2.1.1 Emission estimates

The estimates of global and national fossil CO₂ emissions (E_FF) include the combustion of fossil fuels through a wide range of activities (e.g. transport, heating, and cooling, industry, fossil industry's own use, and gas flaring), the production of cement, and other process emissions (e.g. the production of chemicals and fertilisers). The estimates of E_FF rely primarily on energy consumption data, specifically data on hydrocarbon fuels, collated and archived by several organisations (Andres et al., 2012). We use four main data sets for historical emissions (1751–2017).

1. We use global and national emission estimates for coal, oil, and gas from CDIAC for the time period of 1751–2014 (Boden et al., 2017), as it is the only data set that extends back to 1751 by country.

2. We use official UNFCCC national inventory reports for 1990–2016 for the 42 Annex I countries in the UNFCCC (UNFCCC, 2018). We assess these to be the most accurate estimates because they are compiled by experts within countries that have access to detailed energy data, and they are periodically reviewed.

3. We use the BP Statistical Review of World Energy (BP, 2018), as these are the most up-to-date estimates of national energy statistics.

4. We use global and national cement emissions updated from Andrew (2018), which include revised emission factors.

In the following section we provide more details for each data set and describe the additional modifications that are required to make the data set consistent and usable.

• CDIAC. The CDIAC estimates have been updated annually to the year 2014, derived primarily from energy statistics published by the United Nations (UN, 2017b). Fuel masses and volumes are converted to fuel energy content using country-level coefficients provided by the UN and then converted to CO₂ emissions using conversion factors that take into account the relationship between carbon content and energy (heat) content of the different fuel types (coal, oil, gas, gas flaring) and the combustion efficiency (Marland and Rotty, 1984).

• UNFCCC. Estimates from the UNFCCC national inventory reports follow the IPCC guidelines (IPCC, 2006) but have a slightly larger system boundary than CDIAC by including emissions coming from carbonates other than in cement manufacturing. We reallocate the detailed UNFCCC estimates to the CDIAC definitions of coal, oil, gas, cement, and other to allow consistent comparisons over time and among countries.

• BP. For the most recent period when the UNFCCC (2018) and CDIAC (2015–2017) estimates are not available, we generate preliminary estimates using the BP Statistical Review of World Energy (Andres et al., 2014; Myhre et al., 2009; BP, 2018). We apply the BP growth rates by fuel type (coal, oil, gas) to estimate 2017 emissions based on 2016 estimates (UNFCCC) and to estimate 2015–2017 emissions based on 2014 estimates (CDIAC). BP's data set explicitly covers about 70 countries (96 % of global emissions), and for the remaining countries we use growth rates from the subregion the country belongs to. For the most recent years, flaring is assumed constant from the most recent available year of data (2016 for countries that report to the UNFCCC, 2014 for the remainder).

• Cement. Estimates of emissions from cement production are taken directly from Andrew (2018). Additional calcination and carbonation processes are not included explicitly here, except in national inventories provided by UNFCCC, but are discussed in Sect. 2.8.2.

• Country mappings. The published CDIAC data set includes 256 countries and regions. This list includes countries that no longer exist, such as the USSR and Yugoslavia. We reduce the list to 213 countries by reallocating emissions to the currently defined territories, using mass-preserving aggregation or disaggregation. Examples of aggregation include merging East and West Germany to the currently defined Germany. Examples of disaggregation include reallocating the emissions from the former USSR to the resulting independent countries. For disaggregation, we use the emission shares when the current territories first appeared, and thus historical estimates of disaggregated countries should be treated with extreme care. In addition, we aggregate some overseas territories (e.g. Réunion, Guadeloupe) into their governing nations (e.g. France) to align with UNFCCC reporting.

• Global total. Our global estimate is based on CDIAC for fossil fuel combustion plus Andrew (2018) for cement emissions. This is greater than the sum of emissions from all countries. This is largely attributable to emissions that occur in international territory, in particular, the combustion of fuels used in international shipping and aviation (bunker fuels). The emissions from international bunker fuels are calculated based on where the fuels were loaded, but we do not include them in the national emission estimates. Other differences occur (1) because the sum of imports in all countries is not equal to the sum of exports, and (2) because of inconsistent national reporting, differing treatment of oxidation of non-fuel uses of hydrocarbons (e.g. as solvents, lubricants, feedstocks), and (3) because of changes in fuel stored (Andres et al., 2012).

2.2.1 Emissions embodied in goods and services

CDIAC, UNFCCC, and BP national emission statistics “include greenhouse gas emissions and removals taking place within national territory and offshore areas over which the country has jurisdiction” (Rypdal et al., 2006) and are called territorial emission inventories. Consumption-based emission inventories allocate emissions to products that are consumed within a country and are conceptually calculated as the territorial emissions minus the “embodied” territorial emissions to produce exported products plus the emissions in other countries to produce imported products (consumption = territorial − exports + imports). Consumption-based emission attribution results (e.g. Davis and Caldeira, 2010) provide additional information to territorial-based emissions that can be used to understand emission drivers (Hertwich and Peters, 2009) and quantify emission transfers by the trade of products between countries (Peters et al., 2011b). The consumption-based emissions have the same global total but reflect the trade-driven movement of emissions across the Earth's surface in response to human activities.

We estimate consumption-based emissions from 1990 to 2016 by enumerating the global supply chain using a global model of the economic relationships between economic sectors within and among every country (Andrew and Peters, 2013; Peters et al., 2011a). Our analysis is based on the economic and trade data from the Global Trade and Analysis Project (GTAP; Narayanan et al., 2015), and we make detailed estimates for the years 1997 (GTAP version 5), 2001 (GTAP6), and 2004, 2007, and 2011 (GTAP9.2), covering 57 sectors and 141 countries and regions. The detailed results are then extended into an annual time series from 1990 to the latest year of the gross domestic product (GDP) data (2016 in this budget), using GDP data by expenditure in the current exchange rate of US dollars (USD; from the UN National Accounts Main Aggregrates Database; UN, 2017a) and time series of trade data from GTAP (based on the methodology in Peters et al., 2011b). We estimate the sector-level CO₂ emissions using the GTAP data and methodology, include flaring and cement emissions from CDIAC, and then scale the national totals (excluding bunker fuels) to match the emission estimates from the carbon budget. We do not provide a separate uncertainty estimate for the consumption-based emissions, but based on model comparisons and sensitivity analysis, they are unlikely to be significantly different than for the territorial emission estimates (Peters et al., 2012a).

2.2.2 Growth rate in emissions

We report the annual growth rate in emissions for adjacent years (in per cent per year) by calculating the difference between the two years and then normalising to the emissions in the first year: $\left(E_{\mathrm{FF}}\right(t_{\mathrm{0}+\mathrm{1}})-E_{\mathrm{FF}}(t_{\mathrm{0}}\left)\right)/E_{\mathrm{FF}}\left(t_{\mathrm{0}}\right)\times \mathrm{100}\mathrm{\%}\times \mathrm{100}/\left(\mathrm{1}\mathrm{year}\right)$. $\times \mathrm{100}/\left(\mathrm{1}\text{year}\right)$. We apply a leap-year adjustment when relevant to ensure valid interpretations of annual growth rates. This affects the growth rate by about 0.3 % yr⁻¹ (1∕365) and causes growth rates to go up approximately 0.3 % if the first year is a leap year and down 0.3 % if the second year is a leap year.

The relative growth rate of E_FF over time periods of greater than 1 year can be rewritten using its logarithm equivalent as follows:

Here we calculate relative growth rates in emissions for multi-year periods (e.g. a decade) by fitting a linear trend to ln(E_FF) in Eq. (2), reported in per cent per year.

2.2.3 Emission projections

To gain insight into emission trends for the current year (2018), we provide an assessment of global fossil CO₂ emissions, E_FF, by combining individual assessments of emissions for China, the US, the EU, and India (the four countries/regions with the largest emissions), and the rest of the world.

Our 2018 estimate for China uses (1) the sum of domestic production (NBS, 2018b) and net imports (General Administration of Customs of the People's Republic of China, 2018) for coal, oil and natural gas, and production of cement (NBS, 2018b) from preliminary statistics for January through September of 2018 and (2) historical relationships between January–September statistics for both production and imports and full-year statistics for consumption using final data for 2000–2016 (NBS, 2015, 2017) and preliminary data for 2017 (NBS, 2018a). See also Liu et al. (2018) and Jackson et al. (2018) for details. The uncertainty is based on the variance of the difference between the January–September and full-year data from historical data, as well as typical variance in the preliminary full-year data used for 2017 and typical changes in the energy content of coal for the period of 2013–2016 (NBS, 2017, 2015). We note that developments for the final 3 months this year may be atypical due to the ongoing trade disputes between China and the US, and this additional uncertainty has not been quantified. Results and uncertainties are discussed further in Sect. 3.4.1.

For the US, we use the forecast of the U.S. Energy Information Administration (EIA) for emissions from fossil fuels (EIA, 2018). This is based on an energy forecasting model which is updated monthly (last update to October) and takes into account heating-degree days, household expenditures by fuel type, energy markets, policies, and other effects. We combine this with our estimate of emissions from cement production using the monthly US cement data from the U.S. Geological Survey (USGS) for January–August, assuming changes in cement production over the first part of the year apply throughout the year. While the EIA's forecasts for current full-year emissions have on average been revised downwards, only 10 such forecasts are available, so we conservatively use the full range of adjustments following revision and additionally assume symmetrical uncertainty to give ±2.5 % around the central forecast.

