IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis New Developments in Knowledge of the Carbon Cycle Since the Third Assessment Report

Sections 7.3.2 to 7.3.5 describe where knowledge and understanding have advanced significantly since the Third Assessment Report (TAR). In particular, the budget of anthropogenic CO2 (shown by the red fluxes in Figure 7.3) can be calculated with improved accuracy. In the ocean, newly available high-quality data on the ocean carbon system have been used to construct robust estimates of the cumulative ocean burden of anthropogenic carbon (Sabine et al., 2004a) and associated changes in the carbonate system (Feely et al., 2004). The pH in the surface ocean is decreasing, indicating the need to understand both its interaction with a changing climate and the potential impact on organisms in the ocean (e.g., Orr et al., 2005; Royal Society, 2005). On land, there is a better understanding of the contribution to the buildup of CO2 in the atmosphere since 1750 associated with land use and of how the land surface and the terrestrial biosphere interact with a changing climate. Globally, inverse techniques used to infer the magnitude and location of major fluxes in the global carbon cycle have continued to mature, reflecting both refinement of the techniques and the availability of new observations. During preparation of the TAR, inclusion of the carbon cycle in climate models was new. Now, results from the first C4MIP are available: when the carbon cycle is included, the models consistently simulate climate feedbacks to land and ocean carbon cycles that tend to reduce uptake of CO2 by land and ocean from 1850 to 2100 (see Section 7.3.5).

7.3.2 The Contemporary Carbon Budget Atmospheric Increase

The atmospheric CO2 increase is measured with great accuracy at various monitoring stations (see Chapter 2; and Keeling and Whorf, 2005 updated by S. Piper through 2006). The mean yearly increase in atmospheric CO2 (the CO2 ‘growth rate’) is reported in Table 7.1. Atmospheric CO2 has continued to increase since the TAR (Figure 7.4), and the rate of increase appears to be higher, with the average annual increment rising from 3.2 ± 0.1 GtC yr–1 in the 1990s to 4.1 ± 0.1 GtC yr–1 in the period 2000 to 2005. The annual increase represents the net effect of several processes that regulate global land-atmosphere and ocean-atmosphere fluxes, examined below. The ‘airborne fraction’ (atmospheric increase in CO2 concentration/fossil fuel emissions) provides a basic benchmark for assessing short- and long-term changes in these processes. From 1959 to the present, the airborne fraction has averaged 0.55, with remarkably little variation when block-averaged into five-year bins (Figure 7.4). Thus, the terrestrial biosphere and the oceans together have consistently removed 45% of fossil CO2 for the last 45 years, and the recent higher rate of atmospheric CO2 increase largely reflects increased fossil fuel emissions. Year-to-year fluctuations in the airborne fraction are associated with major climatic events (see Section The annual increase in 1998, 2.5 ppm, was the highest ever observed, but the airborne fraction (0.82) was no higher than values observed several times in prior decades. The airborne fraction dropped significantly below the average in the early 1990s, and preliminary data suggest it may have risen above the average in 2000 to 2005.

Table 7.1. The global carbon budget (GtC yr–1); errors represent ±1 standard deviation uncertainty estimates and not interannual variability, which is larger. The atmospheric increase (first line) results from fluxes to and from the atmosphere: positive fluxes are inputs to the atmosphere (emissions); negative fluxes are losses from the atmosphere (sinks); and numbers in parentheses are ranges. Note that the total sink of anthropogenic CO2 is well constrained. Thus, the ocean-to-atmosphere and land-to-atmosphere fluxes are negatively correlated: if one is larger, the other must be smaller to match the total sink, and vice versa.

 1980s 1990s 2000–2005c 
TAR TAR reviseda TAR AR4 AR4 
Atmospheric Increaseb  3.3 ± 0.1 3.3 ± 0.1 3.2 ± 0.1 3.2 ± 0.1 4.1 ± 0.1 
Emissions (fossil + cement)c 5.4 ± 0.3 5.4 ± 0.3 6.4 ± 0.4 6.4 ± 0.4 7.2 ± 0.3 
Net ocean-to-atmosphere fluxd  –1.9 ± 0.6 –1.8 ± 0.8 –1.7 ± 0.5 –2.2 ± 0.4 –2.2 ± 0.5 
Net land-to-atmosphere fluxe  –0.2 ± 0.7 –0.3 ± 0.9 –1.4 ± 0.7 –1.0 ± 0.6 –0.9 ± 0.6 
Partitioned as follows           
Land use change flux 1.7 (0.6 to 2.5) 1.4 (0.4 to 2.3) n.a. 1.6 (0.5 to 2.7) n.a. 
Residual terrestrial sink –1.9 (–3.8 to –0.3) –1.7 (–3.4 to 0.2) n.a. –2.6 (–4.3 to –0.9) n.a. 


a TAR values revised according to an ocean heat content correction for ocean oxygen fluxes (Bopp et al., 2002) and using the Fourth Assessment Report (AR4) best estimate for the land use change flux given in Table 7.2.

