Anthropogenic and biogenic winter sources of Arctic CO2—a model study

Long‐range transport of anthropogenic and biogenic CO₂ to a remote site in the Arctic is studied. A limited area, off‐line, Eulerian atmospheric transport model is used, and the results are compared to the observed CO₂ concentration at the "Ny‐Alesund International Arctic Research and Monitoring Facility". Inventories of anthropogenic CO₂ emissions and estimates of biogenic CO₂ emissions are used to investigate the respective impact of these emissions on Arctic CO₂ variations during 4 winter months. A direct comparison of the modelled and observed concentrations reveals remarkably good timing in the modelled variations as compared to the observed variations for most of the time. The correlation of observed versus modelled CO₂ concentration is significant at the 95% confidence level. The biogenic and the anthropogenic CO₂ emissions are shown to have approximately equal influence on Arctic CO₂ variations during winter. Europe is found to be the dominant source of anthropogenic CO₂ at the monitoring station, while Siberia and Northern America have little influence on Arctic CO₂, during the months studied. These results contradict Engardt and Holmén whose results indicate that the lower‐Ob region in western Siberia has a large impact on Arctic CO₂.     

一个地球系统模式模拟的气候变化对海洋碳循环的影响

使用维多利亚大学的地球系统模式进行模拟,选取1800-2500年间较高的CO₂浓度情景（RCP8.5）,分析由于CO₂增加引起的气候变化对海洋碳循环的影响。当气候敏感度为3.0 K时,相对于无气候变化,到2100年,由于大气CO₂增加造成的气候变化导致海表面温度升高2.7 K,北大西洋深水流量减少4.5 Sv,海洋对人为碳的年吸收减少0.8 Pg C;比较人为溶解无机碳在海洋中的垂直累积分布,发现气候变化对海洋吸收大气CO₂的影响在北大西洋区域最明显。1800-2500年,相对于不考虑气候变化的情景,模式模拟的气候变化导致整个海洋对人为碳的累积吸收总量减少23.1%,其中北大西洋减少32.0%。此外,比较不同气候敏感度（0～4.5 K,间隔为0.5 K）的模拟结果发现,气候敏感度越高,气候变化对海洋吸收CO₂能力的抑制作用越明显。     

Determination of microbial versus root‐produced CO2 in an agricultural ecosystem by means of δ13CO2 measurements in soil air

The amounts of microbial and root‐respired CO₂ in a maize/winter wheat agricultural system in south western Germany were investigated by measurements of the CO₂ mixing ratio and the ¹³C/¹²C ratio in soil air. CO₂ fluxes at the soil surface for the period of investigation (1993–1995) were also determined. Root respired CO₂ shows a strong correlation with the plant mass above ground surface of the respective vegetation (R²≥0.88); the maximum CO₂ release from roots was in August for the maize (2.0±0.5 mmol m⁻² h⁻¹) and in June for winter wheat (1.5±0.5 mmol m⁻² h⁻¹). Maximum CO₂ production by roots correlate well with the maximum amount of plant root matter. Integrating the CO₂ production over the whole growing season and normalizing to the dry root matter yields, the CO₂ production per gram dry organic root matter (DORM) of maize was found to be 0.14±0.03 gC (g DORM)⁻¹. At the sites investigated, root‐produced CO₂ contributed (16±4)% for maize, and (24±4)% for winter wheat, respectively, to the total annual CO₂ production in the soil (450±50 gC m⁻² for maize, 210±30 gC m⁻² for winter wheat).     

The YAK-AEROSIB transcontinental aircraft campaigns: new insights on the transport of CO2, CO and O3 across Siberia

Two airborne campaigns were carried out to measure the tropospheric concentrations and variability of CO₂, CO and O₃ over Siberia. In order to quantify the influence of remote and regional natural and anthropogenic sources, we analysed a total of 52 vertical profiles of these species collected in April and September 2006, every ∼200 km and up to 7 km altitude. CO₂ and CO concentrations were high in April 2006 (respectively 385–390 ppm CO₂ and 160–200 ppb CO) compared to background values. CO concentrations up to 220 ppb were recorded above 3.5 km over eastern Siberia, with enhancements in 500–1000 m thick layers. The presence of CO enriched air masses resulted from a quick frontal uplift of a polluted air mass exposed to northern China anthropogenic emissions and to fire emissions in northern Mongolia. A dominant Asian origin for CO above 4 km (71.0%) contrasted with a dominant European origin below this altitude (70.9%) was deduced both from a transport model analysis, and from the contrasted ΔCO/ΔCO₂ ratio vertical distribution. In September 2006, a significant O₃ depletion (∼–30 ppb) was repeatedly observed in the boundary layer, as diagnosed from virtual potential temperature profiles and CO₂ gradients, compared to the free troposphere aloft, suggestive of a strong O₃ deposition over Siberian forests.     

