Atmospheric Inverse Modeling Research Articles

Artificial Intelligence Approach for Estimating Dairy Methane Emissions.

California's dairy sector accounts for ∼50% of anthropogenic CH4 emissions in the state's greenhouse gas (GHG) emission inventory. Although California dairy facilities' location and herd size vary over time, atmospheric inverse modeling studies rely on decade-old facility-scale geospatial information. For the first time, we apply artificial intelligence (AI) to aerial imagery to estimate dairy CH4 emissions from California's San Joaquin Valley (SJV), a region with ∼90% of the state's dairy population. Using an AI method, we process 316,882 images to estimate the facility-scale herd size across the SJV. The AI approach predicts herd size that strongly (>95%) correlates with that made by human visual inspection, providing a low-cost alternative to the labor-intensive inventory development process. We estimate SJV's dairy enteric and manure CH4 emissions for 2018 to be 496-763 Gg/yr (mean = 624; 95% confidence) using the predicted herd size. We also apply our AI approach to estimate CH4 emission reduction from anaerobic digester deployment. We identify 162 large (90th percentile) farms and estimate a CH4 reduction potential of 83 Gg CH4/yr for these large facilities from anaerobic digester adoption. The results indicate that our AI approach can be applied to characterize the manure system (e.g., use of an anaerobic lagoon) and estimate GHG emissions for other sectors.

Plant gross primary production, plant respiration and carbonyl sulfide emissions over the globe inferred by atmospheric inverse modelling

Abstract. Carbonyl sulfide (COS), a trace gas showing striking similarity to CO2 in terms of biochemical diffusion pathway into leaves, has been recognized as a promising indicator of the plant gross primary production (GPP), the amount of carbon dioxide that is absorbed through photosynthesis by terrestrial ecosystems. However, large uncertainties about the other components of its atmospheric budget prevent us from directly relating the atmospheric COS measurements to GPP. The largest uncertainty comes from the closure of its atmospheric budget, with a source component missing. Here, we explore the benefit of assimilating both COS and CO2 measurements into the LMDz atmospheric transport model to obtain consistent information on GPP, plant respiration and COS budget. To this end, we develop an analytical inverse system that optimizes biospheric fluxes for the 15 plant functional types (PFTs) defined in the ORCHIDEE global land surface model. Plant uptake of COS is parameterized as a linear function of GPP and of the leaf relative uptake (LRU), which is the ratio of COS to CO2 deposition velocities in plants. A possible scenario for the period 2008–2019 leads to a global biospheric sink of 800 GgS yr−1, with higher absorption in the high latitudes and higher oceanic emissions between 400 and 600 GgS yr−1 most of which is located in the tropics. As for the CO2 budget, the inverse system increases GPP in the high latitudes by a few GtC yr−1 without modifying the respiration compared to the ORCHIDEE fluxes used as a prior. In contrast, in the tropics the system tends to weaken both respiration and GPP. The optimized components of the COS and CO2 budgets have been evaluated against independent measurements over North America, the Pacific Ocean, at three sites in Japan and at one site in France. Overall, the posterior COS concentrations are in better agreement with the COS retrievals at 250 hPa from the MIPAS satellite and with airborne measurements made over North America and the Pacific Ocean. The system seems to have rightly corrected the underestimated GPP over the high latitudes. However, the change in seasonality of GPP in the tropics disagrees with solar-induced fluorescence (SIF) data. The decline in biospheric sink in the Amazon driven by the inversion also disagrees with MIPAS COS retrievals at 250 hPa, highlighting the lack of observational constraints in this region. Moreover, the comparison with the surface measurements in Japan and France suggests misplaced sources in the prior anthropogenic inventory, emphasizing the need for an improved inventory to better partition oceanic and continental sources in Asia and Europe.