For India, we use (1) monthly coal production and sales data from the Ministry of Mines (2018), Coal India Limited (CIL, 2018), and Singareni Collieries Company Limited (SCCL, 2018), combined with import data from the Ministry of Commerce and Industry (MCI, 2018) and power station stocks data from the Central Electricity Authority (CEA, 2018); (2) monthly oil production and consumption data from the Ministry of Petroleum and Natural Gas (PPAC, 2018a); (3) monthly natural gas production and import data from the Ministry of Petroleum and Natural Gas (PPAC, 2018b); and (4) monthly cement production data from the Office of the Economic Advisor (OEA, 2018). All data were available for January to September or October. We use Holt–Winters exponential smoothing with multiplicative seasonality (Chatfield, 1978) on each of these four emission series to project to the end of the current year. This iterative method produces estimates of both trend and seasonality at the end of the observation period that are a function of all prior observations, weighted most strongly to more recent data, while maintaining some smoothing effect. The main source of uncertainty in the projection of India's emissions is the assumption of continued trends and typical seasonality.

For the EU, we use (1) monthly coal supply data from Eurostat for the first 6–9 months of the year (Eurostat, 2018) cross-checked with more recent data on coal-generated electricity from ENTSO-E for January through October (ENTSO-E, 2018); (2) monthly oil and gas demand data for January through August from the Joint Organisations Data Initiative (JODI, 2018); and (3) cement production assumed to be stable. For oil and gas emissions we apply the Holt–Winters method separately to each country and energy carrier to project to the end of the current year, while for coal – which is much less strongly seasonal because of strong weather variations – we assume the remaining months of the year are the same as the previous year in each country.

For the rest of the world, we use the close relationship between the growth in GDP and the growth in emissions (Raupach et al., 2007) to project emissions for the current year. This is based on a simplified Kaya identity, whereby E_FF (GtC yr⁻¹) is decomposed by the product of GDP (USD yr⁻¹) and the fossil fuel carbon intensity of the economy (I_FF; GtC USD⁻¹) as follows:

Taking a time derivative of Eq. (3) and rearranging gives

where the left-hand term is the relative growth rate of E_FF, and the right-hand terms are the relative growth rates of GDP and I_FF, respectively, which can simply be added linearly to give the overall growth rate.

The growth rates are reported in per cent by multiplying each term by 100. As preliminary estimates of annual change in GDP are made well before the end of a calendar year, making assumptions on the growth rate of I_FF allows us to make projections of the annual change in CO₂ emissions well before the end of a calendar year. The I_FF is based on GDP in constant PPP (purchasing power parity) from the International Energy Agency (IEA) up until 2016 (IEA/OECD, 2017) and extended using the International Monetary Fund (IMF) growth rates for 2016 and 2017 (IMF, 2018). Interannual variability in I_FF is the largest source of uncertainty in the GDP-based emission projections. We thus use the standard deviation of the annual I_FF for the period of 2007–2017 as a measure of uncertainty, reflecting a ±1σ as in the rest of the carbon budget. This is ±1.0 % yr⁻¹ for the rest of the world (global emissions minus China, the US, the EU, and India).

The 2018 projection for the world is made of the sum of the projections for China, the US, the EU, India, and the rest of the world. The uncertainty is added in quadrature among the five regions. The uncertainty here reflects the best of our expert opinion.

2.3.1 Bookkeeping models

Land-use change CO₂ emissions and uptake fluxes are calculated by two bookkeeping models. Both are based on the original bookkeeping approach of Houghton (2003) that keeps track of the carbon stored in vegetation and soils before and after a land-use change (transitions between various natural vegetation types, croplands, and pastures). Literature-based response curves describe decay of vegetation and soil carbon, including transfer to product pools of different lifetimes, as well as carbon uptake due to regrowth. In addition, the bookkeeping models represent long-term degradation of primary forest as lowered standing vegetation and soil carbon stocks in secondary forests and also include forest management practices such as wood harvests.

The bookkeeping models do not include land ecosystems' transient response to changes in climate, atmospheric CO₂, and other environmental factors, and the carbon densities are based on contemporary data reflecting environmental conditions at (and up to) that time. Since carbon densities remain fixed over time in bookkeeping models, the additional sink capacity that ecosystems provide in response to CO₂ fertilisation and some other environmental changes is not captured by these models (Pongratz et al., 2014; see Sect. 2.8.4).

The H&N2017 and BLUE models differ in (1) computational units (country level vs. spatially explicit treatment of land-use change), (2) processes represented (see Table A1), and (3) carbon densities assigned to vegetation and soil of each vegetation type. A notable change of H&N2017 over the original approach by Houghton et al. (2003) used in earlier budget estimates is that no shifting cultivation or other back-and-forth transitions below the country level are included. Only a decline in forest area in a country as indicated by the Forest Resource Assessment of the FAO that exceeds the expansion of agricultural area as indicated by the FAO is assumed to represent a concurrent expansion and abandonment of cropland. In contrast, the BLUE model includes sub-grid-scale transitions at the grid level among all vegetation types as indicated by the harmonised land-use change data (LUH2) data set (https://doi.org/10.22033/ESGF/input4MIPs.1127; Hurtt et al., 2011, 2018). Furthermore, H&N2017 assume conversion of natural grasslands to pasture, while BLUE allocates pasture proportionally on all natural vegetation that exists in a grid cell. This is one reason for generally higher emissions in BLUE. H&N2017 add carbon emissions from peat burning based on the Global Fire Emission Database (GFED4s; van der Werf et al., 2017) and peat drainage based on estimates by Hooijer et al. (2010) to the output of their bookkeeping model for the countries of Indonesia and Malaysia. Peat burning and emissions from the organic layers of drained peat soils, which are not captured by bookkeeping methods directly, need to be included to represent the substantially larger emissions and interannual variability due to synergies of land use and climate variability in Southeast Asia, in particular during El Niño events. Similarly to H&N2017, peat burning and drainage-related emissions are also added to the BLUE estimate.

The two bookkeeping estimates used in this study also differ with respect to the land-use change data used to drive the models. H&N2017 base their estimates directly on the Forest Resource Assessment of the FAO, which provides statistics on forest area change and management at intervals of 5 years currently updated until 2015 (FAO, 2015). The data are based on country reporting to the FAO and may include remote-sensing information in more recent assessments. Changes in land use other than forests are based on annual national changes in cropland and pasture areas reported by the FAO (FAOSTAT, 2015). BLUE uses the harmonised land-use change data LUH2 (https://doi.org/10.22033/ESGF/input4MIPs.1127, Hurtt et al., 2011, 2018), which describe land-use change, also based on the FAO data, but downscaled at a quarter-degree spatial resolution, considering sub-grid-scale transitions among primary forest, secondary forest, cropland, pasture, and rangeland. The LUH2 data provide a new distinction between rangelands and pasture. To constrain the models' interpretation on whether rangeland implies the original natural vegetation to be transformed to grassland or not (e.g. browsing on shrubland), a new forest mask was provided with LUH2; forest is assumed to be transformed, while all other natural vegetation remains. This is implemented in BLUE.

The estimate of H&N2017 was extended here by 2 years (to 2017) by adding the anomaly of total tropical emissions (peat drainage from Hooijer et al. (2010), peat burning, and tropical deforestation and degradation fires (from GFED4s) over the previous decade (2006–2015) to the decadal average of the bookkeeping result.

2.3.2 Dynamic global vegetation models (DGVMs)

Land-use change CO₂ emissions have also been estimated using an ensemble of 16 DGVM simulations. The DGVMs account for deforestation and regrowth, the most important components of E_LUC, but they do not represent all processes resulting directly from human activities on land (Table A1). All DGVMs represent processes of vegetation growth and mortality, as well as decomposition of dead organic matter associated with natural cycles, and include the vegetation and soil carbon response to increasing atmospheric CO₂ levels and to climate variability and change. Some models explicitly simulate the coupling of carbon and nitrogen cycles and account for atmospheric N deposition (Table A1). The DGVMs are independent from the other budget terms except for their use of atmospheric CO₂ concentration to calculate the fertilisation effect of CO₂ on plant photosynthesis.

The DGVMs used the HYDE land-use change data set (Klein Goldewijk et al., 2017a, b), which provides annual half-degree fractional data on cropland and pasture. These data are based on annual FAO statistics of change in agricultural land area available until 2012. The FAOSTAT land use database is updated annually, currently covering the period of 1961–2016 (but used here until 2015 because of the timing of data availability). HYDE-applied annual changes in FAO data to the year 2012 from the previous release are used to derive new 2013–2015 data. After the year 2015 HYDE extrapolates cropland, pasture, and urban land use data until the year 2018. Some models also use an update of the more comprehensive harmonised land-use data set (Hurtt et al., 2011), which further includes fractional data on primary and secondary forest vegetation, as well as all underlying transitions between land-use states (Hurtt et al., 2018; Table A1). This new data set is of quarter-degree fractional areas of land use states and all transitions between those states, including a new wood harvest reconstruction, new representation of shifting cultivation, crop rotations, and management information including irrigation and fertiliser application. The land-use states now include five different crop types in addition to the pasture–rangeland split discussed before. Wood harvest patterns are constrained with Landsat tree cover loss data.

DGVMs implement land-use change differently (e.g. an increased cropland fraction in a grid cell can be at the expense of either grassland or shrubs, or forest, the latter resulting in deforestation; land cover fractions of the non-agricultural land differ among models). Similarly, model-specific assumptions are applied to convert deforested biomass or deforested area and other forest product pools into carbon, and different choices are made regarding the allocation of rangelands as natural vegetation or pastures.

The DGVM model runs were forced by either the merged monthly CRU and 6-hourly JRA-55 data set or by the monthly CRU data set, both providing observation-based temperature, precipitation, and incoming surface radiation on a $\mathrm{0.5}{}^{\circ }\times \mathrm{0.5}{}^{\circ }$ grid and updated to 2017 (Harris et al., 2014). The combination of CRU monthly data with 6-hourly forcing is updated this year from NCEP to JRA-55 (Kobayashi et al., 2015), adapting the methodology used in previous years (Viovy, 2016) to the specifics of the JRA-55 data. The forcing data also include global atmospheric CO₂, which changes over time (Dlugokencky and Tans, 2018) and gridded time-dependent N deposition (as used in some models; Table A1).

