overall DMS production is unknown.
Other nutrients such as nitrates, phosphates, and silica as well as iron may cause ocean fertilization. There has been some speculation that using pulses of fertilization (around 20 days in length) may be more effective at getting carbon to ocean floor than sustained fertilization.[32]
There is some controversy over seeding the oceans with iron however, due to the potential for increased toxic phytoplankton growth (e.g. "red tide"), declining water quality due to overgrowth, and increasing anoxia in areas harming other sea-life such as zooplankton, fish, coral, etc.[33][34]
Since the 1850s, a large proportion of the world's grasslands have been tilled and converted to croplands, allowing the rapid oxidation of large quantities of soil organic carbon. However, in the United States in 2004 (the most recent year for which EPA statistics are available), agricultural soils including pasture land sequestered 0.8% (46 teragrams[24]) as much carbon as was released in the United States by the combustion of fossil fuels (5988 teragrams[25]). The annual amount of this sequestration has been gradually increasing since 1998.[24]
Methods that significantly enhance carbon sequestration in soil include no-till farming, residue mulching, cover cropping, and crop rotation, all of which are more widely used in organic farming than in conventional farming.[35][36] Because only 5% of US farmland currently uses no-till and residue mulching, there is a large potential for carbon sequestration.[37] Conversion to pastureland, particularly with good management of grazing, can sequester even more carbon in the soil.
Terra preta, an anthropogenic, high-carbon soil, is also being investigated as a sequestration mechanism. By pyrolysing biomass, about half of its carbon can be reduced to charcoal, which can persist in the soil for centuries, and makes a useful soil amendment, especially in tropical soils (biochar or agrichar).[38] [39]
Controlled burns on far north Australian savannas can result in an overall carbon sink. One working example is the West Arnhem Fire Management Agreement, started to bring "strategic fire management across 28,000 km² of Western Arnhem Land". Deliberately starting controlled burns early in the dry season results in a mosaic of burnt and unburnt country which reduces the area of burning compared with stronger, late dry season fires. In the early dry season there are higher moisture levels, cooler temperatures, and lighter wind than later in the dry season; fires tend to go out overnight. Early controlled burns also results in a smaller proportion of the grass and tree biomass being burnt.[40] Emission reductions of 256,000 tonnes of CO2 have been made as of 2007.[41]
For carbon to be sequestered artificially (i.e. not using the natural processes of the carbon cycle) it must first be captured, or it must be significantly delayed or prevented from being re-released into the atmosphere (by combustion, decay, etc.) from an existing carbon-rich material, by being incorporated into an enduring usage (such as in construction). Thereafter it can be passively stored or remain productively utilized over time in a variety of ways.
For example, upon harvesting, wood (as a carbon-rich material) can be immediately burned or otherwise serve as a fuel, returning its carbon to the atmosphere, or it can be incorporated into construction or a range of other durable products, thus sequestering its carbon over years or even centuries. One ton of dry wood is equivalent to 1.8 tons of carbon dioxide.
Indeed, a very carefully designed and durable, energy-efficient and energy-capturing building has the potential to sequester (in its carbon-rich construction materials), as much as or more carbon than was released by the acquisition and incorporation of all its materials and than will be released by building-function "energy-imports" during the structure's (potentially multi-century) existence. Such a structure might be termed "carbon neutral" or even "carbon negative". Building construction and operation (electricity usage, heating, etc.) are estimated to contribute nearly half of the annual human-caused carbon additions to the atmosphere.[42]
Natural-gas purification plants often already have to remove carbon dioxide, either to avoid dry ice clogging gas tankers or to prevent carbon-dioxide concentrations exceeding the 3% maximum permitted on the natural-gas distribution grid.
Beyond this, one of the most likely early applications of carbon capture is the capture of carbon dioxide from flue gases at power stations (in the case of coal, this is known as "clean coal"). A typical new 1000 MW coal-fired power station produces around 6 million tons of carbon dioxide annually. Adding carbon capture to existing plants can add significantly to the costs of energy production; scrubbing costs aside, a 1000 MW coal plant will require the storage of about 50 million barrels (7,900,000 m3) of carbon dioxide a year. However, scrubbing is relatively affordable when added to new plants based on coal gasification technology, where it is estimated to raise energy costs for households in the United States using only coal-fired electricity sources from 10 cents per kW·h to 12 cents.[43]
Currently, capture of carbon dioxide is performed on a large scale by absorption of carbon dioxide onto various amine-based solvents. Other techniques are currently being investigated, such as pressure swing adsorption, temperature swing adsorption, gas separation membranes, and cryogenics. Recent pilot studies include flue capture and conversion to baking soda and use of algae for conversion to fuel or feed.
