A global gridded net ecosystem exchange (NEE) of CO₂ dataset is vital in global and regional carbon cycle studies. Top-down atmospheric inversion is one of the major methods to estimate the global NEE; however, the existing global NEE datasets generated through inversion from conventional CO₂ observations have large uncertainties in places where observational data are sparse. Here, by assimilating the GOSAT ACOS v9 XCO₂ product, we generate a 10-year (2010–2019) global monthly terrestrial NEE dataset using the Global Carbon Assimilation System, version 2 (GCASv2), which is named GCAS2021. It includes gridded ($\mathrm{1}{}^{\circ }\times \mathrm{1}{}^{\circ }$ ), globally, latitudinally, and regionally aggregated prior and posterior NEE and ocean (OCN) fluxes and prescribed wildfire (FIRE) and fossil fuel and cement (FFC) carbon emissions. Globally, the decadal mean NEE is $-\mathrm{3.73}\pm \mathrm{0.52}$ PgC yr⁻¹, with an interannual amplitude of 2.73 PgC yr⁻¹. Combining the OCN flux and FIRE and FFC emissions, the net biosphere flux (NBE) and atmospheric growth rate (AGR) as well as their inter-annual variabilities (IAVs) agree well with the estimates of the Global Carbon Budget 2020. Regionally, our dataset shows that eastern North America, the Amazon, the Congo Basin, Europe, boreal forests, southern China, and Southeast Asia are carbon sinks, while the western United States, African grasslands, Brazilian plateaus, and parts of South Asia are carbon sources. In the TRANSCOM land regions, the NBEs of temperate N. America, northern Africa, and boreal Asia are between the estimates of CMS-Flux NBE 2020 and CT2019B, and those in temperate Asia, Europe, and Southeast Asia are consistent with CMS-Flux NBE 2020 but significantly different from CT2019B. In the RECCAP2 regions, except for Africa and South Asia, the NBEs are comparable with the latest bottom-up estimate of Ciais et al. (2021). Compared with previous studies, the IAVs and seasonal cycles of NEE of this dataset could clearly reflect the impacts of extreme climates and large-scale climate anomalies on the carbon flux. The evaluations also show that the posterior CO₂ concentrations at remote sites and on a regional scale, as well as on vertical CO₂ profiles in the Asia-Pacific region, are all consistent with independent CO₂ measurements from surface flask and aircraft CO₂ observations, indicating that this dataset captures surface carbon fluxes well. We believe that this dataset can contribute to regional- or national-scale carbon cycle and carbon neutrality assessment and carbon dynamics research. The dataset can be accessed at https://doi.org/10.5281/zenodo.5829774 (Jiang, 2022).

Satellite retrievals of the column-averaged dry air mole fractions of CO₂ (XCO₂) could help to improve carbon flux estimation due to their good spatial coverage. In this study, in order to assimilate the GOSAT (Greenhouse Gases Observing Satellite) XCO₂ retrievals, the Global Carbon Assimilation System (GCAS) is upgraded with new assimilation algorithms, procedures, a localization scheme, and a higher assimilation parameter resolution. This upgraded system is referred to as GCASv2. Based on this new system, the global terrestrial ecosystem (BIO) and ocean (OCN) carbon fluxes from 1 May 2009 to 31 December 2015 are constrained using the GOSAT ACOS (Atmospheric CO₂ Observations from Space) XCO₂ retrievals (Version 7.3). The posterior carbon fluxes from 2010 to 2015 are independently evaluated using CO₂ observations from 52 surface flask sites. The results show that the posterior carbon fluxes could significantly improve the modeling of atmospheric CO₂ concentrations, with global mean bias decreases from a prior value of 1.6 ± 1.8 ppm to −0.5 ± 1.8 ppm. The uncertainty reduction (UR) of the global BIO flux is 17 %, and the highest monthly regional UR could reach 51 %. Globally, the mean annual BIO and OCN carbon sinks and their interannual variations inferred in this study are very close to the estimates of CarbonTracker 2017 (CT2017) during the study period, and the inferred mean atmospheric CO₂ growth rate and its interannual changes are also very close to the observations. Regionally, over the northern lands, the strongest carbon sinks are seen in temperate North America, followed by Europe, boreal Asia, and temperate Asia; in the tropics, there are strong sinks in tropical South America and tropical Asia, but a very weak sink in Africa. This pattern is significantly different from the estimates of CT2017, but the estimated carbon sinks for each continent and some key regions like boreal Asia and the Amazon are comparable or within the range of previous bottom-up estimates. The inversion also changes the interannual variations in carbon fluxes in most TransCom land regions, which have a better relationship with the changes in severe drought area (SDA) or leaf area index (LAI), or are more consistent with previous estimates for the impact of drought. These results suggest that the GCASv2 system works well with the GOSAT XCO₂ retrievals and shows good performance with respect to estimating the surface carbon fluxes; meanwhile, our results also indicate that the GOSAT XCO₂ retrievals could help to better understand the interannual variations in regional carbon fluxes.