Excess 210Po in the coastal atmosphere

Many researchers have reported the widespread occurrence of excess ²¹⁰ Po in the global atmosphere and suggested probable sources such as resuspension of top soils, stratospheric aerosols, sea spray of the surface micro‐layer, volcanic emission, and bio‐volatile ²¹⁰Po species from the productive ocean. We have observed excess ²¹⁰Po on aerosols in the coastal atmosphere of the Chesapeake and Delaware Bays. On‐board measurements in the Chesapeake Bay atmosphere show that the increase of this excess ²¹⁰Po is dependent upon wind speed. Simultaneously measured activity ratios of ⁷Be/²¹⁰Pb and ²¹⁰Pb/²²²Rn argue against either higher altitude air or continental soils as the source of this excess. We hypothesize that the excess ²¹⁰Po originates mainly from surface waters either by the sea‐spray of the surface microlayer, or more likely, by gas exchange. We conclude gas exchange as the mechanism since the polonium excess increases linearly with wind speed over a threshold of 3 m s⁻¹(mean) similar to other gases (i.e., CO₂, SF₆ , and DMS). In addition, higher ²¹⁰Po excess with lower ²²²Rn is observed in on‐shore marine air at Lewes, DE. This suggests sea‐air exchange of volatile Po along with other bio‐volatile species (i.e., DMS, DMSe, and MMHg) in the coastal productive ocean during high wind speeds.     

Aircraft observation of carbon dioxide at 8–13 km altitude over the western Pacific from 1993 to 1999

Abstract The spatial and temporal variations of atmospheric CO₂ at 8–13 km from April 1993 to April 1999 were observed by measuring CO₂ mixing ratios in samples collected biweekly from a commercial airliner between Australia and Japan. The CO₂ growth rate showed a considerable interannual variation, with a maximum of about 3 ppm yr⁻¹ during late 1997. This variation is related to the El Niño/Southern Oscillation (ENSO) events. A year-to-year change related to the ENSO events was also found in the latitudinal distribution pattern of the CO₂ annual mean between 30°N and 30°S. The averaged CO₂ seasonal cycle in the Northern Hemisphere gradually decayed toward the equator, and a relatively complicated variation with a double seasonal maximum appeared in the Southern Hemisphere. A significant yearly change of the seasonal cycle pattern was observed in the Southern Hemisphere. The impact of a tropical biomass-burning injection on the upper tropospheric CO₂ was estimated on the basis of the CO data from the same airliner observation.     

Vertical profiles of biospheric and fossil fuel-derived CO2 and fossil fuel CO2 : CO ratios from airborne measurements of Δ14C, CO2 and CO above Colorado, USA

Measurements of  Δ¹⁴C  in atmospheric CO₂ are an effective method of separating CO₂ additions from fossil fuel and biospheric sources or sinks of CO₂. We illustrate this technique with vertical profiles of CO₂ and  Δ¹⁴C  analysed in whole air flask samples collected above Colorado, USA in May and July 2004. Comparison of lower tropospheric composition to cleaner air at higher altitudes (>5 km) revealed considerable additions from respiration in the morning in both urban and rural locations. Afternoon concentrations were mainly governed by fossil fuel emissions and boundary layer depth, also showing net biospheric CO₂ uptake in some cases. We estimate local industrial CO₂:CO emission ratios using in situ measurements of CO concentration. Ratios are found to vary by 100% and average 57 mole CO₂:1 mole CO, higher than expected from emissions inventories. Uncertainty in CO₂ from different sources was ±1.1 to ±4.1 ppm for addition or uptake of −4.6 to 55.8 ppm, limited by  Δ¹⁴C  measurement precision and uncertainty in background  Δ¹⁴C  and CO₂ levels.     