Analysis of atmospheric greenhouse gases in north Xinjiang

Flask air samples were collected weekly at the Akedala (AKD) regional atmosphere station in north Xinjiang, China, and atmospheric carbon dioxide (CO2) and methane (CH4) mole fractions were measured using cavity-ring down spectroscopy. AKD station is under a typical temperate continental semiarid and arid monsoon climate with westerly air flow. This paper presents the long-term trends and seasonal variations in atmospheric CO2 and CH4 concentrations for the first time and discusses the possible influencing mechanisms. The CO2 and CH4 background mixing ratios were 412.1 ± 0.36 ppm and 1968.9 ± 0.95 ppb, respectively, in 2018. The growth rates were 2.62 ppm y−1 and 9.67 ppb y−1 for CO2 and CH4, respectively, during 2010–2018, which were higher than the surface measurements at Waliguan Baseline Observatory at similar latitudes. The averaged CO2 seasonal cycle showed a minimum in summer and a maximum in winter, and the annual amplitude was high at 18.7 ppm. The CH4 seasonal cycle showed a rapid recovery in summer during July and August. Back-trajectory analysis showed that the atmospheric information at AKD station can provide some implications for uptake from Kazakhstan and Russia, as well as information about natural and anthropogenic emissions in northern Xinjiang. This study revealed that the long-term upward trend of the CO2 concentration in AKD might be attributed to emissions from energy consumption of heavy industry. The comprehensive effects of livestock, geological micro seepage and oil and gas production, and vertical transport also enhanced the atmospheric CH4 concentrations. The results of this study may improve our basic understanding of observed atmospheric CO2 and CH4 in China and Central Asia and the underlying fluxes to be used in atmospheric inversion models.

The CO2 record at the Amazon Tall Tower Observatory: A new opportunity to study processes on seasonal and inter-annual scales.

High-quality atmospheric CO2 measurements are sparse in Amazonia, but can provide critical insights into the spatial and temporal variability of sources and sinks of CO2 . In this study, we present the first 6years (2014-2019) of continuous, high-precision measurements of atmospheric CO2 at the Amazon Tall Tower Observatory (ATTO, 2.1°S, 58.9°W). After subtracting the simulated background concentrations from our observational record, we define a CO2 regional signal ( ) that has a marked seasonal cycle with an amplitude of about 4ppm. At both seasonal and inter-annual scales, we find differences in phase between and the local eddy covariance net ecosystem exchange (EC-NEE), which is interpreted as an indicator of a decoupling between local and non-local drivers of . In addition, we present how the 2015-2016 El Niño-induced drought was captured by our atmospheric record as a positive 2σ anomaly in both the wet and dry season of 2016. Furthermore, we analyzed the observed seasonal cycle and inter-annual variability of together with net ecosystem exchange (NEE) using a suite of modeled flux products representing biospheric and aquatic CO2 exchange. We use both non-optimized and optimized (i.e., resulting from atmospheric inverse modeling) NEE fluxes as input in an atmospheric transport model (STILT). The observed shape and amplitude of the seasonal cycle was captured neither by the simulations using the optimized fluxes nor by those using the diagnostic Vegetation and Photosynthesis Respiration Model (VPRM). We show that including the contribution of CO2 from river evasion improves the simulated shape (not the magnitude) of the seasonal cycle when using a data-driven non-optimized NEE product (FLUXCOM). The simulated contribution from river evasion was found to be 25% of the seasonal cycle amplitude. Our study demonstrates the importance of the ATTO record to better understand the Amazon carbon cycle at various spatial and temporal scales.

Fluxes of Carbon Dioxide From Managed Ecosystems Estimated by National Inventories Compared to Atmospheric Inverse Modeling

AbstractThe UNFCCC reporting guidelines on annual emission inventories have been encouraging comparison of national inventory reports with atmospheric measurements or atmospheric inverse modeling. We have initiated a framework to compare two CO2 atmospheric inversions with the net agricultural, forestry and other land use (AFOLU) fluxes reported by 10 parties to the UNFCCC. Our study highlights large conceptual differences or ambiguities (temporal support, spatial perimeter, scope of the processes) between inventory reports and between these reports and what the atmosphere sees. We recommend that more effort be invested into documenting the inventories in order to allow unambiguous conversion between reported values and the way they are seen by the atmosphere. We also highlight the potential of Bayesian atmospheric inversion products generated in quasi‐near‐real time with robust error bars and consistently across the world, to directly inform international CO2 flux accounting for the AFOLU sector in large countries or groups of countries.