Two sets of simulations were performed with the DGVMs. Both applied historical changes in climate, atmospheric CO₂ concentration, and N deposition. The two sets of simulations differ, however, with respect to land use: one set applies historical changes in land use, the other a time-invariant pre-industrial land cover distribution and pre-industrial wood harvest rates. By difference of the two simulations, the dynamic evolution of vegetation biomass and soil carbon pools in response to land use change can be quantified in each model (E_LUC). We only retain model outputs with positive E_LUC, i.e. a positive flux to the atmosphere, during the 1990s (Table A1). Using the difference between these two DGVM simulations to diagnose E_LUC means the DGVMs account for the loss of additional sink capacity (around 0.3 GtC yr⁻¹; see Sect. 2.8.4), while the bookkeeping models do not.

2.3.3 Uncertainty assessment for E_LUC

Differences between the bookkeeping models and DGVM models originate from three main sources: the different methodologies, the underlying land use/land cover data set, and the different processes represented (Table A1). We examine the results from the DGVM models and from the bookkeeping method and use the resulting variations as a way to characterise the uncertainty in E_LUC.

Table 5Comparison of results from the bookkeeping method and budget residuals with results from the DGVMs and inverse estimates for different periods, the last decade, and the last year available. All values are in GtC yr⁻¹. The DGVM uncertainties represent ±1σ of the decadal or annual (for 2017 only) estimates from the individual DGVMs: for the inverse models the range of available results is given.

* Estimates are corrected for the pre-industrial influence of river fluxes and adjusted to common E_FF (Sect. 2.8.2). Two inversions are available for the 1980s and 1990s.
Two additional inversions are available from 2001 and used from the decade of the 2000s (Table A3).

Download Print Version | Download XLSX

The E_LUC estimate from the DGVMs multi-model mean is consistent with the average of the emissions from the bookkeeping models (Table 5). However there are large differences among individual DGVMs (standard deviation at around 0.6–0.7 GtC yr⁻¹; Table 5), between the two bookkeeping models (average of 0.7 GtC yr⁻¹), and between the current estimate of H&N2017 and its previous model version (Houghton et al., 2012). The uncertainty in E_LUC of ±0.7 GtC yr⁻¹ reflects our best value judgment that there is at least a 68 % chance (±1σ) that the true land-use change emission lies within the given range, for the range of processes considered here. Prior to the year 1959, the uncertainty in E_LUC was taken from the standard deviation of the DGVMs. We assign low confidence to the annual estimates of E_LUC because of the inconsistencies among estimates and of the difficulties to quantify some of the processes in DGVMs.

2.3.4 Emission projections

We project emissions for both H&N2017 and BLUE for 2018 using the same approach as for the extrapolation of H&N2017 for 2016–2017. Peat burning as well as tropical deforestation and degradation are estimated using active fire data (MCD14ML; Giglio et al., 2016), which scales almost linearly with GFED (van der Werf et al., 2017) and thus allows for tracking fire emissions in deforestation and tropical peat zones in near-real time. During most years, emissions during January–October cover most of the fire season in the Amazon and Southeast Asia, where a large part of the global deforestation takes place.

2.4.1 Global growth rate in atmospheric CO₂ concentration

The rate of growth of the atmospheric CO₂ concentration is provided by the US National Oceanic and Atmospheric Administration Earth System Research Laboratory (NOAA/ESRL, 2018; Dlugokencky and Tans, 2018), which is updated from Ballantyne et al. (2012). For the 1959–1979 period, the global growth rate is based on measurements of atmospheric CO₂ concentration averaged from the Mauna Loa and South Pole stations, as observed by the CO₂ Program at the Scripps Institution of Oceanography (Keeling et al., 1976). For the 1980–2017 time period, the global growth rate is based on the average of multiple stations selected from the marine boundary layer sites with well-mixed background air (Ballantyne et al., 2012), after fitting each station with a smoothed curve as a function of time and averaging by latitude band (Masarie and Tans, 1995). The annual growth rate is estimated by Dlugokencky and Tans (2018) from the atmospheric CO₂ concentration by taking the average of the most recent December–January months corrected for the average seasonal cycle and subtracting this same average 1 year earlier. The growth rate in units of ppm yr⁻¹ is converted to units of GtC yr⁻¹ by multiplying by a factor of 2.124 GtC per ppm (Ballantyne et al., 2012).

The uncertainty around the atmospheric growth rate is due to four main factors. The first factor is the long-term reproducibility of reference gas standards (around 0.03 ppm for 1σ from the 1980s). The second factor is that small unexplained systematic analytical errors that may have a duration of several months to 2 years come and go. They have been simulated by randomising both the duration and the magnitude (determined from the existing evidence) in a Monte Carlo procedure. The third factor is the network composition of the marine boundary layer with some sites coming or going, gaps in the time series at each site, etc. (Dlugokencky and Tans, 2018). The latter uncertainty was estimated by NOAA/ESRL with a Monte Carlo method by constructing 100 “alternative” networks (NOAA/ESRL, 2018; Masarie and Tans, 1995). The second and third uncertainties, summed in quadrature, add up to 0.085 ppm on average (Dlugokencky and Tans, 2018). Fourth, the uncertainty associated with using the average CO₂ concentration from a surface network to approximate the true atmospheric average CO₂ concentration (mass weighted, in three dimensions) as needed to assess the total atmospheric CO₂ burden. In reality, CO₂ variations measured at the stations will not exactly track changes in total atmospheric burden, with offsets in magnitude and phasing due to vertical and horizontal mixing. This effect must be very small on decadal and longer timescales, when the atmosphere can be considered well mixed. Preliminary estimates suggest this effect would increase the annual uncertainty, but a full analysis is not yet available. We therefore maintain an uncertainty around the annual growth rate based on the multiple stations' data set ranges between 0.11 and 0.72 GtC yr⁻¹, with a mean of 0.61 GtC yr⁻¹ for 1959–1979 and 0.18 GtC yr⁻¹ for 1980–2017, when a larger set of stations were available as provided by Dlugokencky and Tans (2018), but recognise further exploration of this uncertainty is required. At this time, we estimate the uncertainty of the decadal averaged growth rate after 1980 at 0.02 GtC yr⁻¹ based on the calibration and the annual growth rate uncertainty, but stretched over a 10-year interval. For years prior to 1980, we estimate the decadal averaged uncertainty to be 0.07 GtC yr⁻¹ based on a factor proportional to the annual uncertainty prior to and after 1980 ($\mathrm{0.61}/\mathrm{0.18}\times \mathrm{0.02}$ GtC yr⁻¹).

We assign a high confidence to the annual estimates of G_ATM because they are based on direct measurements from multiple and consistent instruments and stations distributed around the world (Ballantyne et al., 2012).

In order to estimate the total carbon accumulated in the atmosphere since 1750 or 1870, we use an atmospheric CO₂ concentration of 277±3 ppm or 288±3 ppm, respectively, based on a cubic spline fit to ice core data (Joos and Spahni, 2008). The uncertainty of ±3 ppm (converted to ±1σ) is taken directly from the IPCC's assessment (Ciais et al., 2013). Typical uncertainties in the growth rate in atmospheric CO₂ concentration from ice core data are equivalent to ±0.1–0.15 GtC yr⁻¹ as evaluated from the Law Dome data (Etheridge et al., 1996) for individual 20-year intervals over the period from 1870 to 1960 (Bruno and Joos, 1997).

2.4.2 Atmospheric growth rate projection

We provide an assessment of G_ATM for 2018 based on the observed increase in atmospheric CO₂ concentration at the Mauna Loa station for January to October and a mean growth rate over the past 5 years for the months November to December. Growth at Mauna Loa is closely correlated with the global growth (r=0.95) and is used here as a proxy for global growth, but the regression is not 1 to 1. We also adjust the projected global growth rate to take this into account. The assessment method used this year differs from the forecast method used in Le Quéré et al. (2018) based on the relationship between annual CO₂ growth rate and sea surface temperatures (SSTs) in the Niño3.4 region of Betts et al. (2016). A change was introduced because although the observed growth rate for 2017 of 2.2 ppm was within the projection range of 2.5±0.5 ppm of last year ( Le Quéré et al., 2018), the forecast values for 2018 for January to October are too high by approximately 0.4 ppm above observed values on average. The reasons for the difference are being investigated. The use of observed growth at Mauna Loa Observatory, Hawaii, for the first half of the year is thought to be more robust because of its high correlation with the global growth rate. Furthermore, additional analysis suggests that the first half of the year shows more interannual variability than the second half of the year, so that the exact projection method applied to November–December has only a small impact (<0.1 ppm) on the projection of the full year. Uncertainty is estimated from past variability using the standard deviation of the last 5 years' monthly growth rates.

2.5.1 Observation-based estimates

We use the observational constraints assessed by IPCC of a mean ocean CO₂ sink of 2.2±0.4 GtC yr⁻¹ for the 1990s (Denman et al., 2007) to verify that the GOBMs provide a realistic assessment of S_OCEAN. This is based on indirect observations with seven different methodologies and their uncertainties, using the methods that are deemed most reliable for the assessment of this quantity (Denman et al., 2007). The IPCC confirmed this assessment in 2013 (Ciais et al., 2013). The observational-based estimates use the ocean–land CO₂ sink partitioning from observed atmospheric O₂∕N₂ concentration trends (Manning and Keeling, 2006; updated in Keeling and Manning 2014), an oceanic inversion method constrained by ocean biogeochemistry data (Mikaloff Fletcher et al., 2006), and a method based on a penetration timescale for chlorofluorocarbons (McNeil et al., 2003). The IPCC estimate of 2.2 GtC yr⁻¹ for the 1990s is consistent with a range of methods (Wanninkhof et al., 2013).