In coal-fired power stations, the main alternatives to retrofitting amine-based absorbers to existing power stations are two new technologies: coal gasification combined-cycle and oxy-fuel combustion. Gasification first produces a "syngas" primarily of hydrogen and carbon monoxide, which is burned, with carbon dioxide filtered from the flue gas. Oxy-fuel combustion burns the coal in oxygen instead of air, producing only carbon dioxide and water vapour, which are relatively easily separated. Some of the combustion products must be returned to the combustion chamber, either before or after separation, otherwise the temperatures would be too high for the turbine.
Another long-term option is carbon capture directly from the air using hydroxides. The air would literally be scrubbed of its CO2 content. This idea offers an alternative to non-carbon-based fuels for the transportation sector.
Examples of carbon sequestration at coal plants include converting carbon from smokestacks into baking soda,[44][45] and algae-based carbon capture, circumventing storage by converting algae into fuel or feed.[46]
Another proposed form of carbon sequestration in the ocean is direct injection. In this method, carbon dioxide is pumped directly into the water at depth, and expected to form "lakes" of liquid CO2 at the bottom. Experiments carried out in moderate to deep waters (350–3600 m) indicate that the liquid CO2 reacts to form solid CO2 clathrate hydrates, which gradually dissolve in the surrounding waters.
This method, too, has potentially dangerous environmental consequences. The carbon dioxide does react with the water to form carbonic acid, H2CO3; however, most (as much as 99%) remains as dissolved molecular CO2. The equilibrium would no doubt be quite different under the high pressure conditions in the deep ocean. In addition, if deep-sea bacterial methanogens that reduce carbon dioxide were to encounter the carbon dioxide sinks, levels of methane gas may increase, leading to the generation of an even worse greenhouse gas.[47] The resulting environmental effects on benthic life forms of the bathypelagic, abyssopelagic and hadopelagic zones are unknown. Even though life appears to be rather sparse in the deep ocean basins, energy and chemical effects in these deep basins could have far-reaching implications. Much more work is needed here to define the extent of the potential problems.
Carbon storage in or under oceans may not be compatible with the Convention on the Prevention of Marine Pollution by Dumping of Wastes and Other Matter.[48]
An additional method of long-term ocean-based sequestration is to gather crop residue such as corn stalks or excess hay into large weighted bales of biomass and deposit it in the alluvial fan areas of the deep ocean basin. Dropping these residues in alluvial fans would cause the residues to be quickly buried in silt on the sea floor, sequestering the biomass for very long time spans. Alluvial fans exist in all of the world's oceans and seas where river deltas fall off the edge of the continental shelf such as the Mississippi alluvial fan in the gulf of Mexico and the Nile alluvial fan in the Mediterranean Sea. A downside, however, would be an increase in aerobic bacteria growth due to the introduction of biomass, leading to more competition for oxygen resources in the deep sea, similar to the oxygen minimum zone.
The method of geo-sequestration or geological storage involves injecting carbon dioxide directly into underground geological formations. Declining oil fields, saline aquifers, and unminable coal seams have been suggested as storage sites. Caverns and old mines that are commonly used to store natural gas are not considered, because of a lack of storage safety.
CO2 has been injected into declining oil fields for more than 40 years, to increase oil recovery. This option is attractive because the storage costs are offset by the sale of additional oil that is recovered. Typically, 10-15% additional recovery of the original oil in place is possible. Further benefits are the existing infrastructure and the geophysical and geological information about the oil field that is available from the oil exploration. Another benefit of injecting CO2 into Oil fields is that CO2 is soluble in oil. Dissolving CO2 in oil lowers the viscosity of the oil and reduces its interfacial tension which increases the oils mobility. All oil fields have a geological barrier preventing upward migration of oil. As most oil and gas has been in place for millions to tens of millions of years, depleted oil and gas reservoirs can contain carbon dioxide for millennia. Identified possible problems are the many 'leak' opportunities provided by old oil wells, the need for high injection pressures and acidification which can damage the geological barrier. Other disadvantages of old oil fields are their limited geographic distribution and depths, which require high injection pressures for sequestration. Below a depth of about 1000 m, carbon dioxide is injected as a supercritical fluid, a material with the density of a liquid, but the viscosity and diffusivity of a gas. Unminable coal seams can be used to store CO2, because CO2 absorbs to the coal surface, ensuring safe long-term storage. In the process it releases methane that was previously adsorbed to the coal surface and that may be recovered. Again the sale of the methane can be used to offset the cost of the CO2 storage. Release or burning of methane would of course at least partially offset the obtained sequestration result – except when the gas is allowed to escape into the atmosphere in significant quantities: methane has a higher global warming potential than CO2.