A recent build-up of atmospheric CO2 over Europe. Part 1: observed signals and possible explanations

We analysed interannual and decadal changes in the atmospheric CO₂ concentration gradient (ΔCO₂) between Europe and the Atlantic Ocean over the period 1995–2007. Fourteen measurement stations are used, with Mace-Head being used to define background conditions. The variability of ΔCO₂ reflects fossil fuel emissions and natural sinks activity over Europe, as well as atmospheric transport variability. The mean ΔCO₂ increased by 1–2 ppm at Eastern European stations (∼30% growth), between 1990–1995 and 2000–2005. This built up of CO₂ over the continent is predominantly a winter signal. If the observed increase of ΔCO₂ is explained by changes in ecosystem fluxes, a loss of about 0.46 Pg C per year would be required during 2000–2005. Even if severe droughts have impacted Western Europe in 2003 and 2005, a sustained CO₂ loss of that magnitude is unlikely to be true. We sought alternative explanations for the observed CO₂ build-up into transport changes and into regional redistribution of fossil fuel CO₂ emissions. Boundary layer heights becoming shallower can only explain 32% of the variance of the signal. Regional changes of emissions may explain up to 27% of the build-up. More insights are given in the Aulagnier et al. companion paper.     

Reduction of biosphere life span as a consequence of geodynamics

The long‐term co‐evolution of the geosphere–biospere complex from the Proterozoic up to 1.5 billion years into the planet's future is investigated using a conceptual earth system model including the basic geodynamic processes. The model focusses on the global carbon cycle as mediated by life and driven by increasing solar luminosity and plate tectonics. The main CO₂ sink, the weathering of silicates, is calculated as a function of biologic activity, global run‐off and continental growth. The main CO₂ source, tectonic processes dominated by sea‐floor spreading, is determined using a novel semi‐empirical scheme. Thus, a geodynamic extension of previous geostatic approaches can be achieved. As a major result of extensive numerical investigations, the "terrestrial life corridor", i.e., the biogeophysical domain supporting a photosynthesis‐based ecosphere in the planetary past and in the future, can be identified. Our findings imply, in particular, that the remaining life‐span of the biosphere is considerably shorter (by a few hundred million years) than the value computed with geostatic models by other groups. The "habitable‐zone concept" is also revisited, revealing the band of orbital distances from the sun warranting earth‐like conditions. It turns out that this habitable zone collapses completely in some 1.4 billion years from now as a consequence of geodynamics.     

Tropospheric carbon dioxide concentrations at a northern boreal site in Finland: basic variations and source areas

Diurnal and annual variations of CO₂, O_3, SO₂, black carbon and condensation nuclei and their source areas were studied by utilizing air parcel trajectories and tropospheric concentration measurements at a boreal GAW site in Pallas, Finland. The average growth trend of CO₂ was about 2.5 ppm yr⁻¹ according to a 4-yr measurement period starting in October 1996. The annual cycle of CO₂ showed concentration difference of about 19 ppm between the summer minimum and winter maximum. The diurnal cycle was most pronounced during July and August. The variation between daily minimum and maximum was about 5 ppm. There was a diurnal cycle in aerosol concentrations during spring and summer. Diurnal variation in ozone concentrations was weak. According to trajectory analysis the site was equally affected by continental and marine air masses. During summer the contribution of continental air increased, although the southernmost influences decreased. During daytime in summer the source areas of CO₂ were mainly located in the northern parts of the Central Europe, while during winter the sources were more evenly distributed. Ozone showed similar source areas during summer, while during winter, unlike CO₂, high concentrations were observed in air arriving from the sea. Sulfur dioxide sources were more northern (Kola peninsula and further east) and CO₂ sources west-weighted in comparison to sources of black carbon. Source areas of black carbon were similar to source areas of aerosols during winter. Aerosol source area distributions showed signs of marine sources during spring and summer.     