We also use two estimates of the ocean CO₂ sink and its variability based on interpolations of measurements of surface ocean fugacity of CO₂ (pCO₂ corrected for the non-ideal behaviour of the gas; Pfeil et al., 2013). We refer to these as pCO₂-based flux estimates. The measurements are from the Surface Ocean CO₂ Atlas version 6, which is an update of version 3 (Bakker et al., 2016) and contains quality-controlled data until 2017 (see data attribution Table A4). The SOCAT v6 data were mapped using a data-driven diagnostic method (Rödenbeck et al., 2013) and a combined self-organising map and feed-forward neural network (Landschützer et al., 2014). The global pCO₂-based flux estimates were adjusted to remove the pre-industrial ocean source of CO₂ to the atmosphere of 0.78 GtC yr⁻¹ from river input to the ocean (Resplandy et al., 2018), per our definition of S_OCEAN. Several other ocean sink products based on observations are also available but they continue to show large unresolved discrepancies with observed variability. Here we used the two pCO₂-based flux products that had the best fit to observations for their representation of tropical and global variability (Rödenbeck et al., 2015).

We further use results from two diagnostic ocean models of Khatiwala et al. (2013) and DeVries (2014) to estimate the anthropogenic carbon accumulated in the ocean prior to 1959. The two approaches assume constant ocean circulation and biological fluxes, with S_OCEAN estimated as a response in the change in atmospheric CO₂ concentration calibrated to observations. The uncertainty in cumulative uptake of ±20 GtC (converted to ±1σ) is taken directly from the IPCC's review of the literature (Rhein et al., 2013), or about ±30 % for the annual values (Khatiwala et al., 2009).

2.5.2 Global ocean biogeochemistry models (GOBMs)

The ocean CO₂ sink for 1959–2017 is estimated using seven GOBMs (Table A2). The GOBMs represent the physical, chemical, and biological processes that influence the surface ocean concentration of CO₂ and thus the air–sea CO₂ flux. The GOBMs are forced by meteorological reanalysis and atmospheric CO₂ concentration data available for the entire time period. They mostly differ in the source of the atmospheric forcing data (meteorological reanalysis), spin-up strategies, and their horizontal and vertical resolutions (Table A2). GOBMs do not include the effects of anthropogenic changes in nutrient supply, which could lead to an increase in the ocean sink of up to about 0.3 GtC yr⁻¹ over the industrial period (Duce et al., 2008). They also do not include the perturbation associated with changes in riverine organic carbon (see Sect. 2.8.3).

2.5.3 GOBM evaluation and uncertainty assessment for S_OCEAN

The mean ocean CO₂ sink for all GOBMs falls within 90 % confidence of the observed range, or 1.6 to 2.8 GtC yr⁻¹ for the 1990s. Here we have adjusted the confidence interval to the IPCC confidence interval of 90 % to avoid rejecting models that may be outliers but are still plausible.

The GOBMs and flux products have been further evaluated using fCO₂ from the SOCAT v6 database. We focused this initial evaluation on the interannual mismatch metric proposed by Rödenbeck et al. (2015) for the comparison of flux products. The metric provides a measure of the mismatch between observations and models or flux products on the x axis as well as a measure of the amplitude of the interannual variability on the y axis. A smaller number on the x axis indicates a better fit with observations. The amplitude of the interannual variability in S_OCEAN (y axis) is calculated as the temporal standard deviation of the CO₂ flux time series.

The calculation for the x axis is carried out as follows: (1) the mismatch between the observed and the modelled fCO₂ is calculated for the period 1985 to 2017 (except for the IPSL model, which uses 1985 to 2015 due to data availability), but only for grid points for which actual observations exist. (2) The interannual variability in this mismatch is calculated as the temporal standard deviation of the mismatch. (3) To put numbers into perspective, the interannual variability in the mismatch is reported relative to the interannual variability in the mismatch between a benchmark fCO₂ field and the observations. The benchmark fCO₂ field is designed to have no interannual variability, i.e. it is calculated as the mean seasonal cycle at each grid point over the full period plus the deseasonalised atmospheric fCO₂ increase over time. By definition, the interannual variability in the misfit between benchmark and observations is large as the benchmark field does not contain any interannual variability from the ocean. A smaller relative interannual variability mismatch indicates a better fit between observed and modelled fCO₂. This metric is chosen because it is the most direct measure of the year-to-year variability in S_OCEAN in ocean biogeochemistry models. We apply the metric globally and by latitude bands. Results are shown in Fig. B1 and discussed in Sect. 3.1.3.

The uncertainty around the mean ocean sink of anthropogenic CO₂ was quantified by Denman et al. (2007) for the 1990s (see Sect. 2.5.1). To quantify the uncertainty around annual values, we examine the standard deviation of the GOBM ensemble, which averages between 0.2 and 0.3 GtC yr⁻¹ during 1959–2017. We estimate that the uncertainty in the annual ocean CO₂ sink is about ±0.5 GtC yr⁻¹ from the combined uncertainty of the mean flux based on observations of ±0.4 GtC yr⁻¹ and the standard deviation across GOBMs of up to ±0.3 GtC yr⁻¹, reflecting the uncertainty in both the mean sink from observations during the 1990s (Denman et al., 2007; Sect. 2.5.1) and the interannual variability as assessed by GOBMs.

We examine the consistency between the variability in the model-based and the pCO₂-based flux products to assess confidence in S_OCEAN. The interannual variability in the ocean fluxes (quantified as the standard deviation) of the two pCO₂-based flux products for 1985–2017 (where they overlap) is ±0.36 GtC yr⁻¹ (Rödenbeck et al., 2014) and ±0.38 GtC yr⁻¹ (Landschützer et al., 2015), compared to ±0.29 GtC yr⁻¹ for the GOBM ensemble. The standard deviation includes a component of trend and decadal variability in addition to interannual variability, and their relative influence differs across estimates. Individual estimates (both GOBM and flux products) generally produce a higher ocean CO₂ sink during strong El Niño events. The annual pCO₂-based flux products correlate with the ocean CO₂ sink estimated here with a correlation of r=0.75 (0.59 to 0.79 for individual GOBMs) and r=0.80 (0.71 to 0.81) for the pCO₂-based flux products of Rödenbeck et al. (2014) and Landschützer et al. (2015), respectively (simple linear regression), with their mutual correlation at 0.73. The agreement between models and the flux products reflects some consistency in their representation of underlying variability since there is little overlap in their methodology or use of observations. The use of annual data for the correlation may reduce the strength of the relationship because the dominant source of variability associated with El Niño events is less than 1 year. We assess a medium confidence level to the annual ocean CO₂ sink and its uncertainty because it is based on multiple lines of evidence, and the results are consistent in that the interannual variability in the GOBMs and data-based estimates are all generally small compared to the variability in the growth rate of atmospheric CO₂ concentration.

2.6.1 DGVM simulations

The terrestrial land sink (S_LAND) is thought to be due to the combined effects of fertilisation by rising atmospheric CO₂ and N deposition on plant growth, as well as the effects of climate change such as the lengthening of the growing season in northern temperate and boreal areas. S_LAND does not include land sinks directly resulting from land use and land-use change (e.g. regrowth of vegetation) as these are part of the land use flux (E_LUC), although system boundaries make it difficult to exactly attribute CO₂ fluxes on land between S_LAND and E_LUC (Erb et al., 2013).

S_LAND is estimated from the multi-model mean of the DGVMs (Table 4). As described in Sect. 2.3.2, DGVM simulations include all climate variability and CO₂ effects over land, with some DGVMs also including the effect of N deposition. The DGVMs do not include the perturbation associated with changes in river organic carbon, which is discussed in Sect. 2.8.

2.6.2 DGVM evaluation and uncertainty assessment for S_LAND

We apply three criteria for minimum DGVM realism by including only those DGVMs with (1) steady state after spin-up; (2) net land fluxes (S_LAND–E_LUC) that are the atmosphere-to-land carbon flux over the 1990s ranging between −0.3 and 2.3 GtC yr⁻¹, within 90 % confidence of constraints by global atmospheric and oceanic observations (Keeling and Manning, 2014; Wanninkhof et al., 2013); and (3) global E_LUC that is a carbon source to the atmosphere over the 1990s. All 16 DGVMs meet the three criteria.

In addition, the DGVM results are now also evaluated using the International Land Model Benchmarking System (ILAMB; Collier et al., 2018). This evaluation is provided here to document, encourage, and support model improvements through time. ILAMB variables cover key processes that are relevant for the quantification of S_LAND and resulting aggregated outcomes. The selected variables are vegetation biomass, gross primary productivity, leaf area index, net ecosystem exchange, ecosystem respiration, evapotranspiration, and runoff (see Fig. B2 for the results and for the list of observed databases). Results are shown in Fig. B2 and discussed in Sect. 3.1.3.

For the uncertainty, we use the standard deviation of the annual CO₂ sink across the DGVMs, which averages to ±0.8 GtC yr⁻¹ for the period from 1959 to 2017. We attach a medium confidence level to the annual land CO₂ sink and its uncertainty because the estimates from the residual budget and averaged DGVMs match well within their respective uncertainties (Table 5).