Anthropogenic and natural controls on atmospheric <i>δ</i><sup>13</sup>C-CO<sub>2</sub> variations in the Yangtze River delta: insights from a carbon isotope modeling framework

Abstract. The atmospheric carbon dioxide (CO2) mixing ratio and its carbon isotope (δ13C-CO2) composition contain important CO2 sink and source information spanning from ecosystem to global scales. The observation and simulation for both CO2 and δ13C-CO2 can be used to constrain regional emissions and better understand the anthropogenic and natural mechanisms that control δ13C-CO2 variations. Such work remains rare for urban environments, especially megacities. Here, we used near-continuous CO2 and δ13C-CO2 measurements, from September 2013 to August 2015, and inverse modeling to constrain the CO2 budget and investigate the main factors that dominated δ13C-CO2 variations for the Yangtze River delta (YRD) region, one of the largest anthropogenic CO2 hotspots and densely populated regions in China. We used the WRF-STILT model framework with category-specified EDGAR v4.3.2 CO2 inventories to simulate hourly CO2 mixing ratios and δ13C-CO2, evaluated these simulations with observations, and constrained the total anthropogenic CO2 emission. We show that (1) top-down and bottom-up estimates of anthropogenic CO2 emissions agreed well (bias < 6 %) on an annual basis, (2) the WRF-STILT model can generally reproduce the observed diel and seasonal atmospheric δ13C-CO2 variations, and (3) anthropogenic CO2 emissions played a much larger role than ecosystems in controlling the δ13C-CO2 seasonality. When excluding ecosystem respiration and photosynthetic discrimination in the YRD area, δ13C-CO2 seasonality increased from 1.53 ‰ to 1.66 ‰. (4) Atmospheric transport processes in summer amplified the cement CO2 enhancement proportions in the YRD area, which dominated monthly δs (the mixture of δ13C-CO2 from all regional end-members) variations. These findings show that the combination of long-term atmospheric carbon isotope observations and inverse modeling can provide a powerful constraint on the carbon cycle of these complex megacities.

The consolidated European synthesis of CH4 and N2O emissions for the European Union and United Kingdom: 1990–2017

Abstract. Reliable quantification of the sources and sinks of greenhouse gases, together with trends and uncertainties, is essential to monitoring the progress in mitigating anthropogenic emissions under the Paris Agreement. This study provides a consolidated synthesis of CH4 and N2O emissions with consistently derived state-of-the-art bottom-up (BU) and top-down (TD) data sources for the European Union and UK (EU27 + UK). We integrate recent emission inventory data, ecosystem process-based model results and inverse modeling estimates over the period 1990–2017. BU and TD products are compared with European national greenhouse gas inventories (NGHGIs) reported to the UN climate convention UNFCCC secretariat in 2019. For uncertainties, we used for NGHGIs the standard deviation obtained by varying parameters of inventory calculations, reported by the member states (MSs) following the recommendations of the IPCC Guidelines. For atmospheric inversion models (TD) or other inventory datasets (BU), we defined uncertainties from the spread between different model estimates or model-specific uncertainties when reported. In comparing NGHGIs with other approaches, a key source of bias is the activities included, e.g., anthropogenic versus anthropogenic plus natural fluxes. In inversions, the separation between anthropogenic and natural emissions is sensitive to the geospatial prior distribution of emissions. Over the 2011–2015 period, which is the common denominator of data availability between all sources, the anthropogenic BU approaches are directly comparable, reporting mean emissions of 20.8 Tg CH4 yr−1 (EDGAR v5.0) and 19.0 Tg CH4 yr−1 (GAINS), consistent with the NGHGI estimates of 18.9 ± 1.7 Tg CH4 yr−1. The estimates of TD total inversions give higher emission estimates, as they also include natural emissions. Over the same period regional TD inversions with higher-resolution atmospheric transport models give a mean emission of 28.8 Tg CH4 yr−1. Coarser-resolution global TD inversions are consistent with regional TD inversions, for global inversions with GOSAT satellite data (23.3 Tg CH4 yr−1) and surface network (24.4 Tg CH4 yr−1). The magnitude of natural peatland emissions from the JSBACH–HIMMELI model, natural rivers and lakes emissions, and geological sources together account for the gap between NGHGIs and inversions and account for 5.2 Tg CH4 yr−1. For N2O emissions, over the 2011–2015 period, both BU approaches (EDGAR v5.0 and GAINS) give a mean value of anthropogenic emissions of 0.8 and 0.9 Tg N2O yr−1, respectively, agreeing with the NGHGI data (0.9 ± 0.6 Tg N2O yr−1). Over the same period, the average of the three total TD global and regional inversions was 1.3 ± 0.4 and 1.3 ± 0.1 Tg N2O yr−1, respectively. The TD and BU comparison method defined in this study can be operationalized for future yearly updates for the calculation of CH4 and N2O budgets both at the EU+UK scale and at the national scale. The referenced datasets related to figures are visualized at https://doi.org/10.5281/zenodo.4590875 (Petrescu et al., 2020b).