b Determined from atmospheric CO2 measurements (Keeling and Whorf, 2005, updated by S. Piper until 2006) at Mauna Loa (19°N) and South Pole (90°S) stations, consistent with the data shown in Figure 7.4, using a conversion factor of 2.12 GtC yr–1 = 1 ppm.

c Fossil fuel and cement emission data are available only until 2003 (Marland et al., 2006). Mean emissions for 2004 and 2005 were extrapolated from energy use data with a trend of 0.2 GtC yr–1.

d For the 1980s, the ocean-to-atmosphere and land-to-atmosphere fluxes were estimated using atmospheric O2:N2 and CO2 trends, as in the TAR. For the 1990s, the ocean-to-atmosphere flux alone is estimated using ocean observations and model results (see Section, giving results identical to the atmospheric O2:N2 method (Manning and Keeling, 2006), but with less uncertainty. The net land-to-atmosphere flux then is obtained by subtracting the ocean-to-atmosphere flux from the total sink (and its errors estimated by propagation). For 2000 to 2005, the change in ocean-to-atmosphere flux was modelled (Le Quéré et al., 2005) and added to the mean ocean-to-atmosphere flux of the 1990s. The error was estimated based on the quadratic sum of the error of the mean ocean flux during the 1990s and the root mean square of the five-year variability from three inversions and one ocean model presented in Le Quéré et al. (2003).

e Balance of emissions due to land use change and a residual land sink. These two terms cannot be separated based on current observations.

Figure 7.4

Figure 7.4. Changes in global atmospheric CO2 concentrations. (a) Annual (bars) and five-year mean (lower black line) changes in global CO2 concentrations, from Scripps Institution of Oceanography observations (mean of South Pole and Mauna Loa; Keeling and Whorf, 2005, updated). The upper stepped line shows annual increases that would occur if 100% of fossil fuel emissions (Marland et al., 2006, updated as described in Chapter 2) remained in the atmosphere, and the red line shows five-year mean annual increases from National Oceanic and Atmospheric Administration (NOAA) data (mean of Samoa and Mauna Loa; Tans and Conway, 2005, updated). (b) Fraction of fossil fuel emissions remaining in the atmosphere (‘airborne fraction’) each year (bars), and five-year means (solid black line) (Scripps data) (mean since 1958 is 0.55). Note the anomalously low airborne fraction in the early 1990s.

The inter-hemispheric gradient of CO2 provides additional evidence that the increase in atmospheric CO2 is caused primarily by NH sources. The excess atmospheric CO2 in the NH compared with the Southern Hemisphere (SH), ΔCO2N-S, has increased in proportion to fossil fuel emission rates (which are predominantly in the NH) at about 0.5 ppm per (GtC yr–1) (Figure 7.5). The intercept of the best-fit line indicates that, without anthropogenic emissions, atmospheric CO2 would be 0.8 ppm higher in the SH than in the NH, presumably due to transport of CO2 by the ocean circulation. The consistency of the airborne fraction and the relationship between

ΔCO2N-S and fossil fuel emissions suggest broad consistency in the functioning of the carbon cycle over the period. There are interannual fluctuations in ΔCO2N-S as large as ±0.4 ppm, at least some of which may be attributed to changes in atmospheric circulation (Dargaville et al., 2000), while others may be due to shifts in sources and sinks, such as large forest fires.

Figure 7.5

Figure 7.5. The difference between CO2 concentration in the NH and SH (y axis), computed as the difference between annual mean concentrations (ppm) at Mauna Loa and the South Pole (Keeling and Whorf, 2005, updated), compared with annual fossil fuel emissions (x axis; GtC; Marland, et al., 2006), with a line showing the best fit. The observations show that the north-south difference in CO2 increases proportionally with fossil fuel use, verifying the global impact of human-caused emissions. Fossil fuel and cement emissions

Fossil fuel and cement emissions rose from 5.4 ± 0.3 GtC yr–1 in the 1980s to 6.4 ± 0.4 GtC yr–1 in the 1990s (Marland et al., 2006). They have continued to increase between the 1990s and 2000 to 2005, climbing to 7.2 ± 0.3 GtC yr–1. These numbers are estimated based upon international energy statistics for the 1980 to 2003 period (Marland et al., 2006) with extrapolated trends for 2004 to 2005 (see Table 7.1). The error (±1 standard deviation) for fossil fuel and cement emissions is of the order of 5% globally. Cement emissions are small compared with fossil fuel emissions (roughly 3% of the total). Land use change

