scholarly journals Towards better error statistics for atmospheric inversions of methane surface fluxes

2013 ◽  
Vol 13 (14) ◽  
pp. 7115-7132 ◽  
Author(s):  
A. Berchet ◽  
I. Pison ◽  
F. Chevallier ◽  
P. Bousquet ◽  
S. Conil ◽  
...  
Keyword(s):  
Expert Knowledge ◽  
Transport Model ◽  
Vertical Mixing ◽  
Real Data ◽  
Surface Fluxes ◽  
Error Statistics ◽  
The North ◽  
Error Covariance ◽  
Methane Fluxes ◽  
General Statistical

Abstract. We adapt general statistical methods to estimate the optimal error covariance matrices in a regional inversion system inferring methane surface emissions from atmospheric concentrations. Using a minimal set of physical hypotheses on the patterns of errors, we compute a guess of the error statistics that is optimal in regard to objective statistical criteria for the specific inversion system. With this very general approach applied to a real-data case, we recover sources of errors in the observations and in the prior state of the system that are consistent with expert knowledge while inferred from objective criteria and with affordable computation costs. By not assuming any specific error patterns, our results depict the variability and the inter-dependency of errors induced by complex factors such as the misrepresentation of the observations in the transport model or the inability of the model to reproduce well the situations of steep gradients of concentrations. Situations with probable significant biases (e.g., during the night when vertical mixing is ill-represented by the transport model) can also be diagnosed by our methods in order to point at necessary improvement in a model. By additionally analysing the sensitivity of the inversion to each observation, guidelines to enhance data selection in regional inversions are also proposed. We applied our method to a recent significant accidental methane release from an offshore platform in the North Sea and found methane fluxes of the same magnitude than what was officially declared.

scholarly journals Towards better error statistics for atmospheric inversions of methane surface fluxes

2013 ◽  
Vol 13 (2) ◽  
pp. 3735-3782
Author(s):  
A. Berchet ◽  
I. Pison ◽  
F. Chevallier ◽  
P. Bousquet ◽  
S. Conil ◽  
...  
Keyword(s):  
Expert Knowledge ◽  
Transport Model ◽  
Real Data ◽  
Surface Fluxes ◽  
Mass Composition ◽  
Error Statistics ◽  
The North ◽  
Error Covariance ◽  
Data Framework ◽  
General Statistical

Abstract. In this study, we adapt general statistical methods to compute the optimal error covariance matrices in a regional inversion system inferring methane surface emissions from atmospheric concentrations. We optimally estimate the error statistics with a minimal set of physical hypotheses on the patterns of errors. With this very general approach applied within a real-data framework, we recover sources of errors in the observations and in the prior state of the system that are consistent with expert knowledge. By not assuming any specific error patterns, our results show the variability and the inter-dependency of errors induced by complex factors such as the mis-representation of the observations in the transport model or the inability of the model to reproduce well the situations of steep gradients of air mass composition in the atmosphere. By analyzing the sensitivity of the inversion to each observation, ways to improve data selection in regional inversions are also proposed. We applied our method to a recent significant accidental methane release from an offshore platform in the North Sea.

scholarly journals A 4D-Var inversion system based on the icosahedral grid model (NICAM-TM 4D-Var v1.0) – Part 2: Optimization scheme and identical twin experiment of atmospheric CO<sub>2</sub> inversion

2017 ◽  
Vol 10 (6) ◽  
pp. 2201-2219 ◽  
Author(s):  
Yosuke Niwa ◽  
Yosuke Fujii ◽  
Yousuke Sawa ◽  
Yosuke Iida ◽  
Akihiko Ito ◽  
...  
Keyword(s):  
Transport Model ◽  
Surface Fluxes ◽  
Identical Twin ◽  
Inversion Problem ◽  
Model Errors ◽  
Optimization Scheme ◽  
Error Covariance ◽  
Optimal Setting ◽  
Source Sink ◽  
Quasi Newton