2.7.1 Atmospheric inversions

The four inversion systems used in this release are CarbonTracker Europe (CTE; van der Laan-Luijkx et al., 2017), Jena CarboScope (Rödenbeck, 2005), the Copernicus Atmosphere Monitoring Service (CAMS; Chevallier et al., 2005), and MIROC (Patra et al., 2018). See Table A3 for version numbers. The inversions are based on the same Bayesian inversion principles that interpret the same, for the most part, observed time series (or subsets thereof) but use different methodologies (Table A3). These differences mainly concern the selection of atmospheric CO₂ data, the used prior fluxes, spatial breakdown (i.e. grid size), assumed correlation structures, and mathematical approach. The details of these approaches are documented extensively in the references provided above. Each system uses a different transport model, which was demonstrated to be a driving factor behind differences in atmospheric-based flux estimates, and specifically their distribution across latitudinal bands (e.g. Gaubert et al., 2018).

The inversions use atmospheric CO₂ observations from various flask and in situ networks, as detailed in Table A3. They prescribe global E_FF, which is scaled to the present study for CAMS and CTE, while slightly lower E_FF values based on alternative emission compilations were used in CarboScope and MIROC. Since this is known to result directly in lower total CO₂ uptake in atmospheric inversions (Gaubert et al., 2018; Peylin et al., 2013), we adjusted the land sink of each inversion estimate (where most of the emissions occur) by its fossil fuel difference to the CAMS model. These differences amount to as much as 0.7 GtC for certain years (CarboScope inversion region NH) and are thus an important consideration in an inverse flux comparison.

The land–ocean CO₂ fluxes from atmospheric inversions contain anthropogenic perturbation and natural pre-industrial CO₂ fluxes. Natural pre-industrial fluxes are land CO₂ sinks corresponding to carbon transported to the ocean by rivers. These land CO₂ sinks are compensated for over the globe by ocean CO₂ sources corresponding to the outgassing of riverine carbon inputs to the ocean. We apply the distribution of land CO₂ fluxes in three latitude bands using estimates from Resplandy et al. (2018), which are constrained by ocean heat transport to a total sink of 0.78 GtC yr⁻¹. The latitude distribution of river-induced ocean CO₂ sources is derived from a simulation of the IPSL GOBM using the river flux constrained by heat transport of Resplandy et al. (2018) as an input. We adjusted the land–ocean fluxes per latitude band based on these results.

The atmospheric inversions are now evaluated using vertical profiles of atmospheric CO₂ concentrations (Fig. B3). More than 50 aircraft programmes over the globe, either regular or occasional, have been used in order to draw a picture of the model performance but the space–time data coverage is irregular, denser around 2009 or in the 0–45^∘ N latitude band. The four models are compared to independent CO₂ measurements made onboard aircraft over many places of the world between 1 and 7 km above sea level, between 2008 and 2016. Results are shown in Fig. B3 and discussed in Sect. 3.1.3.

2.8.1 Contribution of anthropogenic CO and CH₄ to the global carbon budget

Equation (1) only partly includes the net input of CO₂ to the atmosphere from the chemical oxidation of reactive carbon-containing gases from sources other than the combustion of fossil fuels, such as (1) cement process emissions, since these do not come from combustion of fossil fuels, (2) the oxidation of fossil fuels, and (3) the assumption of immediate oxidation of vented methane in oil production. Equation (1) omits however any other anthropogenic carbon-containing gases that are eventually oxidised in the atmosphere, such as anthropogenic emissions of CO and CH₄. An attempt is made in this section to estimate their magnitude and identify the sources of uncertainty. Anthropogenic CO emissions are from incomplete fossil fuel and biofuel burning and deforestation fires. The main anthropogenic emissions of fossil CH₄ that matter for the global carbon budget are the fugitive emissions of coal, oil, and gas upstream from sectors (see below). These emissions of CO and CH₄ contribute a net addition of fossil carbon to the atmosphere. In our estimate of E_FF we assumed (Sect. 2.1.1) that all the fuel burned is emitted as CO₂; thus CO anthropogenic emissions associated with incomplete combustion and their atmospheric oxidation into CO₂ within a few months are already counted implicitly in E_FF and should not be counted twice (same for E_LUC and anthropogenic CO emissions by deforestation fires). Anthropogenic emissions of fossil CH₄ are not included in E_FF because these fugitive emissions are not included in the fuel inventories. Yet they contribute to the annual CO₂ growth rate after CH₄ is oxidised into CO₂. Anthropogenic emissions of fossil CH₄ represent 15 % of total CH₄ emissions (Kirschke et al., 2013), that is 0.061 GtC yr⁻¹ for the past decade. Assuming steady state, these emissions are all converted to CO₂ by OH oxidation and thus explain 0.06 GtC yr⁻¹ of the global CO₂ growth rate in the past decade, or 0.07–0.1 GtC yr⁻¹ using the higher CH₄ emissions reported recently (Schwietzke et al., 2016).

Figure 2Schematic representation of the overall perturbation of the global carbon cycle caused by anthropogenic activities, averaged globally for the decade 2008–2017. See legends for the corresponding arrows and units. The uncertainty in the atmospheric CO₂ growth rate is very small (±0.02 GtC yr⁻¹) and is neglected for the figure. The anthropogenic perturbation occurs on top of an active carbon cycle, with fluxes and stocks represented in the background and taken from Ciais et al. (2013) for all numbers, with the ocean fluxes updated to 90 GtC yr⁻¹ to account for the increase in atmospheric CO₂ since publication, and except for the carbon stocks at the coasts, which are from a literature review of coastal marine sediments (Price and Warren, 2016).

Download

Other anthropogenic changes in the sources of CO and CH₄ from wildfires, vegetation biomass, wetlands, ruminants, or permafrost changes are similarly assumed to have a small effect on the CO₂ growth rate. The CH₄ emissions and sinks are published and analysed separately in the Global Methane Budget publication that follows an approach similar to that presented here (Saunois et al., 2016).

2.8.2 Contribution of other carbonates to CO₂ emissions

The contribution of fossil carbonates other than cement production is not systematically included in estimates of E_FF, except at the national level at which they are accounted for in the UNFCCC national inventories. The missing processes include CO₂ emissions associated with the calcination of lime and limestone outside cement production and the reabsorption of CO₂ by the rocks and concrete from carbonation through their lifetime (Xi et al., 2016). Carbonates are used in various industries, including in iron and steel manufacture and in agriculture. They are found naturally in some coals. Carbonation from the cement life cycle, including demolition and crushing, was estimated by one study to be around 0.25 GtC yr⁻¹ for the year 2013 (Xi et al., 2016). Carbonation emissions from the cement life cycle would offset calcination emissions from lime and limestone production. The balance of these two processes is not clear.

2.8.3 Anthropogenic carbon fluxes in the land-to-ocean aquatic continuum

The approach used to determine the global carbon budget refers to the mean, variations, and trends in the perturbation of CO₂ in the atmosphere, referenced to the pre-industrial era. Carbon is continuously displaced from the land to the ocean through the land–ocean aquatic continuum (LOAC) comprising freshwaters, estuaries, and coastal areas (Bauer et al., 2013; Regnier et al., 2013). A significant fraction of this lateral carbon flux is entirely “natural” and is thus a steady-state component of the pre-industrial carbon cycle. We account for this pre-industrial flux where appropriate in our study. However, changes in environmental conditions and land use change have caused an increase in the lateral transport of carbon into the LOAC – a perturbation that is relevant for the global carbon budget presented here.

The results of the analysis of Regnier et al. (2013) can be summarised in two points of relevance for the anthropogenic CO₂ budget. First, the anthropogenic perturbation has increased the organic carbon export from terrestrial ecosystems to the hydrosphere at a rate of 1.0±0.5 GtC yr⁻¹, mainly owing to enhanced carbon export from soils. Second, this exported anthropogenic carbon is partly respired through the LOAC, partly sequestered in sediments along the LOAC, and to a lesser extent transferred to the open ocean where it may accumulate. The increase in storage of land-derived organic carbon in the LOAC and open ocean combined is estimated by Regnier et al. (2013) at 0.65±0.35 GtC yr⁻¹. We do not attempt to incorporate the changes in LOAC in our study.

The inclusion of freshwater fluxes of anthropogenic CO₂ affects the estimates of, and partitioning between, S_LAND and S_OCEAN in Eq. (1), but does not affect the other terms. This effect is not included in the GOBMs and DGVMs used in our global carbon budget analysis presented here.

2.8.4 Loss of additional sink capacity

Historical land-cover change was dominated by transitions from vegetation types that can provide a large sink per area unit (typically forests) to others less efficient in removing CO₂ from the atmosphere (typically croplands). The resultant decrease in land sink, called the “loss of sink capacity”, is calculated as the difference between the actual land sink under changing land cover and the counterfactual land sink under pre-industrial land cover. An efficient protocol has yet to be designed to estimate the magnitude of the loss of additional sink capacity in DGVMs. Here, we provide a quantitative estimate of this term to be used in the discussion. Our estimate uses the compact Earth system model OSCAR whose land carbon cycle component is designed to emulate the behaviour of DGVMs (Gasser et al., 2017). We use OSCAR v2.2.1 (an update of v2.2 with minor changes) in a probabilistic setup identical to the one of Arneth et al. (2017) but with a Monte Carlo ensemble of 2000 simulations. For each, we calculate S_LAND and the loss of additional sink capacity separately. We then constrain the ensemble by weighting each member to obtain a distribution of cumulative S_LAND over 1850–2005 close to the DGVMs used here. From this ensemble, we estimate a loss of additional sink capacity of 0.4±0.3 GtC yr⁻¹ on average over 2005–2014 and 20±15 GtC accumulated between 1870 and 2017 (using a linear extrapolation of the trend to estimate the last few years).