During the past two decades, the CO2 flux caused by land use changes has been dominated by tropical deforestation. Agriculture and exploitation of forest resources have reached into formerly remote areas of old growth forest in the tropics, in contrast to mid-latitudes where exploitation previously eliminated most old growth forests. The land use change fluxes reported in this section include explicitly some accumulation of carbon by regrowing vegetation (e.g., Houghton et al., 2000). In the TAR, the global land use flux, adapted from Houghton (1999), was estimated to be 1.7 (0.6–2.5) GtC yr–1 for the 1980s. No estimate was available at the time for the 1990s. This estimate is based on a ‘bookkeeping’ carbon model prescribed with deforestation statistics (Houghton, 1999). A markedly lower estimate of the land use flux in the 1980s (Table 7.2) was obtained by McGuire et al. (2001) from four process-driven terrestrial carbon models, prescribed with changes in cropland area from Ramankutty and Foley (1999). The higher land use emissions of Houghton (2003a) may reflect both the additional inclusion of conversion of forest to pasture and the use of a larger cropland expansion rate than the one of Ramankutty and Foley (1999), as noted by Jain and Yang (2005). Houghton (2003a) updated the land use flux to 2.0 ± 0.8 GtC yr–1 for the 1980s and 2.2 ± 0.8 GtC yr–1 for the 1990s (see Table 7.2). This update gives higher carbon losses from tropical deforestation than those in the TAR (Houghton 2003b).

In addition, DeFries et al. (2002) estimated a tropical land use flux of 0.7 (0.4–1.0) GtC yr–1 for the 1980s and 1.0 (0.5–1.6) GtC yr–1 for the 1990s, using the same bookkeeping approach as Houghton (1999) but driven by remotely sensed data on deforested areas. A similar estimate was independently produced by Achard et al. (2004) for the 1990s, also based on remote sensing. These different land use emissions estimates are reported in Table 7.2. Although the two recent satellite-based estimates point to a smaller source than that of Houghton (2003a), it is premature to say that Houghton’s numbers are overestimated. The land use carbon source has the largest uncertainties in the global carbon budget. If a high value for the land use source is adopted in the global budget, then the residual land uptake over undisturbed ecosystems should be a large sink, and vice versa. For evaluating the global carbon budget, the mean of DeFries et al. (2002) and Houghton (2003a), which both cover the 1980s and the 1990s (Table 7.2), was chosen and the full range of uncertainty is reported. The fraction of carbon emitted by fossil fuel burning, cement production and land use changes that does not accumulate in the atmosphere must be taken up by land ecosystems and by the oceans.

Table 7.2. Land to atmosphere emissions resulting from land use changes during the 1990s and the 1980s (GtC yr–1). The Fourth Assessment Report (AR4) estimates used in the global carbon budget (Table 7.1) are shown in bold. Positive values indicate carbon losses from land ecosystems. Uncertainties are reported as ±1 standard deviation. Numbers in parentheses are ranges of uncertainty.

 Tropical Americas Tropical Africa Tropical Asia Pan-Tropical Non-tropics Total Globe 
Houghton (2003a)a 0.8 ± 0.3 0.4 ± 0.2 1.1 ± 0.5 2.2 ± 0.6 –0.02 ± 0.5  2.2 ± 0.8 
DeFries et al. (2002)b 0.5 (0.2 to 0.7) 0.1 (0.1 to 0.2) 0.4 (0.2 to 0.6) 1.0 (0.5 to 1.6) n.a. n.a. 
Achard et al. (2004)c 0.3 (0.3 to 0.4) 0.2 (0.1 to 0.2) 0.4 (0.3 to 0.5) 0.9 (0.5 to 1.4) n.a. n.a. 
AR4d 0.7 (0.4 to 0.9) 0.3 (0.2 to 0.4) 0.8 (0.4 to 1.1) 1.6 (1.0 to 2.2) –0.02 (–0.5 to +0.5) 1.6 (0.5 to 2.7) 
Houghton (2003a)a 0.8 ± 0.3 0.3 ± 0.2 0.9 ± 0.5 1.9 ± 0.6 0.06 ± 0.5 2.0 ± 0.8 
DeFries et al. (2002)b 0.4 (0.2 to 0.5) 0.1 (0.08 to 0.14) 0.2 (0.1 to 0.3) 0.7 (0.4 to 1.0) n.a. n.a. 
McGuire et al. (2001)e       0.6 to 1.2 –0.1 to +0.4 (0.6 to 1.0) 
Jain and Yang (2005)f 0.22 to 0.24 0.08 to 0.48 0.58 to 0.34 1.33 to 2.06 
TARg           1.7 (0.6 to 2.5) 
AR4d 0.6 (0.3 to 0.8) 0.2 (0.1 to 0.3) 0.6 (0.3 to 0.9) 1.3 (0.9 to 1.8) 0.06 (–0.4 to +0.6) 1.4 (0.4 to 2.3) 


a His Table 2.

b Their Table 3.

c Their Table 2 for mean estimates with the range indicated in parentheses corresponding to their reported minimum and maximum estimates.

d Best estimate calculated from the mean of Houghton (2003a) and DeFries et al. (2002), the only two studies covering both the 1980s and the 1990s. For non-tropical regions where DeFries et al. have no estimate, Houghton has been used.

e Their Table 5; range is obtained from four terrestrial carbon models.

f The range indicated in parentheses corresponds to two simulations using the same model, but forced with different land cover change datasets from Houghton (2003a) and DeFries et al. (2002).

g In the TAR estimate, no values were available for the 1990s.