Abstract. A four-dimensional variational method (4D-Var) is a popular technique for source/sink inversions of atmospheric constituents, but it is not without problems. Using an icosahedral grid transport model and the 4D-Var method, a new atmospheric greenhouse gas (GHG) inversion system has been developed. The system combines offline forward and adjoint models with a quasi-Newton optimization scheme. The new approach is then used to conduct identical twin experiments to investigate optimal system settings for an atmospheric CO2 inversion problem, and to demonstrate the validity of the new inversion system. In this paper, the inversion problem is simplified by assuming the prior flux errors to be reasonably well known and by designing the prior error correlations with a simple function as a first step. It is found that a system of forward and adjoint models with smaller model errors but with nonlinearity has comparable optimization performance to that of another system that conserves linearity with an exact adjoint relationship. Furthermore, the effectiveness of the prior error correlations is demonstrated, as the global error is reduced by about 15 % by adding prior error correlations that are simply designed when 65 weekly flask sampling observations at ground-based stations are used. With the optimal setting, the new inversion system successfully reproduces the spatiotemporal variations of the surface fluxes, from regional (such as biomass burning) to global scales. The optimization algorithm introduced in the new system does not require decomposition of a matrix that establishes the correlation among the prior flux errors. This enables us to design the prior error covariance matrix more freely.

scholarly journals Development of CarbonTracker Europe-CH<sub>4</sub> – Part 2: global methane emission estimates and their evaluation for 2000–2012

2016 ◽  
Author(s):  
Aki Tsuruta ◽  
Tuula Aalto ◽  
Leif Backman ◽  
Janne Hakkarainen ◽  
Ingrid T. van der Laan-Luijkx ◽  
...  
Keyword(s):  
North American ◽  
Transport Model ◽  
Vertical Mixing ◽  
Total Carbon ◽  
In Situ Observation ◽  
Convection Scheme ◽  
Temperate Region ◽  
Northern Latitude ◽  
The North

Abstract. Gobal methane emissions were estimated for 2000–2012 using the CarbonTracker Europe-CH4 (CTE-CH4) data assimilation system. In CTE-CH4, the anthropogenic and biosphere emissions of CH4 are simultaneously constrained by global atmospheric in-situ methane mole fraction observations. We use three configurations developed in Tsuruta et al. (2016) to assess the sensitivity of the CH4 flux estimates to (a) the number of unknown flux scaling factors to be optimized which in turn depends on the choice of underlying land-ecosystem map, and (b) on the parametrization of vertical mixing in the 30 atmospheric transport model TM5. The posterior emission estimates were evaluated by comparing simulations to surface in-situ observation sites, to profile observations made by aircraft, to dry air total column-averaged mole fractions (XCH4) observations from the Total Carbon Column Observing Network (TCCON), and to XCH4 retrievals from the Greenhouse gases Observing SATellite (GOSAT). Our estimated posterior mean global total emissions during 2000–2012 are 516 ± 51 Tg CH4 yr−1, and emission estimates during 2007–2012 are 18 Tg CH4 yr−1 greater than those from 2001–2006, mainly driven by an 35 increase in emissions from the south America temperate region, the Asia temperate region and Asia tropics. The sensitivity of the flux estimates to the underlying ecosystem map was large for the Asia temperate region and Australia, but not significant in the northern latitude regions, i.e. the north American boreal region, the north American temperate region and Europe. Instead, the posterior estimates for the northern latitude regions show larger sensitivity to the choice of convection scheme in TM5. The Gregory et al. (2000) mixing scheme with faster interhemispheric exchange leads to higher estimated CH4 emissions at northern latitudes, and lower emissions in southern latitudes, compared to the estimates using Tiedtke (1989) convection scheme. Our evaluation with non-assimilated observations showed that posterior mole fractions were better matched with the 5 observations when Gregory et al. (2000) convection scheme was used.

scholarly journals Estimating regional methane surface fluxes: the relative importance of surface and GOSAT mole fraction measurements