Gridded fossil CO2 emissions and related O2 combustion consistent with national inventories 1959\u20132018

Quantification of CO2 fluxes at the Earth’s surface is required to evaluate the causes and drivers of observed increases in atmospheric CO2 concentrations. Atmospheric inversion models disaggregate observed variations in atmospheric CO2 concentration to variability in CO2 emissions and sinks. They require prior constraints fossil CO2 emissions. Here we describe GCP-GridFED (version 2019.1), a gridded fossil emissions dataset that is consistent with the national CO2 emissions reported by the Global Carbon Project (GCP). GCP-GridFEDv2019.1 provides monthly fossil CO2 emissions estimates for the period 1959–2018 at a spatial resolution of 0.1°. Estimates are provided separately for oil, coal and natural gas, for mixed international bunker fuels, and for the calcination of limestone during cement production. GCP-GridFED also includes gridded estimates of O2 uptake based on oxidative ratios for oil, coal and natural gas. It will be updated annually and made available for atmospheric inversions contributing to GCP global carbon budget assessments, thus aligning the prior constraints on top-down fossil CO2 emissions with the bottom-up estimates compiled by the GCP.

Responses of vegetation greenness and carbon cycle to extreme droughts in China

Increasing drought stress under climate change exerting large impacts on vegetation has aroused growing concern, yet how vegetation greenness and carbon cycle respond to extreme drought is not well understood. In this paper, we studied this issue with various data sources including satellite observation, flux sites upscaling estimation, atmospheric inversion, and Dynamic Global Vegetation Model (DGVM) simulations over China during the period 2000–2015. We found a highly consistent negative response of greenness and carbon fluxes to extreme droughts, with the largest reduction in Gross Primary Productivity (GPP, a standardized anomaly of -1.26 ~ -0.33), a slightly smaller extent decrease of Net Ecosystem Productivity (NEP, -0.74 ~ -0.14) due to concurrent decrease of respiration, and reduced Leaf Area Index (LAI, -0.39 ~ -0.20). Besides, there was a large spatial heterogeneity across different regions and biomes, where Loess Plateau was one of the most vulnerable regions. The ambient climate conditions were found to modulate the drought response. Drought sensitivity of greenness and GPP peaked where mean annual temperature is around 25 °C, while that of NEP decreased along increasing precipitation gradients.

Eight-Year Estimates of Methane Emissions from Oil and Gas Operations in Western Canada Are Nearly Twice Those Reported in Inventories.

The provinces of Alberta and Saskatchewan account for 70% of Canada's methane emissions from the oil and gas sector. In 2018, the Government of Canada introduced methane regulations to reduce emissions from the sector by 40-45% from the 2012 levels by 2025. Complementary to inventory accounting methods, the effectiveness of regulatory practices to reduce emissions can be assessed using atmospheric measurements and inverse models. Total anthropogenic (oil and gas, agriculture, and waste) emission rates of methane from 2010 to 2017 in Alberta and Saskatchewan were derived using hourly atmospheric methane measurements over a six-month winter period from October to March. Scaling up the winter estimate to annual indicated an anthropogenic emission rate of 3.7 ± 0.7 MtCH4/year, about 60% greater than that reported in Canada's National Inventory Report (2.3 MtCH4). This discrepancy is tied primarily to the oil and gas sector emissions as the reported emissions from livestock operations (0.6 MtCH4) are well substantiated in both top-down and bottom-up estimates and waste management (0.1 MtCH4) emissions are small. The resulting estimate of 3.0 MtCH4 from the oil and gas sector is nearly twice that reported in Canada's National Inventory (1.6 MtCH4).

