Natural wetlands constitute the largest and most uncertain source of methane (CH₄) to the atmosphere and a large fraction of them are found in the northern latitudes. These emissions are typically estimated using process (“bottom-up”) or inversion (“top-down”) models. However, estimates from these two types of models are not independent of each other since the top-down estimates usually rely on the a priori estimation of these emissions obtained with process models. Hence, independent spatially explicit validation data are needed. Here we utilize a random forest (RF) machine-learning technique to upscale CH₄ eddy covariance flux measurements from 25 sites to estimate CH₄ wetland emissions from the northern latitudes (north of 45^∘ N). Eddy covariance data from 2005 to 2016 are used for model development. The model is then used to predict emissions during 2013 and 2014. The predictive performance of the RF model is evaluated using a leave-one-site-out cross-validation scheme. The performance (Nash–Sutcliffe model efficiency =0.47) is comparable to previous studies upscaling net ecosystem exchange of carbon dioxide and studies comparing process model output against site-level CH₄ emission data. The global distribution of wetlands is one major source of uncertainty for upscaling CH₄. Thus, three wetland distribution maps are utilized in the upscaling. Depending on the wetland distribution map, the annual emissions for the northern wetlands yield 32 (22.3–41.2, 95 % confidence interval calculated from a RF model ensemble), 31 (21.4–39.9) or 38 (25.9–49.5) Tg(CH₄) yr⁻¹. To further evaluate the uncertainties of the upscaled CH₄ flux data products we also compared them against output from two process models (LPX-Bern and WetCHARTs), and methodological issues related to CH₄ flux upscaling are discussed. The monthly upscaled CH₄ flux data products are available at https://doi.org/10.5281/zenodo.2560163 (Peltola et al., 2019).

We present inverse modelling (top down) estimates of European methane (CH₄) emissions for 2006–2012 based on a new quality-controlled and harmonised in situ data set from 18 European atmospheric monitoring stations. We applied an ensemble of seven inverse models and performed four inversion experiments, investigating the impact of different sets of stations and the use of a priori information on emissions. The inverse models infer total CH₄ emissions of 26.8 (20.2–29.7) Tg CH₄ yr⁻¹ (mean, 10th and 90th percentiles from all inversions) for the EU-28 for 2006–2012 from the four inversion experiments. For comparison, total anthropogenic CH₄ emissions reported to UNFCCC (bottom up, based on statistical data and emissions factors) amount to only 21.3 Tg CH₄ yr⁻¹ (2006) to 18.8 Tg CH₄ yr⁻¹ (2012). A potential explanation for the higher range of top-down estimates compared to bottom-up inventories could be the contribution from natural sources, such as peatlands, wetlands, and wet soils. Based on seven different wetland inventories from the Wetland and Wetland CH₄ Inter-comparison of Models Project (WETCHIMP), total wetland emissions of 4.3 (2.3–8.2) Tg CH₄ yr⁻¹ from the EU-28 are estimated. The hypothesis of significant natural emissions is supported by the finding that several inverse models yield significant seasonal cycles of derived CH₄ emissions with maxima in summer, while anthropogenic CH₄ emissions are assumed to have much lower seasonal variability. Taking into account the wetland emissions from the WETCHIMP ensemble, the top-down estimates are broadly consistent with the sum of anthropogenic and natural bottom-up inventories. However, the contribution of natural sources and their regional distribution remain rather uncertain. Furthermore, we investigate potential biases in the inverse models by comparison with regular aircraft profiles at four European sites and with vertical profiles obtained during the Infrastructure for Measurement of the European Carbon Cycle (IMECC) aircraft campaign. We present a novel approach to estimate the biases in the derived emissions, based on the comparison of simulated and measured enhancements of CH₄ compared to the background, integrated over the entire boundary layer and over the lower troposphere. The estimated average regional biases range between −40 and 20 % at the aircraft profile sites in France, Hungary and Poland.