2013 ◽  
Vol 13 (11) ◽  
pp. 5697-5713 ◽  
Author(s):  
A. Fraser ◽  
P. I. Palmer ◽  
L. Feng ◽  
H. Boesch ◽  
A. Cogan ◽  
...  
Keyword(s):  
Mole Fraction ◽  
Transport Model ◽  
Correlation Coefficients ◽  
Methane Flux ◽  
Surface Fluxes ◽  
Total Carbon ◽  
Tropical Regions ◽  
Spatial Coverage ◽  
Methane Fluxes ◽  
Atmospheric Sampling

Abstract. We use an ensemble Kalman filter (EnKF), together with the GEOS-Chem chemistry transport model, to estimate regional monthly methane (CH4) fluxes for the period June 2009–December 2010 using proxy dry-air column-averaged mole fractions of methane (XCH4) from GOSAT (Greenhouse gases Observing SATellite) and/or NOAA ESRL (Earth System Research Laboratory) and CSIRO GASLAB (Global Atmospheric Sampling Laboratory) CH4 surface mole fraction measurements. Global posterior estimates using GOSAT and/or surface measurements are between 510–516 Tg yr−1, which is less than, though within the uncertainty of, the prior global flux of 529 ± 25 Tg yr−1. We find larger differences between regional prior and posterior fluxes, with the largest changes in monthly emissions (75 Tg yr−1) occurring in Temperate Eurasia. In non-boreal regions the error reductions for inversions using the GOSAT data are at least three times larger (up to 45%) than if only surface data are assimilated, a reflection of the greater spatial coverage of GOSAT, with the two exceptions of latitudes >60° associated with a data filter and over Europe where the surface network adequately describes fluxes on our model spatial and temporal grid. We use CarbonTracker and GEOS-Chem XCO2 model output to investigate model error on quantifying proxy GOSAT XCH4 (involving model XCO2) and inferring methane flux estimates from surface mole fraction data and show similar resulting fluxes, with differences reflecting initial differences in the proxy value. Using a series of observing system simulation experiments (OSSEs) we characterize the posterior flux error introduced by non-uniform atmospheric sampling by GOSAT. We show that clear-sky measurements can theoretically reproduce fluxes within 10% of true values, with the exception of tropical regions where, due to a large seasonal cycle in the number of measurements because of clouds and aerosols, fluxes are within 15% of true fluxes. We evaluate our posterior methane fluxes by incorporating them into GEOS-Chem and sampling the model at the location and time of surface CH4 measurements from the AGAGE (Advanced Global Atmospheric Gases Experiment) network and column XCH4 measurements from TCCON (Total Carbon Column Observing Network). The posterior fluxes modestly improve the model agreement with AGAGE and TCCON data relative to prior fluxes, with the correlation coefficients (r2) increasing by a mean of 0.04 (range: −0.17 to 0.23) and the biases decreasing by a mean of 0.4 ppb (range: −8.9 to 8.4 ppb).

scholarly journals The potential of Orbiting Carbon Observatory-2 data to reduce the uncertainties in CO<sub>2</sub> surface fluxes over Australia using a variational assimilation scheme

2020 ◽  
Vol 20 (14) ◽  
pp. 8473-8500
Author(s):  
Yohanna Villalobos ◽  
Peter Rayner ◽  
Steven Thomas ◽  
Jeremy Silver
Keyword(s):  
Boundary Conditions ◽  
Dispersion Model ◽  
Grid Point ◽  
Surface Fluxes ◽  
Co2 Fluxes ◽  
Error Statistics ◽  
Error Covariance ◽  
Horizontal Grid ◽  
Assimilation Scheme ◽  
Total Column