Bayesian spatio-temporal inference of trace gas emissions using an integrated nested Laplacian approximation and Gaussian Markov random fields

Abstract. We present a method to infer spatially and spatio-temporally correlated emissions of greenhouse gases from atmospheric measurements and a chemical transport model. The method allows fast computation of spatial emissions using a hierarchical Bayesian framework as an alternative to Markov chain Monte Carlo algorithms. The spatial emissions follow a Gaussian process with a Matérn correlation structure which can be represented by a Gaussian Markov random field through a stochastic partial differential equation approach. The inference is based on an integrated nested Laplacian approximation (INLA) for hierarchical models with Gaussian latent fields. Combining an autoregressive temporal correlation and the Matérn field provides a full spatio-temporal correlation structure. We first demonstrate the method on a synthetic data example and follow this using a well-studied test case of inferring UK methane emissions from tall tower measurements of atmospheric mole fraction. Results from these two test cases show that this method can accurately estimate regional greenhouse gas emissions, accounting for spatio-temporal uncertainties that have traditionally been neglected in atmospheric inverse modelling.

A Geostatistical Framework for Quantifying the Imprint of Mesoscale Atmospheric Transport on Satellite Trace Gas Retrievals

AbstractNational Aeronautics and Space Administration's Orbiting Carbon Observatory‐2 (OCO‐2) satellite provides observations of total column‐averaged CO2 mole fractions ( ) at high spatial resolution that may enable novel constraints on surface‐atmosphere carbon fluxes. Atmospheric inverse modeling provides an approach to optimize surface fluxes at regional scales, but the accuracy of the fluxes from inversion frameworks depends on key inputs, including spatially and temporally dense CO2 observations and reliable representations of atmospheric transport. Since observations are sensitive to both synoptic and mesoscale variations within the free troposphere, horizontal atmospheric transport imparts substantial variations in these data and must be either resolved explicitly by the atmospheric transport model or accounted for within the error covariance budget provided to inverse frameworks. Here, we used geostatistical techniques to quantify the imprint of atmospheric transport in along‐track OCO‐2 soundings. We compare high‐pass‐filtered (<250 km, spatial scales that primarily isolate mesoscale or finer‐scale variations) along‐track spatial variability in and from OCO‐2 tracks to temporal synoptic and mesoscale variability from ground‐based and observed by nearby Total Carbon Column Observing Network sites. Mesoscale atmospheric transport is found to be the primary driver of along‐track, high‐frequency variability for OCO‐2 . For , both mesoscale transport variability and spatially coherent bias associated with other elements of the OCO‐2 retrieval state vector are important drivers of the along‐track variance budget.