The CO₂ source and sink distribution across Europe can be estimated in principle through inverse methods by combining CO₂ observations and atmospheric transport models. Uncertainties of such estimates are mainly due to insufficient spatiotemporal coverage of CO₂ observations and biases of the models. In order to assess the biases related to the use of different models the CO₂ concentration field over Europe has been simulated with five different Eulerian atmospheric transport models as part of the EU-funded AEROCARB project, which has the main goal to estimate the carbon balance of Europe. In contrast to previous comparisons, here both global coarse-resolution and regional higher-resolution models are included. Continuous CO₂ observations from continental, coastal and mountain sites as well as flasks sampled on aircrafts are used to evaluate the models' ability to capture the spatiotemporal variability and distribution of lower troposphere CO₂ across Europe. ¹⁴CO₂ is used in addition to evaluate separately fossil fuel signal predictions. The simulated concentrations show a large range of variation, with up to ~10 ppm higher surface concentrations over Western and Central Europe in the regional models with highest (mesoscale) spatial resolution. The simulation – data comparison reveals that generally high-resolution models are more successful than coarse models in capturing the amplitude and phasing of the observed short-term variability. At high-altitude stations the magnitude of the differences between observations and models and in between models is less pronounced, but the timing of the diurnal cycle is not well captured by the models. The data comparisons show also that the timing of the observed variability on hourly to daily time scales at low-altitude stations is generally well captured by all models. However, the amplitude of the variability tends to be underestimated. While daytime values are quite well predicted, nighttime values are generally underpredicted. This is a reflection of the different mixing regimes during day and night combined with different vertical resolution between models. In line with this finding, the agreement among models is increased when sampling in the afternoon hours only and when sampling the mixed portion of the PBL, which amounts to sampling at a few hundred meters above ground. The main recommendations resulting from the study for constraining land carbon sources and sinks using high-resolution concentration data and state-of-the art transport models through inverse methods are given in the following: 1) Low altitude stations are presently preferable in inverse studies. If high altitude stations are used then the model level that represents the specific sites should be applied, 2) at low altitude sites only the afternoon values of concentrations can be represented sufficiently well by current models and therefore afternoon values are more appropriate for constraining large-scale sources and sinks in combination with transport models, 3) even when using only afternoon values it is clear that data sampled several hundred meters above ground can be represented substantially more robustly in models than surface station records, which emphasize the use of tower data in inverse studies and finally 4) traditional large scale transport models seem not sufficient to resolve fine-scale features associated with fossil fuel emissions, as well as larger-scale features like the concentration distribution above the south-western Europe. It is therefore recommended to use higher resolution models for interpretation of continental data in future studies.

European CH4 and N2O emissions are estimated for 2006 and 2007 using four inverse modelling systems, based on different global and regional Eulerian and Lagrangian transport models. This ensemble approach is designed to provide more realistic estimates of the overall uncertainties in the derived emissions, which is particularly important for verifying bottom-up emission inventories. We use continuous observations from 10 European stations (including 5 tall towers) for CH4 and 9 continuous stations for N2O, complemented by additional European and global discrete air sampling sites. The available observations mainly constrain CH4 and N2O emissions from north-western and eastern Europe. The inversions are strongly driven by the observations and the derived total emissions of larger countries show little dependence on the emission inventories used a priori. Three inverse models yield 26–56% higher total CH4 emissions from north-western and eastern Europe compared to bottom-up emissions reported to the UNFCCC, while one model is close to the UNFCCC values. In contrast, the inverse modelling estimates of European N2O emissions are in general close to the UNFCCC values, with the overall range from all models being much smaller than the UNFCCC uncertainty range for most countries. Our analysis suggests that the reported uncertainties for CH4 emissions might be underestimated, while those for N2O emissions are likely overestimated.

Significance: We quantified a 36-y trend of advanced spring recovery of carbon uptake across the northern hemisphere boreal evergreen forest zone. From this trend, we estimated the corresponding change in global gross primary production (GPP) and further quantified the magnitude and spatiotemporal variability of spring GPP, that is, the cross-photosynthetic carbon uptake by forest. Our main findings are the following: ( i ) We developed a proxy indicator for spring recovery from in situ flux data on CO ₂ exchange and recent satellite snowmelt products and ( ii ) we established a relation between spring recovery and carbon uptake to assess changes in springtime carbon exchange showing a major advance in the CO ₂ sink.