3.1.1 CO₂ emissions

Global fossil CO₂ emissions have increased every decade from an average of 3.1±0.2 GtC yr⁻¹ in the 1960s to an average of 9.4±0.5 GtC yr⁻¹ during 2008–2017 (Table 6, Figs. 2 and 5). The growth rate in these emissions decreased between the 1960s and the 1990s, from 4.5 % yr⁻¹ in the 1960s (1960–1969) to 2.8 % yr⁻¹ in the 1970s (1970–1979), 1.9 % yr⁻¹ in the 1980s (1980–1989), and 1.0 % yr⁻¹ in the 1990s (1990–1999). After this period, the growth rate began increasing again in the 2000s at an average growth rate of 3.2 % yr⁻¹, decreasing to 1.5 % yr⁻¹ for the last decade (2008–2017), with a 3-year period of no or low growth during 2014–2016 (Fig. 5).

Figure 5Fossil CO₂ emissions for (a) the globe, including an uncertainty of ±5 % (grey shading), and the emissions extrapolated using BP energy statistics (black dots); (b) global emissions by fuel type, including coal (salmon), oil (olive), gas (turquoise), and cement (purple), and excluding gas flaring, which is small (0.6 % in 2013); (c) territorial (solid lines) and consumption (dashed lines) emissions for the top three country emitters (US – olive; China – salmon; India – purple) and for the European Union (EU; turquoise for the 28 member states of the EU as of 2012); and (d) per capita emissions for the top three country emitters, the EU (all colours as in panel c), and the world (black). In (b–c), the dots show the data that were extrapolated from BP energy statistics for 2014–2016. All time series are in GtC yr⁻¹ except the per capita emissions (d), which are in tonnes of carbon per person per year (tC person⁻¹ yr⁻¹). Territorial emissions are primarily from Boden et al. (2017) except national data for the US and EU28 (the 28 member states of the EU) for 1990–2016, which are reported by the countries to the UNFCCC as detailed in the text; consumption-based emissions are updated from Peters et al. (2011a). See Sect. 2.1.1 for details of the calculations and data sources.

Download

In contrast, CO₂ emissions from land use, land-use change, and forestry have remained relatively constant, at around 1.3±0.7 GtC yr⁻¹ over the past half-century but with large spread across estimates (Fig. 6). These emissions are also relatively constant in the DGVM ensemble of models, except during the last decade when they increase to 1.9±0.6 GtC yr⁻¹. However, there is no agreement on this recent increase between the two bookkeeping models, each suggesting an opposite trend (Fig. 6).

Figure 6CO₂ exchanges between the atmosphere and the terrestrial biosphere as used in the global carbon budget (black with ±1σ uncertainty in grey shading), for (a) CO₂ emissions from land-use change (E_LUC), also individually showing the two bookkeeping models (two brown lines) and the DGVM model results (green) and their multi-model mean (dark green). The dashed line identifies the pre-satellite period before the inclusion of peatland burning. (b) Land CO₂ sink (S_LAND) with individual DGVMs (green); (c) total land CO₂ fluxes (b–a) with individual DGVMs (green) and their multi-model mean (dark green).

Download

3.1.2 Partitioning among the atmosphere, ocean, and land

The growth rate in atmospheric CO₂ level increased from 1.7±0.07 GtC yr⁻¹ in the 1960s to 4.7±0.02 GtC yr⁻¹ during 2008–2017 with important decadal variations (Table 6 and Fig. 2). Both ocean and land CO₂ sinks increased roughly in line with the atmospheric increase, but with significant decadal variability on land (Table 6), and possibly in the ocean (Fig. 7).

The ocean CO₂ sink increased from 1.0±0.5 GtC yr⁻¹ in the 1960s to 2.4±0.5 GtC yr⁻¹ during 2008–2017, with interannual variations of the order of a few tenths of GtC yr⁻¹ generally showing an increased ocean sink during large El Niño events (i.e. 1997–1998) (Fig. 7; Rödenbeck et al., 2014). Although there is some coherence among the GOBMs and pCO₂-based flux products regarding the mean, there is poor agreement for interannual variability, and the ocean models underestimate decadal variability (Sect. 2.5.3 and Fig. 7; DeVries et al., 2017).

Figure 7Comparison of the anthropogenic atmosphere–ocean CO₂ flux showing the budget values of S_OCEAN (black; with ±1σ uncertainty in grey shading), individual ocean models (blue), and the two ocean pCO₂-based flux products (dark blue; see Table 4). Both pCO₂-based flux products were adjusted for the pre-industrial ocean source of CO₂ from river input to the ocean, which is not present in the ocean models, by adding a sink of 0.78 GtC yr⁻¹ (Resplandy et al., 2018), to make them comparable to S_OCEAN. This adjustment does not take into account the anthropogenic contribution to river fluxes (see Sect. 2.8.3).

Download

The terrestrial CO₂ sink increased from 1.2±0.5 GtC yr⁻¹ in the 1960s to 3.2±0.7 GtC yr⁻¹ during 2008–2017, with important interannual variations of up to 2 GtC yr⁻¹ generally showing a decreased land sink during El Niño events (Fig. 6), responsible for the corresponding enhanced growth rate in atmospheric CO₂ concentration. The larger land CO₂ sink during 2008–2017 compared to the 1960s is reproduced by all the DGVMs in response to the combined atmospheric CO₂ increase and changes in climate, and consistent with constraints from the other budget terms (Table 5).

Estimates of total atmosphere-to-land fluxes (S_LAND–E_LUC) from the DGVMs are consistent with the budget constraints (Table 5), except during 2008–2017, when the DGVM ensemble estimates a total atmosphere-to-land flux of 1.3±0.5 GtC yr⁻¹, likely below the budget constraints of 2.1±0.7 GtC yr⁻¹ and outside the range of the inversions (Table 5). This comparison suggests that the DGVMs could overestimate E_LUC emissions and/or underestimate the terrestrial sink S_LAND during the last decade.

3.1.3 Model evaluation

The evaluation of ocean estimates (Fig. B1) shows a relative interannual mismatch of 15 % and 17 % for the two pCO₂-based flux products over the globe, relative to the pCO₂ observations from the SOCAT v6 database for the period 1985–2017. A 0 % mismatch would indicate a perfect model, and a field with no interannual variability would result in a 100 % mismatch. A mismatch larger than 100 % is possible when the method produces a larger mismatch than the benchmark field with no interannual variability (see Sect. 2.5.3). This mismatch by the pCO₂-based flux products is improved compared with earlier published versions of these two flux products of around 20 %–25 % for the 1992–2009 time period (Rödenbeck et al., 2015), likely because of the larger data availability after 2009. The GOBMs show a global relative interannual mismatch between 50 % and 60 %, with one model at 94 % and one at 193 %. The GOBM mismatch is of the same order as the mismatch calculated in an ensemble of 14 flux products but larger than the two flux products used in this report (Fig. 5 in Rödenbeck et al., 2015). The mismatch is generally larger at high latitudes compared to the tropics, for both the flux products and the GOBMs. The two flux products have a similar mismatch of around 10 %–15 % in the tropics, around 25 % in the north, and 30 %–55 % in the south. The GOBM mismatch is more spread across regions, ranging from 29 % to 178 % in the tropics, 70 % to 192 % in the north, and 108 % to 304 % in the south. The higher mismatch occurs in regions with stronger climate variability, such as the northern and southern high latitudes (poleward of the subtropical gyres) and the equatorial Pacific. The latter is also apparent in the model mismatch but is hidden in Fig. B1 due to the averaging over 30^∘ S to 30^∘ N (see also Sect. 4).

The evaluation of the DGVMs (Fig. B2) shows generally high skill scores across models for runoff, and to a lesser extent for vegetation biomass, GPP, and ecosystem respiration (Fig. B2, left panel). The skill score was lowest for leaf area index and net ecosystem exchange, with the widest disparity among models for soil carbon. Further analysis of the results will be provided separately, focusing on the strengths and weaknesses in the DGVM ensemble and its validity for use in the global carbon budget.

The evaluation of the atmospheric inversions (Fig. B3) shows long-term mean biases in the free troposphere better than 0.8 ppm in absolute values for each product. CAMS and CTE biases show some dependency on latitude (a trend of $-\mathrm{0.0018}\pm \mathrm{0.0005}$ and 0.0043±0.0004 ppm per degree for CAMS and CTE, respectively). These latitude-dependent biases may reveal biases in the surface fluxes (e.g. Houweling et al., 2015) but the link is not straightforward and will be analysed separately. The biases for MIROC and CarboScope behave similarly together in relative values, but they are less regular than the two other products, which hampers the interpretation. Lesser model performance for specific aircraft programmes, like for the 4-year DISCOVER-AQ campaign in the continental US (https://discover-aq.larc.nasa.gov/, last access: 28 November 2018), contributes to this variability.

3.1.4 Budget imbalance

The carbon budget imbalance (B_IM; Eq. 1) quantifies the mismatch between the estimated total emissions and the estimated changes in the atmosphere, land, and ocean reservoirs. The mean budget imbalance from 1959 to 2017 is small (0.14 GtC yr⁻¹) and shows no trend over the full time series. The process models (GOBMs and DGVMs) have been selected to match observational constraints in the 1990s but no further constraints have been applied to their representation of trend and variability. Therefore, the near-zero mean and trend in the budget imbalance are indirect evidence of a coherent community understanding of the emissions and their partitioning on those timescales (Fig. 4). However, the budget imbalance shows substantial variability on the order of ±1 GtC yr⁻¹, particularly over semi-decadal timescales, although most of the variability is within the uncertainty of the estimates. The positive carbon imbalance during the 1960s, early 1990s, and in the last decade suggests that either the emissions were overestimated or the sinks were underestimated during these periods. The reverse is true for the 1970s and around 1995–2000 (Fig. 4).