In this study, both the Greenhouse Gases Observing Satellite (GOSAT) and the Orbiting Carbon Observatory 2 (OCO-2) XCO₂ retrievals produced by the NASA Atmospheric CO₂ Observations from Space (ACOS) project (version b7.3) are assimilated within the GEOS-Chem 4D-Var assimilation framework to constrain the terrestrial ecosystem carbon flux during 1 October 2014 to 31 December 2015. One inversion for the comparison, using in situ CO₂ observations, and another inversion as a benchmark for the simulated atmospheric CO₂ distributions of the real inversions, using global atmospheric CO₂ trends and referred to as the poor-man inversion, are also conducted. The estimated global and regional carbon fluxes for 2015 are shown and discussed. CO₂ observations from surface flask sites and XCO₂ retrievals from Total Carbon Column Observing Network (TCCON) sites are used to evaluate the simulated concentrations with the posterior carbon fluxes. Globally, the terrestrial ecosystem carbon sink (excluding biomass burning emissions) estimated from GOSAT data is stronger than that inferred from OCO-2 data, weaker than the in situ inversion and matches the poor-man inversion the best. Regionally, in most regions, the land sinks inferred from GOSAT data are also stronger than those from OCO-2 data, and in North America, Asia and Europe, the carbon sinks inferred from GOSAT inversion are comparable to those from in situ inversion. For the latitudinal distribution of land sinks, the satellite-based inversions suggest a smaller boreal and tropical sink but larger temperate sinks in both the Northern and Southern Hemisphere than the in situ inversion. However, OCO-2 and GOSAT generally do not agree on which continent contains the smaller or larger sinks. Evaluations using flask and TCCON observations and the comparisons with in situ and poor-man inversions suggest that only GOSAT and the in situ inversions perform better than a poor-man solution. GOSAT data can effectively improve the carbon flux estimates in the Northern Hemisphere, while OCO-2 data, with the specific version used in this study, show only slight improvement. The differences of inferred land fluxes between GOSAT and OCO-2 inversions in different regions are mainly related to the spatial coverage, the data amount and the biases of these two satellite XCO₂ retrievals.

The interannual variability (IAV) in atmospheric CO₂ growth rate (CGR) is closely connected with the El Niño–Southern Oscillation. However, sensitivities of CGR to temperature and precipitation remain largely uncertain. This paper analyzed the relationship between Mauna Loa CGR and tropical land climatic elements. We find that Mauna Loa CGR lags precipitation by 4 months with a correlation coefficient of −0.63, leads temperature by 1 month (0.77), and correlates with soil moisture (−0.65) with zero lag. Additionally, precipitation and temperature are highly correlated (−0.66), with precipitation leading by 4–5 months. Regression analysis shows that sensitivities of Mauna Loa CGR to temperature and precipitation are 2.92 ± 0.20 PgC yr⁻¹ K⁻¹ and −0.46 ± 0.07 PgC yr⁻¹ 100 mm⁻¹, respectively. Unlike some recent suggestions, these empirical relationships favor neither temperature nor precipitation as the dominant factor of CGR IAV. We further analyzed seven terrestrial carbon cycle models, from the TRENDY project, to study the processes underlying CGR IAV. All models capture well the IAV of tropical land–atmosphere carbon flux (CF_TA). Sensitivities of the ensemble mean CF_TA to temperature and precipitation are 3.18 ± 0.11 PgC yr⁻¹ K⁻¹ and −0.67 ± 0.04 PgC yr⁻¹ 100 mm⁻¹, close to Mauna Loa CGR. Importantly, the models consistently show the variability in net primary productivity (NPP) dominates CGR, rather than heterotrophic respiration. Because previous studies have proved that NPP is largely driven by precipitation in tropics, it suggests a key role of precipitation in CGR IAV despite the higher CGR correlation with temperature. Understanding the relative contribution of CO₂ sensitivity to precipitation and temperature has important implications for future carbon-climate feedback using such ''emergent constraint''.

The CO₂ efflux from soil (soil respiration (SR)) is one of the largest fluxes in the global carbon (C) cycle and its response to climate change could strongly influence future atmospheric CO₂ concentrations. Still, a large divergence of global SR estimates and its autotrophic (AR) and heterotrophic (HR) components exists among process based terrestrial ecosystem models. Therefore, alternatively derived global benchmark values are warranted for constraining the various ecosystem model output. In this study, we developed models based on the global soil respiration database (version 5.0), using the random forest (RF) method to generate the global benchmark distribution of total SR and its components. Benchmark values were then compared with the output of ten different global terrestrial ecosystem models. Our observationally derived global mean annual benchmark rates were 85.5 ± 40.4 (SD) Pg C yr⁻¹ for SR, 50.3 ± 25.0 (SD) Pg C yr⁻¹ for HR and 35.2 Pg C yr⁻¹ for AR during 1982–2012, respectively. Evaluating against the observations, the RF models showed better performance in both of SR and HR simulations than all investigated terrestrial ecosystem models. Large divergences in simulating SR and its components were observed among the terrestrial ecosystem models. The estimated global SR and HR by the ecosystem models ranged from 61.4 to 91.7 Pg C yr⁻¹ and 39.8 to 61.7 Pg C yr⁻¹, respectively. The most discrepancy lays in the estimation of AR, the difference (12.0–42.3 Pg C yr⁻¹) of estimates among the ecosystem models was up to 3.5 times. The contribution of AR to SR highly varied among the ecosystem models ranging from 18% to 48%, which differed with the estimate by RF (41%). This study generated global SR and its components (HR and AR) fluxes, which are useful benchmarks to constrain the performance of terrestrial ecosystem models.