Methane budget of East Asia, 1990–2015: A bottom-up evaluation

The regional budget of methane (CH4) emissions for East Asia, a crucial region in the global greenhouse gas budget, was quantified for 1990–2015 with a bottom-up method based on inventories and emission model simulations. Anthropogenic emissions associated with fossil fuel extraction, industrial activities, waste management, and agricultural activities were derived from the Emission Database for Global Atmospheric Research version 4.3.2 and compared with other inventories. Emissions from natural wetlands and CH4 uptake by upland soil oxidation were estimated using the Vegetation Integrative SImulator for Trace gases (VISIT), a biogeochemical model that considers historical land use and climatic conditions. Emissions from biomass burning and termites were calculated using satellite and land-use data combined with empirical emission factors. The resulting average annual estimated CH4 budget for 2000–2012 indicated that East Asia was a net source of 67.3 Tg CH4 yr−1, of which 88.8% was associated with anthropogenic emissions. The uncertainty (±standard deviation) of this estimate, ±14 Tg CH4 yr−1, stemmed from data and model inconsistencies. The increase of the net flux from 60.2 Tg CH4 yr−1 in 1990 to 78.0 Tg CH4 yr−1 in 2012 was due mainly to increased emissions by the fossil fuel extraction and livestock sectors. Our results showed that CH4 was a crucial component of the regional greenhouse gas budget. A spatial analysis using 0.25° × 0.25° grid cells revealed emission hotspots in urban areas, agricultural areas, and wetlands. These hotspots were surrounded by weak sinks in upland areas. The estimated natural and anthropogenic emissions fell within the range of independent estimates, including top-down estimates from atmospheric inversion models. Such a regional accounting is an effective way to elucidate climatic forcings and to develop mitigation policies. Further studies, however, are required to reduce the uncertainties in the budget.

Quantifying the UK's carbon dioxide flux: an atmospheric inverse modelling approach using a regional measurement network

Abstract. We present a method to derive atmospheric-observation-based estimates of carbon dioxide (CO2) fluxes at the national scale, demonstrated using data from a network of surface tall-tower sites across the UK and Ireland over the period 2013–2014. The inversion is carried out using simulations from a Lagrangian chemical transport model and an innovative hierarchical Bayesian Markov chain Monte Carlo (MCMC) framework, which addresses some of the traditional problems faced by inverse modelling studies, such as subjectivity in the specification of model and prior uncertainties. Biospheric fluxes related to gross primary productivity and terrestrial ecosystem respiration are solved separately in the inversion and then combined a posteriori to determine net ecosystem exchange of CO2. Two different models, Data Assimilation Linked Ecosystem Carbon (DALEC) and Joint UK Land Environment Simulator (JULES), provide prior estimates for these fluxes. We carry out separate inversions to assess the impact of these different priors on the posterior flux estimates and evaluate the differences between the prior and posterior estimates in terms of missing model components. The Numerical Atmospheric dispersion Modelling Environment (NAME) is used to relate fluxes to the measurements taken across the regional network. Posterior CO2 estimates from the two inversions agree within estimated uncertainties, despite large differences in the prior fluxes from the different models. With our method, averaging results from 2013 and 2014, we find a total annual net biospheric flux for the UK of 8±79 Tg CO2 yr−1 (DALEC prior) and 64±85 Tg CO2 yr−1 (JULES prior), where negative values represent an uptake of CO2. These biospheric CO2 estimates show that annual UK biospheric sources and sinks are roughly in balance. These annual mean estimates consistently indicate a greater net release of CO2 than the prior estimates, which show much more pronounced uptake in summer months.

Quantifying the Impact of Atmospheric Transport Uncertainty on CO2 Surface Flux Estimates.

We show that transport differences between two commonly used global chemical transport models, GEOS‐Chem and TM5, lead to systematic space‐time differences in modeled distributions of carbon dioxide and sulfur hexafluoride. The distribution of differences suggests inconsistencies between the transport simulated by the models, most likely due to the representation of vertical motion. We further demonstrate that these transport differences result in systematic differences in surface CO2 flux estimated by a collection of global atmospheric inverse models using TM5 and GEOS‐Chem and constrained by in situ and satellite observations. While the impact on inferred surface fluxes is most easily illustrated in the magnitude of the seasonal cycle of surface CO2 exchange, it is the annual carbon budgets that are particularly relevant for carbon cycle science and policy. We show that inverse model flux estimates for large zonal bands can have systematic biases of up to 1.7 PgC/year due to large‐scale transport uncertainty. These uncertainties will propagate directly into analysis of the annual meridional CO2 flux gradient between the tropics and northern midlatitudes, a key metric for understanding the location, and more importantly the processes, responsible for the annual global carbon sink. The research suggests that variability among transport models remains the largest source of uncertainty across global flux inversion systems and highlights the importance both of using model ensembles and of using independent constraints to evaluate simulated transport.