Potential impact of climate change on marine dimethyl sulfide emissions

Dimethyl sulfide (DMS) is a biogenic compound produced in sea-surface water and outgased to the atmosphere. Once in the atmosphere, DMS is a significant source of cloud condensation nuclei in the unpolluted marine atmosphere. It has been postulated that climate may be partly modulated by variations in DMS production through a DMS-cloud condensation nuclei-albedo feedback. We present here a modelled estimation of the response of DMS sea-water concentrations and DMS fluxes to climate change, following previous work on marine DMS modeling ( Aumont et al., 2002 ) and on the global warming impact on marine biology ( Bopp et al., 2001 ). An atmosphere–ocean general circulation model (GCM) was coupled to a marine biogeochemical scheme and used without flux correction to simulate climate response to increased greenhouse gases (a 1% increase per year in atmospheric CO₂ until it has doubled). The predicted global distribution of DMS at  1 × CO₂  compares reasonably well with observations; however, in the high latitudes, very elevated concentrations of DMS due to spring and summer blooms of Phaeocystis can not be reproduced. At  2 × CO₂  , the model estimates a small increase of global DMS flux to the atmosphere (+2%) but with large spatial heterogeneities (from −15% to +30% for the zonal mean). Mechanisms affecting DMS fluxes are changes in (1) marine biological productivity, (2) relative abundance of phytoplankton species and (3) wind intensity. The mean DMS flux perturbation we simulate represents a small negative feedback on global warming; however, the large regional changes may significantly impact regional temperature and precipitation patterns.     

Simulating effects of land use changes on carbon fluxes: past contributions to atmospheric CO2 increases and future commitments due to losses of terrestrial sink capacity

The impact of land use on the global carbon cycle and climate is assessed. The Bern carbon cycle-climate model was used with land use maps from HYDE3.0 for 1700 to 2000 A.D. and from post-SRES scenarios for this century. Cropland and pasture expansion each cause about half of the simulated net carbon emissions of 188 Gt C over the industrial period and 1.1 Gt C yr⁻¹ in the 1990s, implying a residual terrestrial sink of 113 Gt C and of 1.8 Gt C yr⁻¹, respectively. Direct CO₂ emissions due to land conversion as simulated in book-keeping models dominate carbon fluxes due to land use in the past. They are, however, mitigated by 25% through the feedback of increased atmospheric CO₂ stimulating uptake. CO₂ stimulated sinks are largely lost when natural lands are converted. Past land use change has eliminated potential future carbon sinks equivalent to emissions of 80–150 Gt C over this century. They represent a commitment of past land use change, which accounts for 70% of the future land use flux in the scenarios considered. Pre-industrial land use emissions are estimated to 45 Gt C at most, implying a maximum change in Holocene atmospheric CO₂ of 3 ppm. This is not compatible with the hypothesis that early anthropogenic CO₂ emissions prevented a new glacial period.     

Dimethylsulphide production in the subantarctic southern ocean under enhanced greenhouse conditions

Dimethylsulphide (DMS) is an important sulphur‐containing trace gas produced by enzymatic cleavage of its precursor compound, dimethylsulphoniopropionate (DMSP), which is released by marine phytoplankton in the upper ocean. After ventilation to the atmosphere, DMS is oxidised to form sulphate aerosols which in the unpolluted marine atmosphere are a major source of cloud condensation nuclei (CCN). Because the micro‐physical properties of clouds relevant to climate change are sensitive to CCN concentration in air, it has been postulated that marine sulphur emissions may play a rôle in climate regulation. The Subantarctic Southern Ocean (41–53°S) is relatively free of anthropogenic sulphur emissions, thus sulphate aerosols will be mainly derived from the biogenic source of DMS, making it an ideal region in which to evaluate the DMS‐climate regulation hypothesis. We have extended a previous modelling analysis of the DMS cycle in this region by employing a coupled general circulation model (CGCM) which has been run in transient mode to provide a more realistic climate scenario. The CGCM output provided meteorological data under the IPCC/IS92a radiative forcing scenario. A DMS production model has been forced with the CGCM climate data to simulate the trend in the sea‐to‐air DMS flux for the period 1960 to 2080, corresponding to equivalent CO₂ tripling relative to pre‐industrial levels. The results confirm a minor but non‐negligible increase in DMS flux in this region, in the range +1% to +6% predicted over the period simulated. Uncertainty analysis of the DMS model predictions have confirmed the positive sign for the change in DMS flux, that is a negative DMS feedback on warming.     

Future changes of the atmospheric composition and the impact of climate change

The development of the future atmospheric chemical composition is investigated with respect to NO _y and O₃ by means of the off‐line coupled dynamic‐chemical general circulation model ECHAM3/CHEM. Two time slice experiments have been performed for the years 1992 and 2015, which include changes in sea surface temperatures, greenhouse gas concentrations, emissions of CFCs, NO _x and other species, i.e., the 2015 simulation accounts for changes in chemically relevant emissions and for a climate change and its impact on air chemistry. The 2015 simulation clearly shows a global increase in ozone except for large areas of the lower stratosphere, where no significant changes or even decreases in the ozone concentration are found. For a better understanding of the importance of (A) emissions like NO _x and CFCs, (B) future changes of air temperature and water vapour concentration, and (C) other dynamical parameters, like precipitation and changes in the circulation, diabatic circulation, stratosphere‐troposphere‐exchange, the simulation of the future atmosphere has been performed stepwise. This method requires a climate‐chemistry model without interactive coupling of chemical species. Model results show that the direct effect of emissions (A) plays a major rôle for the composition of the future atmosphere, but they also clearly show that climate change (B and C) has a significant impact and strongly reduces the NO _y and ozone concentration in the lower stratosphere.     