We cannot attribute the cause of the variability in the budget imbalance with our analysis, only to note that the budget imbalance is unlikely to be explained by errors or biases in the emissions alone because of its large semi-decadal variability component, a variability that is untypical of emissions and has not changed in the past 50 years in spite of a nearly tripling in emissions (Fig. 4). Errors in S_LAND and S_OCEAN are more likely to be the main cause for the budget imbalance. For example, underestimation of S_LAND by DGVMs has been reported following the eruption of Mount Pinatubo in 1991 possibly due to missing responses to changes in diffuse radiation (Mercado et al., 2009) or other yet unknown factors, and DGVMs are suspected to overestimate the land sink in response to the wet decade of the 1970s (Sitch et al., 2008). Decadal and semi-decadal variability in the ocean sink has been also reported recently (DeVries et al., 2017; Landschützer et al., 2015), with the pCO₂-based ocean flux products suggesting a smaller-than-expected ocean CO₂ sink in the 1990s and a larger-than-expected sink in the 2000s (Fig. 7), possibly caused by changes in ocean circulation (DeVries et al., 2017) not captured in coarse-resolution GOBMs used here (Dufour et al., 2013). The absence of internal variability could also be at fault. Internal variability is not captured by single realisations of coarse-resolution model simulations (Li and Ilyina, 2018) and is thought to be largest in regions with strong seasonal and interannual climate variability, i.e. the high-latitude ocean regions (poleward of the subtropical gyres) and the equatorial Pacific (McKinley et al., 2016). Some of these errors could be driven by errors in the climatic forcing data, particularly precipitation (for S_LAND) and wind (for S_OCEAN), rather than in the models.

3.2.1 CO₂ emissions

Global fossil CO₂ emissions grew at a rate of 1.5 % yr⁻¹ for the last decade (2008–2017). China's emissions increased by +3.0 % yr⁻¹ on average (increasing by +0.64 GtC yr⁻¹ during the 10-year period), dominating the global trends, followed by India's emissions increase by +5.2 % yr⁻¹ (increasing by +0.25 GtC yr⁻¹), while emissions decreased in the EU28 by −1.8 % yr⁻¹ (decreasing by −0.17 GtC yr⁻¹), and in the US by 0.9 % yr⁻¹ (decreasing by −0.18 GtC yr⁻¹). In the past decade, fossil CO₂ emissions decreased significantly (at the 95 % level) in 25 countries: Aruba, Barbados, Croatia, Czech Republic, North Korea, Denmark, France, Greece, Greenland, Iceland, Ireland, Malta, the Netherlands, Romania, Slovakia, Slovenia, Sweden, Switzerland, Syria, Trinidad and Tobago, Ukraine, the United Kingdom, the US, Uzbekistan, and Venezuela. Notable was Germany, whose emissions did not decrease significantly.

In contrast, there is no apparent trend in CO₂ emissions from land-use change (Fig. 6), though the data are very uncertain, with the two bookkeeping estimates showing opposite trends over the last decade. Larger emissions are expected increasingly over time for DGVM-based estimates as they include the loss of additional sink capacity, while the bookkeeping estimates do not. The LUH2 data set also features large dynamics in land use in particular in the tropics in recent years, causing higher emissions in DGVMs and BLUE than in H&N.

3.2.2 Partitioning among the atmosphere, ocean, and land

The growth rate in atmospheric CO₂ concentration increased during 2008–2017, in contrast to more constant levels the previous decade and reflecting a similar decrease in the land sink compared to an increase in the previous decade, albeit with large interannual variability (Fig. 4). During the same period, the ocean CO₂ sink appears to have intensified, an effect which is particularly apparent in the pCO₂-based flux products (Fig. 7) and is thought to originate at least in part in the Southern Ocean (Landschützer et al., 2015).

The budget imbalance (Table 6) and the residual sink from the global budget (Table 5) include an error term due to the inconsistency that arises from using E_LUC from bookkeeping models but S_LAND from DGVMs. This error term includes the fundamental differences between bookkeeping models and DGVMs, most notably the loss of additional sink capacity. Other differences include an incomplete account of land-use change practices and processes in DGVMs, while they are all accounted for in bookkeeping models by using observed carbon densities, and bookkeeping error of keeping present-day carbon densities fixed in the past. That the budget imbalance shows no clear trend towards larger values over time is an indication that the loss of additional sink capacity plays a minor role compared to other errors in S_LAND or S_OCEAN (discussed in Sect. 3.1.4).

3.2.3 Regional distribution

Figure 8 shows the partitioning of the total atmosphere-to-surface fluxes excluding fossil CO₂ emissions ($S_{\mathrm{LAND}}+S_{\mathrm{OCEAN}}-E_{\mathrm{LUC}}$) according to the multi-model average of the process models in the ocean and on land (GOBMs and DGVMs) and to the atmospheric inversions. Figure 8 provides information on the regional distribution of those fluxes by latitude bands. The global mean total atmosphere-to-surface CO₂ fluxes from process models for 2008–2017 is 3.7±1.2 GtC yr⁻¹. This is below but still within the uncertainty range of a global mean atmosphere-to-surface flux of 4.6±0.5 GtC yr⁻¹ inferred from the carbon budget (E_FF–G_ATM in Eq. 1; Table 6). The total atmosphere-to-surface CO₂ fluxes from the four inversions are very similar, ranging from 4.7 to 5.0 GtC yr⁻¹, consistent with the carbon budget as expected from the constraints on the inversions and the adjustments to the same E_FF distribution (see Sect. 2.7).

Figure 8CO₂ fluxes between the atmosphere and the surface ($S_{\mathrm{OCEAN}}+S_{\mathrm{LAND}}-E_{\mathrm{LUC}}$) by latitude bands for the (top) globe (second row), north (north of 30^∘ N), (third row) tropics (30^∘ S–30^∘ N), and (bottom) south (south of 30^∘ S) and (left) total, (middle) land only (S_LAND−E_LUC), and (right) ocean only. Positive values indicate a flux from the atmosphere to the land and/or ocean. Estimates from the combination of the process models for the land and oceans are shown (black for the total, green for the land, blue for the ocean) with ±1σ of the model ensemble (in grey). Results from the atmospheric inversions (pink lines) and from the pCO₂-based flux products (dark blue lines) are also shown.

Download

In the south (south of 30^∘ S), the atmospheric inversions suggest an atmosphere-to-surface flux for 2008–2017 of around 1.6–1.7 GtC yr⁻¹, close to the process models' estimate of 1.4±0.7 GtC yr⁻¹ (Fig. 8). The interannual variability in the south is low because of the dominance of ocean area with low variability compared to land areas. The split between land (S_LAND−E_LUC) and ocean (S_OCEAN) shows a small contribution to variability in the south coming from the land, with no consistency between the DGVMs and the inversions or among inversions. This is expected due to the difficulty of separating exactly the land and oceanic fluxes when viewed from atmospheric observations alone. The oceanic variability in the south is estimated to be significant in the two flux products and in at least one of the inversions, with decadal variability in around 0.5 GtC yr⁻¹. The GOBMs do not reproduce this variability.

In the tropics (30^∘ S–30^∘ N), both the atmospheric inversions and process models suggest the total carbon balance in this region is close to neutral on average over the past decade, with atmosphere-to-surface fluxes for the 2008–2017 average ranging between −0.4 and +0.4 GtC yr⁻¹. The agreement between inversions and models is significantly better for the last decade than for any previous decade, although the reasons for this better agreement are still unclear. Both the process models and the inversions consistently allocate more year-to-year variability in CO₂ fluxes to the tropics compared to the north (north of 30^∘ N; Fig. 8). The split between the land and ocean indicates the land is the origin of most of the tropical variability, consistently among models (both for the land and for the ocean) and inversions. The oceanic variability in the tropics is similar among models and with the two ocean flux products, reflected in their lower observational mismatch (Sect. 3.1.3). While the inversions indicate that atmosphere-to-land CO₂ fluxes are more variable than atmosphere-to-ocean CO₂ fluxes in the tropics, the correspondence between the inversions and the ocean flux products or GOBMs is much poorer.

In the north (north of 30^∘ N), the inversions and process models show less agreement on the magnitude of the atmosphere-to-land flux, with the ensemble mean of the process models suggesting a total Northern Hemisphere sink for 2008–2017 of 2.2±0.6 GtC yr⁻¹, likely below the estimates from the inversions ranging from 2.6 to 3.6 GtC yr⁻¹ (Fig. 8). The discrepancy in the north-tropics distribution of CO₂ fluxes between the inversions and models arises from the differences in mean fluxes over the northern land. This discrepancy is also evidenced over the previous decade and highlights not only persistent issues with the quantification of the drivers of the net land CO₂ flux (Arneth et al., 2017; Huntzinger et al., 2017) but also the distribution of atmosphere-to-land fluxes between the tropics and higher latitudes that is particularly marked in previous decades, as highlighted previously (Stephens et al., 2007; Baccini et al., 2017; Schimel et al., 2015).

Differences between inversions may be related for example to differences in their interhemispheric transport, and other inversion settings (Table A3). Separate analysis has shown that the influence of the chosen prior land and ocean fluxes is minor compared to other aspects of each inversion. In comparison to the previous global carbon budget publication, the fossil fuel inputs were adjusted to match those of E_FF used in this analysis (see Sect. 2.7), therefore removing differences due to prior fossil emissions. Differences between inversions and the ensemble of process models in the north cannot be simply explained. They could either reflect a bias in the inversions or missing processes or biases in the process models, such as the lack of adequate parameterisations for forest management in the north and for forest degradation emissions in the tropics for the DGVMs. The estimated contribution of the north and its uncertainty from process models is sensitive to both the ensemble of process models used and the specifics of each inversion.