Characterizing uncertainties in atmospheric inversions of fossil fuel CO<sub>2</sub> emissions in California

Abstract. Atmospheric inverse modelling has become an increasingly useful tool for evaluating emissions of greenhouse gases including methane, nitrous oxide, and synthetic gases such as hydrofluorocarbons (HFCs). Atmospheric inversions for emissions of CO2 from fossil fuel combustion (ffCO2) are currently being developed. The aim of this paper is to investigate potential errors and uncertainties related to the spatial and temporal prior representation of emissions and modelled atmospheric transport for the inversion of ffCO2 emissions in the US state of California. We perform simulation experiments based on a network of ground-based observations of CO2 concentration and radiocarbon in CO2 (a tracer of ffCO2), combining prior (bottom-up) emission models and transport models currently used in many atmospheric studies. The potential effect of errors in the spatial and temporal distribution of prior emission estimates is investigated in experiments by using perturbed versions of the emission estimates used to create the pseudo-data. The potential effect of transport error was investigated by using three different atmospheric transport models for the prior and pseudo-data simulations. We find that the magnitude of biases in posterior total state emissions arising from errors in the spatial and temporal distribution in prior emissions in these experiments are 1 %–15 % of posterior total state emissions and are generally smaller than the 2σ uncertainty in posterior emissions. Transport error in these experiments introduces biases of −10 % to +6 % into posterior total state emissions. Our results indicate that uncertainties in posterior total state ffCO2 estimates arising from the choice of prior emissions or atmospheric transport model are on the order of 15 % or less for the ground-based network in California we consider. We highlight the need for temporal variations to be included in prior emissions and for continuing efforts to evaluate and improve the representation of atmospheric transport for regional ffCO2 inversions.

Atmospheric observation-based estimation of fossil fuel CO2 emissions from regions of central and southern California

Combustion of fossil fuel is the dominant source of greenhouse gas emissions to the atmosphere in California. Here, we describe radiocarbon (14CO2) measurements and atmospheric inverse modeling to estimate fossil fuel CO2 (ffCO2) emissions for 2009–2012 from a site in central California, and for June 2013–May 2014 from two sites in southern California. A priori predicted ffCO2 mixing ratios are computed based on regional atmospheric transport model (WRF-STILT) footprints and an hourly ffCO2 prior emission map (Vulcan 2.2). Regional inversions using observations from the central California site suggest that emissions from the San Francisco Bay Area (SFBA) are higher in winter and lower in summer. Taking all years together, the average of a total of fifteen 3-month inversions from 2009 to 2012 suggests ffCO2 emissions from SFBA were within 6 ± 35% of the a priori estimate for that region, where posterior emission uncertainties are reported as 95% confidence intervals. Results for four 3-month inversions using measurements in Los Angeles South Coast Air Basin (SoCAB) during June 2013–May 2014 suggest that emissions in SoCAB are within 13 ± 28% of the a priori estimate for that region, with marginal detection of any seasonality. While emissions from the SFBA and SoCAB urban regions (containing ~50% of prior emissions from California) are constrained by the observations, emissions from the remaining regions are less constrained, suggesting that additional observations will be valuable to more accurately estimate total ffCO2 emissions from California as a whole.

Global atmospheric CO2 inverse models converging on neutral tropical land exchange, but disagreeing on fossil fuel and atmospheric growth rate.