Long-term effects of anthropogenic CO<Subscript>2</Subscript> emissions simulated with a complex earth system model

A new complex earth system model consisting of an atmospheric general circulation model, an ocean general circulation model, a three-dimensional ice sheet model, a marine biogeochemistry model, and a dynamic vegetation model was used to study the long-term response to anthropogenic carbon emissions. The prescribed emissions follow estimates of past emissions for the period 1751–2000 and standard IPCC emission scenarios up to the year 2100. After 2100, an exponential decrease of the emissions was assumed. For each of the scenarios, a small ensemble of simulations was carried out. The North Atlantic overturning collapsed in the high emission scenario (A2) simulations. In the low emission scenario (B1), only a temporary weakening of the deep water formation in the North Atlantic is predicted. The moderate emission scenario (A1B) brings the system close to its bifurcation point, with three out of five runs leading to a collapsed North Atlantic overturning circulation. The atmospheric moisture transport predominantly contributes to the collapse of the deep water formation. In the simulations with collapsed deep water formation in the North Atlantic a substantial cooling over parts of the North Atlantic is simulated. Anthropogenic climate change substantially reduces the ability of land and ocean to sequester anthropogenic carbon. The simulated effect of a collapse of the deep water formation in the North Atlantic on the atmospheric CO₂ concentration turned out to be relatively small. The volume of the Greenland ice sheet is reduced, but its contribution to global mean sea level is almost counterbalanced by the growth of the Antarctic ice sheet due to enhanced snowfall. The modifications of the high latitude freshwater input due to the simulated changes in mass balance of the ice sheet are one order of magnitude smaller than the changes due to atmospheric moisture transport. After the year 3000, the global mean surface temperature is predicted to be almost constant due to the compensating effects of decreasing atmospheric CO₂ concentrations due to oceanic uptake and delayed response to increasing atmospheric CO₂ concentrations before.     

Model analysis of the influence of gas diffusivity in soil on CO and H2 uptake

CO and H₂ uptake by soil was studied as a diffusion process. A diffusion model was used to determine how the surface fluxes (net deposition velocities) were controlled by in‐situ microbial uptake rates and soil gas diffusivity calculated from the 3‐phase system (solid, liquid, gas) in the soil. Analytical solutions of the diffusion model assuming vertical uniformity of soil properties showed that physical properties such as air‐filled porosity and soil gas diffusivity were more important in the uptake process than in the emission process. To incorporate the distribution of in‐situ microbial uptake, we used a 2‐layer model incorporating "a microbiologically inactive layer and an active layer" as suggested from experimental results. By numerical simulation using the 2‐layer model, we estimated the effect of several factors on deposition velocities. The variations in soil gas diffusivity due to physical properties, i.e., soil moisture and air‐filled porosity, as well as to the depth of the inactive layer and in‐situ microbial uptake, were found to be important in controlling deposition velocities. This result shows that the diffusion process in soil is critically important for CO and H₂ uptake by soil, at least in soils with higher in‐situ uptake rates and/or with large variation in soil moisture. Similar uptake rates and the difference in deposition velocity between CO and H₂ may be attributable to differences in CO and H₂ molecular diffusivity. The inactive layer is resistant to diffusion and creates uptake limits in CO and H₂ by soil. The coupling of high temperature and a thick inactive layer, common in arid soils, markedly lowers net CO deposition velocity. The temperature for maximum uptake of CO changes with depth of the inactive layer.     