Resolving the differences in the Northern Hemisphere land sink will require the consideration and inclusion of larger volumes of semi-continuous observations from tall towers close to the surface CO₂ exchange. Some of these data are becoming available but not used in the current inverse models, sometimes due to the short records and sometimes because the coarse transport models cannot adequately represent these time series. Improvements in model resolution and atmospheric transport realism together with expansion of the observational record (also in the data-sparse Boreal Eurasian area) will help anchor the mid-latitude fluxes per continent. In addition, new metrics could potentially differentiate between the more and less realistic realisations of the Northern Hemisphere land sink shown in Fig. 8.

3.2.4 Budget imbalance

The budget imbalance was +0.5 GtC yr⁻¹ on average over 2008–2017. Although the uncertainties are large in each term, the sustained imbalance over this last decade suggests an overestimation of the emissions and/or an underestimation of the sinks. An origin in the land and/or ocean sink may be more likely, given the large variability in the land sink and the suspected underestimation of decadal variability in the ocean sink. An underestimate of S_LAND would also reconcile model results with inversion estimates for fluxes in the total land during the past decade (Fig. 8; Table 5). However, we cannot exclude that the budget imbalance over the last decade could partly be due to an overestimation of CO₂ emissions from land-use change, given their large uncertainty, as has been suggested elsewhere (Piao et al., 2018). More integrated use of observations in the Global Carbon Budget, either on their own or for further constraining model results, should help resolve some of the budget imbalance (Peters et al., 2017; Sect. 4).

3.3.1 CO₂ emissions

Preliminary estimates of global fossil CO₂ emissions based on BP energy statistics are for emissions growing by 1.6 % between 2016 and 2017 to 9.9±0.5 GtC in 2017 (Fig. 5), distributed among coal (40 %), oil (35 %), gas (20 %), cement (4 %), and gas flaring (0.7 %). Compared to the previous year, emissions from coal increased by 1.6 %, while emissions from oil, gas, and cement increased by 1.7 %, 3.0 %, and 1.2 %, respectively. All growth rates presented are adjusted for the leap year, unless stated otherwise.

The growth in emissions of 1.6 % in 2017 is within the range of the projected growth of 2.0 % (range of 0.8 to 3.0 %) published in Le Quéré et al. (2018) based on national emission projections for China, the US, and India and projections of gross domestic product corrected for I_FF trends for the rest of the world. The growth in emissions in 2017 for China, the US, and the rest of the world is also within their previously projected range, while the growth in India was slightly above the projection (Table 7).

Table 7Comparison of the projection with realised fossil CO₂ emissions (E_FF). The “Actual” values are the first estimate available using actual data, and the “Projected” values refer to the estimate made before the end of the year for each publication. Projections based on a different method from that described here during 2008–2014 are available in Le Quéré et al. (2016). All values are adjusted for leap years.

^a Jackson et al. (2016) and Le Quéré et al. (2015a). ^b Le Quéré et al. (2016). ^c Le Quéré et al. (2018). ^d This study.

Download Print Version | Download XLSX

In 2017, the largest absolute contributions to global CO₂ emissions were from China (27 %), the US (15 %), the EU (28 member states; 10 %), and India (7 %) while the rest of the world contributed 42 %. The percentages are the fraction of the global emissions including bunker fuels (3.1 %). These four regions account for 59 % of global CO₂ emissions. Growth rates for these countries from 2016 to 2017 were +1.5 % (China), −0.5 % (US), +1.2 % (EU28), and +3.9 % (India), with +1.9 % for the rest of the world. The per capita CO₂ emissions in 2017 were 1.1 tC person⁻¹ yr⁻¹ for the globe and were 4.4 (US), 2.0 (China), 1.9 (EU28), and 0.5 (India) tC person⁻¹ yr⁻¹ for the four highest emitting countries (Fig. 5).

In 2016 (the last year available), the largest absolute contributions to global CO₂ emissions from a consumption perspective were China (25 %), the US (16 %), the EU (12 %), and India (6 %). The difference between territorial and consumption emissions (the net emission transfer via international trade) has generally increased from 1990 to around 2005 and remained relatively stable afterwards until the last year available (2016; Fig. 5).

The global CO₂ emissions from land-use change are estimated as 1.4±0.7 GtC in 2017, close to the previous decade but with low confidence in the annual change. This brings the total CO₂ emissions from fossil fuels plus land-use change (E_FF+E_LUC) to 11.3±0.9 GtC (41.2±3 GtCO₂).

3.3.2 Partitioning among the atmosphere, ocean, and land

The growth rate in atmospheric CO₂ concentration was 4.6±0.2 GtC in 2017 (2.16±0.09 ppm; Fig. 4; Dlugokencky and Tans, 2018). This is near the 2008–2017 average of 4.7±0.1 GtC yr⁻¹ and reflects the return to normal conditions after the El Niño of 2015–2016.

The estimated ocean CO₂ sink was 2.5±0.5 GtC in 2017. All models and data products estimate a small reduction or no change in the sink (average of 0.1, ranging from +0.02 to −0.4 GtC), consistent with the return to normal conditions after the El Niño, which caused an enhanced sink in previous years (Fig. 7).

The terrestrial CO₂ sink from the model ensemble was 3.8±0.8 GtC in 2017, above the decadal average (Fig. 4) and consistent with constraints from the rest of the budget (Table 5).

The budget imbalance was +0.3 GtC in 2017, indicating, as for the last decade, a small overestimation of the emissions and/or underestimation of the sinks for that year. This imbalance is indicative only, given the large uncertainties in the estimation of the B_IM.

3.4.1 CO₂ emissions

Based on available data as of 7 November 2018 (see Sect. 2.1.5), fossil CO₂ emissions (E_FF) for 2018 are projected to increase by +2.7 % (range of 1.8 % to +3.7 %; Table 7). Our method contains several assumptions that could influence the estimate beyond the given range, and as such, it has an indicative value only. Within the given assumptions, global emissions would be 10.1±0.5 GtC (37.1±1.8 GtCO₂) in 2018. The interpretation of the 2018 emission projection is provided elsewhere (Figueres et al., 2018; Jackson et al., 2018).

Figure 9Cumulative changes during 1870–2017 and mean fluxes during 2008–2017 for the anthropogenic perturbation as defined in the legend.

Download

For China, the expected change is for an increase in emissions of +4.7 % (range of +2.0 % to +7.4 %) in 2018 compared to 2017. This is based on estimated growth in coal (+4.5 %; the main fuel source in China), oil (+3.6 %), natural gas (+17.7 %) consumption, and cement production (+1.0 %). The uncertainty range considers the variations in the difference between preliminary January–September data and final full-year data, the uncertainty in the preliminary data used for the 2017 base, and uncertainty in the evolution of energy density and carbon content of coal. See also Liu et al. (2018) for further analysis of China's projected emissions.

Table 8Cumulative CO₂ for different time periods in gigatonnes of carbon (GtC). All uncertainties are reported as ±1σ. The budget imbalance provides a measure of the discrepancies among the nearly independent estimates. Its uncertainty exceeds ±60 GtC. The method used here does not capture the loss of additional sink capacity from reduced forest cover, which is about 20 GtC and would exacerbate the budget imbalance (see Sect. 2.8.4). All values are rounded to the nearest 5 GtC and therefore columns do not necessarily add to zero.

^a Using projections for the year 2018 (Sect. 3.3). ^b This value was incorrectly reported as 145 in Le Quéré et al. (2018).

Download Print Version | Download XLSX

For the US, the EIA emission projection for 2018 combined with cement data from USGS give an increase of 2.5 % (range of +0.5 to +4.5 %) compared to 2017.

For the European Union, our projection for 2018 is for a decrease of −0.7 % (range of −2.6 % to +1.3 %) over 2017. This is based on estimates for coal of −1.2 %, oil of +1.2 %, gas of −2.9 %, and stable cement emissions.

For India, our projection for 2018 is for an increase of +6.3 % (range of 4.3 % to +8.3 %) over 2017. This is based on separate projections for coal (+7.1 %), oil (+2.9 %), gas (+6.0 %), and cement (+13.4 %).

For the rest of the world, the expected growth for 2018 is +1.8 % (range of +0.5 % to +3.0 %). This is computed using the GDP projection for the world excluding China, the US, the EU, and India of 2.8 % made by the IMF (IMF, 2018) and a decrease in I_FF of −1.0 % yr⁻¹, which is the average from 2008 to 2017. The uncertainty range is based on the standard deviation of the interannual variability in I_FF during 2008–2017 of ±0.7 % yr⁻¹ and our estimate of uncertainty in the IMF's GDP forecast of ±0.5 %.

Preliminary estimates of fire emissions in deforestation zones indicate that emissions from land-use change (E_LUC) for 2018 were below average until October and are expected to range between 0.1 and 0.2 lower than the 2008–2017 average. We therefore expect E_LUC emissions of around 1.2 GtC in 2018, for total CO₂ emissions of 11.3±0.9 GtC (41.5±3 GtCO₂).

3.4.2 Partitioning among the atmosphere, ocean, and land

The 2018 growth in atmospheric CO₂ concentration (G_ATM) is projected to be 4.9±0.7 GtC (2.3±0.3 ppm) based on MLO observations until the end of October 2018, bringing the atmospheric CO₂ concentration to an expected level of 407 ppm averaged over the year. Combining projected E_FF, E_LUC, and G_ATM suggests a combined land and ocean sink (S_LAND+S_OCEAN) of about 6.5 GtC for 2018. Although each term has large uncertainty, the oceanic sink S_OCEAN has generally low interannual variability and is likely to remain close to its 2017 value of around 2.5 GtC, leaving a rough estimated land sink S_LAND of around 4.0 GtC. If realised, it would be among the largest S_LAND values over the historical period. However, the possible onset of an El Niño at the end of 2018 could reduce S_LAND, with G_ATM returning to a high growth rate towards the end of the year.