We have compared a suite of recent global CO2 atmospheric inversion results to independent airborne observations and to each other, to assess their dependence on differences in northern extratropical (NET) vertical transport and to identify some of the drivers of model spread. We evaluate posterior CO2 concentration profiles against observations from the High-Performance Instrumented Airborne Platform for Environmental Research (HIAPER) Pole-to-Pole Observations (HIPPO) aircraft campaigns over the mid-Pacific in 2009–2011. Although the models differ in inverse approaches, assimilated observations, prior fluxes, and transport models, their broad latitudinal separation of land fluxes has converged significantly since the Atmospheric Carbon Cycle Inversion Intercomparison (TransCom 3) and the REgional Carbon Cycle Assessment and Processes (RECCAP) projects, with model spread reduced by 80% since TransCom 3 and 70% since RECCAP. Most modeled CO2 fields agree reasonably well with the HIPPO observations, specifically for the annual mean vertical gradients in the Northern Hemisphere. Northern Hemisphere vertical mixing no longer appears to be a dominant driver of northern versus tropical (T) annual flux differences. Our newer suite of models still gives northern extratropical land uptake that is modest relative to previous estimates (Gurney et al., 2002; Peylin et al., 2013) and near-neutral tropical land uptake for 2009–2011. Given estimates of emissions from deforestation, this implies a continued uptake in intact tropical forests that is strong relative to historical estimates (Gurney et al., 2002; Peylin et al., 2013). The results from these models for other time periods (2004–2014, 2001–2004, 1992–1996) and reevaluation of the TransCom 3 Level 2 and RECCAP results confirm that tropical land carbon fluxes including deforestation have been near neutral for several decades. However, models still have large disagreements on ocean–land partitioning. The fossil fuel (FF) and the atmospheric growth rate terms have been thought to be the best-known terms in the global carbon budget, but we show that they currently limit our ability to assess regional-scale terrestrial fluxes and ocean–land partitioning from the model ensemble.

Reducing errors in aircraft atmospheric inversion estimates of point-source emissions: the Aliso Canyon natural gas leak as a natural tracer experiment

Urban greenhouse gas (GHG) flux estimation with atmospheric measurements and modeling, i.e. the ‘top-down’ approach, can potentially support GHG emission reduction policies by assessing trends in surface fluxes and detecting anomalies from bottom-up inventories. Aircraft-collected GHG observations also have the potential to help quantify point-source emissions that may not be adequately sampled by fixed surface tower-based atmospheric observing systems. Here, we estimate CH4 emissions from a known point source, the Aliso Canyon natural gas leak in Los Angeles, CA from October 2015–February 2016, using atmospheric inverse models with airborne CH4 observations from twelve flights ≈4 km downwind of the leak and surface sensitivities from a mesoscale atmospheric transport model. This leak event has been well-quantified previously using various methods by the California Air Resources Board, thereby providing high confidence in the mass-balance leak rate estimates of (Conley et al ), used here for comparison to inversion results. Inversions with an optimal setup are shown to provide estimates of the leak magnitude, on average, within a third of the mass balance values, with remaining errors in estimated leak rates predominantly explained by modeled wind speed errors of up to 10 m s−1, quantified by comparing airborne meteorological observations with modeled values along the flight track. An inversion setup using scaled observational wind speed errors in the model-data mismatch covariance matrix is shown to significantly reduce the influence of transport model errors on spatial patterns and estimated leak rates from the inversions. In sum, this study takes advantage of a natural tracer release experiment (i.e. the Aliso Canyon natural gas leak) to identify effective approaches for reducing the influence of transport model error on atmospheric inversions of point-source emissions, while suggesting future potential for integrating surface tower and aircraft atmospheric GHG observations in top-down urban emission monitoring systems.

Atmospheric inverse modeling via sparse reconstruction

Abstract. Many applications in atmospheric science involve ill-posed inverse problems. A crucial component of many inverse problems is the proper formulation of a priori knowledge about the unknown parameters. In most cases, this knowledge is expressed as a Gaussian prior. This formulation often performs well at capturing smoothed, large-scale processes but is often ill equipped to capture localized structures like large point sources or localized hot spots. Over the last decade, scientists from a diverse array of applied mathematics and engineering fields have developed sparse reconstruction techniques to identify localized structures. In this study, we present a new regularization approach for ill-posed inverse problems in atmospheric science. It is based on Tikhonov regularization with sparsity constraint and allows bounds on the parameters. We enforce sparsity using a dictionary representation system. We analyze its performance in an atmospheric inverse modeling scenario by estimating anthropogenic US methane (CH4) emissions from simulated atmospheric measurements. Different measures indicate that our sparse reconstruction approach is better able to capture large point sources or localized hot spots than other methods commonly used in atmospheric inversions. It captures the overall signal equally well but adds details on the grid scale. This feature can be of value for any inverse problem with point or spatially discrete sources. We show an example for source estimation of synthetic methane emissions from the Barnett shale formation.