Climate sensitivity to cloud optical properties

A radiative–convective model was developed to investigate the sensitivity of climate to cloud optical properties and the related feedback processes. This model demonstrates that the Earth's surface temperature increases with cloud optical depth when the clouds are very thin but decreases with cloud optical depth when the cloud shortwave (solar) radiative forcing is larger than the cloud longwave (terrestrial) radiative forcing. When clouds are included in the model, the magnitude of the greenhouse effect due to a doubling of the CO₂ concentration varies with the cloudoptical depth: the thicker the clouds, the weaker the greenhouse warming. In addition, a small variation in the cloud droplet size has a larger impact on the equilibrium state temperature in the lower atmosphere than the warming caused by a doubling of the CO₂ concentration: a 2% increase in the average cloud droplet size per degree increase in temperature doubles the warming caused by the doubling of the CO₂ concentration. These findings suggest that physically reliable correlations between the cloud droplet size and macrophysical meteorological variables such as temperature, wind and water vapor fields are needed on a global climate scale to assess the climate impact of increases in greenhouse gases.     

Solar irradiance during the last 1200 years based on cosmogenic nuclides

Based on a quantitative study of the common fluctuations of ¹⁴C and ¹⁰Be production rates, we have derived a time series of the solar magnetic variability over the last 1200 years. This record is converted into irradiance variations by linear scaling based on previous studies of sun‐like stars and of the sun's behavior over the last few centuries. The new solar irradiance record exhibits low values during the well‐known solar minima centered at about 1900, 1810 (Dalton) and 1690 ad (Maunder). Further back in time, a rather long period between 1450 and 1750 ad is characterized by low irradiance values. A shorter period is centered at about 1200 ad , with irradiance slightly higher or similar to present day values. It is tempting to correlate these periods with the so‐called "little ice age" and "medieval warm period", respectively. An accurate quantification of the climatic impact of this new irradiance record requires the use of coupled atmosphere–ocean general circulation models (GCMs). Nevertheless, our record is already compatible with a global cooling of about 0.5‐1°C during the "little ice age", and with a general cooling trend during the past millenium followed by global warming during the 20th century (Mann et al., 1999).     

Short-wave aerosol radiative efficiency over the global oceans derived from satellite data

Using 5 yr (December 2000–November 2005) of satellite data from the clouds and the earths radiant energy system (CERES) and moderate resolution imaging spectroradiometer (MODIS), we examine the instantaneous short-wave radiative efficiency ( E_τ ) of aerosols during the morning Terra satellite overpass time over the global oceans (60°N–60°S). We calculate E_τ using two commonly used methods. The first method uses the MODIS aerosol optical thickness (AOT) at 0.55 μm with radiative transfer calculations, whereas the second method utilizes the same AOT values along with a new generation of aerosol angular distribution models to convert the CERES-measured broad-band radiances to fluxes. Over the 5 yr, the global mean instantaneous E_τ between the methods is remarkably consistent and within 5 W m⁻²τ⁻¹ with a mean value of –70 W m⁻²τ⁻¹. The largest differences between the methods occur in high-latitude regions, primarily in the Southern Hemisphere, where AOT is low. In dust dominated regions, there is an excellent agreement between the methods with differences of <3 W m⁻²τ⁻¹. These differences are largely due to assumptions in aerosol models and definition of clear sky backgrounds. Independent assessments of aerosol radiative effects from different satellite sensors and methods are extremely valuable and should be used to verify numerical modelling simulations.     

Reconciliation of excess 14C-constrained global CO2 piston velocity estimates

Oceanic excess radiocarbon data is widely used as a constraint for air–sea gas exchange. However, recent estimates of the global mean piston velocity  〈 k 〉  from Naegler et al., Krakauer et al., Sweeney et al. and Müller et al. differ substantially despite the fact that they all are based on excess radiocarbon data from the GLODAP data base. Here I show that these estimates of  〈 k 〉  can be reconciled if first, the changing oceanic radiocarbon inventory due to net uptake of CO₂ is taken into account; second, if realistic reconstructions of sea surface  Δ¹⁴C  are used and third, if  〈 k 〉  is consistently reported with or without normalization to a Schmidt number of 660. These corrections applied, unnormalized estimates of  〈 k 〉  from these studies range between 15.1 and 18.2 cm h⁻¹. However, none of these estimates can be regarded as the only correct value for  〈 k 〉  . I thus propose to use the 'average' of the corrected values of  〈 k 〉  presented here (16.5 ± 3.2 cm h⁻¹) as the best available estimate of the global mean unnormalized piston velocity  〈 k 〉  , resulting in a gross ocean-to-atmosphere CO₂ flux of 76 ± 15 PgC yr⁻¹ for the mid-1990s.