Saline aquifers contain highly mineralized brines and have so far been considered of no benefit to humans except in a few cases where they have been used for the storage of chemical waste. Their advantages include a large potential storage volume and relatively common occurrence reducing the distance over which CO2 has to be transported. The major disadvantage of saline aquifers is that relatively little is known about them compared to oil fields. Another disadvantage of saline aquifers is that as the salinity of the water increases, less CO2 can be dissolved into aqueous solution. To keep the cost of storage acceptable the geophysical exploration may be limited, resulting in larger uncertainty about the structure of a given aquifer. Unlike storage in oil fields or coal beds, no side product will offset the storage cost. Leakage of CO2 back into the atmosphere may be a problem in saline-aquifer storage. However, current research shows that several trapping mechanisms immobilize the CO2 underground, reducing the risk of leakage.
A major research project examining the geological sequestration of carbon dioxide is currently being performed at an oil field at Weyburn in south-eastern Saskatchewan. In the North Sea, Norway's Statoil natural-gas platform Sleipner strips carbon dioxide out of the natural gas with amine solvents and disposes of this carbon dioxide by geological sequestration. Sleipner reduces emissions of carbon dioxide by approximately one million tonnes a year. The cost of geological sequestration is minor relative to the overall running costs. As of April 2005, BP is considering a trial of large-scale sequestration of carbon dioxide stripped from power plant emissions in the Miller oilfield as its reserves are depleted.
In October 2007, the Bureau of Economic Geology at The University of Texas at Austin received a 10-year, $38 million subcontract to conduct the first intensively monitored, long-term project in the United States studying the feasibility of injecting a large volume of CO2 for underground storage.[49] The project is a research program of the Southeast Regional Carbon Sequestration Partnership (SECARB), funded by the National Energy Technology Laboratory of the U.S. Department of Energy (DOE). The SECARB partnership will demonstrate CO2 injection rate and storage capacity in the Tuscaloosa-Woodbine geologic system that stretches from Texas to Florida. Beginning in fall 2007, the project will inject CO2 at the rate of one million tons per year, for up to 1.5 years, into brine up to 10,000 feet (3,000 m) below the land surface near the Cranfield oil field about 15 miles (24 km) east of Natchez, Mississippi. Experimental equipment will measure the ability of the subsurface to accept and retain CO2.
Mineral sequestration aims to trap carbon in the form of solid carbonate salts. This process occurs slowly in nature and is responsible for the deposition and accumulation of limestone over geologic time. Carbonic acid in groundwater slowly reacts with complex silicates to dissolve calcium, magnesium, alkalis and silica and leave a residue of clay minerals. The dissolved calcium and magnesium react with bicarbonate to precipitate calcium and magnesium carbonates, a process that organisms use to make shells. When the organisms die, their shells are deposited as sediment and eventually turn into limestone. Limestones have accumulated over billions of years of geologic time and contain much of Earth's carbon. Ongoing research aims to speed up similar reactions involving alkali carbonates.[50]
Several serpentinite deposits are being investigated as potentially large scale CO2 storage sinks such as those found in NSW, Australia, where the first mineral carbonation pilot plant project is underway.[51] Beneficial re-use of magnesium carbonate from this process could provide feedstock for new products developed for the built environment and agriculture without returning the carbon into the atmosphere and so acting as a carbon sink.
One proposed reaction is that of the olivine-rich rock dunite, or its hydrated equivalent serpentinite with carbon dioxide to form the carbonate mineral magnesite, plus silica and iron oxide (magnetite).
Serpentinite sequestration is favored because of the non-toxic and stable nature of magnesium carbonate. The ideal reactions involve the magnesium endmember components of the olivine (reaction 1) or serpentine (reaction 2), the latter derived from earlier olivine by hydration and silicification (reaction 3). The presence of iron in the olivine or serpentine reduces the efficiency of sequestration, since the iron components of these minerals break down to iron oxide and silica (reaction 4).
Reaction 1
Mg-olivine + carbon dioxide → magnesite + silica + water
Reaction 2
Serpentine + carbon dioxide → magnesite + silica + water
Reaction 3
Mg-olivine + water + silica → serpentine
Reaction 4
Fe-olivine + water → magnetite + silica + hydrogen
Zeolitic imidazolate frameworks is a metal-organic framework carbon dioxide sink which could be used to keep industrial emissions of carbon dioxide out of the atmosphere.[52]
According to a report in Nature magazine, (November, 2009) the first year-by-year accounting of this mechanism during the industrial era, and the first time scientists have actually measured it, suggests "the oceans are struggling to keep up with rising emissions—a finding with potentially wide implications for future climate." With total world emissions from fossil fuels growing rapidly, the proportion of fossil-fuel emissions absorbed by the oceans since 2000 may have declined by as much as 10%, indicating that over time the ocean will become "a less efficient sink of manmade carbon." Samar Khatiwala, an oceanographer at Columbia University concludes that the studies suggest "we cannot count on these sinks operating in the future as they have in the past, and keep on subsidizing our ever-growing appetite for fossil fuels."[12] However, a recent paper by Wolfgang Knorr indicates that the fraction of CO2 absorbed by carbon sinks has not changed since 1850.[53]