Abstract. This paper addresses the question of how much uncertainties in CO2 fluxes over Australia can be reduced by assimilation of total-column carbon dioxide retrievals from the Orbiting Carbon Observatory-2 (OCO-2) satellite instrument. We apply a four-dimensional variational data assimilation system, based around the Community Multiscale Air Quality (CMAQ) transport-dispersion model. We ran a series of observing system simulation experiments to estimate posterior error statistics of optimized monthly-mean CO2 fluxes in Australia. Our assimilations were run with a horizontal grid resolution of 81 km using OCO-2 data for 2015. Based on four representative months, we find that the integrated flux uncertainty for Australia is reduced from 0.52 to 0.13 Pg C yr−1. Uncertainty reductions of up to 90 % were found at grid-point resolution over productive ecosystems. Our sensitivity experiments show that the choice of the correlation structure in the prior error covariance plays a large role in distributing information from the observations. We also found that biases in the observations would significantly impact the inverted fluxes and could contaminate the final results of the inversion. Biases in prior fluxes are generally removed by the inversion system. Biases in the boundary conditions have a significant impact on retrieved fluxes, but this can be mitigated by including boundary conditions in our retrieved parameters. In general, results from our idealized experiments suggest that flux inversions at this unusually fine scale will yield useful information on the carbon cycle at continental and finer scales.

scholarly journals A 4D-Var inversion system based on the icosahedral grid model (NICAM-TM 4D-Var v1.0): 2. Optimization scheme and identical twin experiment of atmospheric CO<sub>2</sub> inversion

2016 ◽  
Author(s):  
Yosuke Niwa ◽  
Yosuke Fujii ◽  
Yousuke Sawa ◽  
Yosuke Iida ◽  
Akihiko Ito ◽  
...  
Keyword(s):  
Transport Model ◽  
Global Scale ◽  
Surface Fluxes ◽  
Identical Twin ◽  
Model Errors ◽  
Optimization Scheme ◽  
New Approach ◽  
Error Covariance ◽  
Optimal Setting ◽  
Quasi Newton

Abstract. A 4-dimensional variational method (4D-Var) is a popular technique for inverse modeling of atmospheric constituents, but it is not without problems. Using an icosahedral grid transport model and the 4D-Var method, a new atmospheric greenhouse gas (GHG) inversion system has been developed. The system combines off-line forward and adjoint models with a quasi-Newton optimization scheme. The new approach is then used to conduct identical twin experiments to investigate optimal system settings for an atmospheric CO2 inversion problem, and to demonstrate the validity of the new inversion system. It is found that a system of forward and adjoint models that has less model errors but with non-linearity performs better than another system that conserves linearity with exact adjoint relationship. Furthermore, the effectiveness of the prior error correlations is confirmed; the global error is reduced by about 15 % by adding prior error correlations that are simply designed. With the optimal setting, the new inversion system successfully reproduces the spatiotemporal variations of the surface fluxes, from regional (such as biomass burning) to a global scale. The optimization algorithm introduced in the new system does not require difficult decomposition of a matrix that establishes the correlation among the prior flux errors. This enables us to design the prior error covariance matrix more freely.

scholarly journals Objectified quantification of uncertainties in Bayesian atmospheric inversions

2015 ◽  
Vol 8 (5) ◽  
pp. 1525-1546 ◽  
Author(s):  
A. Berchet ◽  
I. Pison ◽  
F. Chevallier ◽  
P. Bousquet ◽  
J.-L. Bonne ◽  
...  
Keyword(s):  
Expert Knowledge ◽  
Spatial Scales ◽  
Surface Fluxes ◽  
Large Set ◽  
Error Statistics ◽  
Observation Operator ◽  
Probability Of Occurrence ◽  
Error Distributions ◽  
The Impact ◽  
Atmospheric Inversions

Abstract. Classical Bayesian atmospheric inversions process atmospheric observations and prior emissions, the two being connected by an observation operator picturing mainly the atmospheric transport. These inversions rely on prescribed errors in the observations, the prior emissions and the observation operator. When data pieces are sparse, inversion results are very sensitive to the prescribed error distributions, which are not accurately known. The classical Bayesian framework experiences difficulties in quantifying the impact of mis-specified error distributions on the optimized fluxes. In order to cope with this issue, we rely on recent research results to enhance the classical Bayesian inversion framework through a marginalization on a large set of plausible errors that can be prescribed in the system. The marginalization consists in computing inversions for all possible error distributions weighted by the probability of occurrence of the error distributions. The posterior distribution of the fluxes calculated by the marginalization is not explicitly describable. As a consequence, we carry out a Monte Carlo sampling based on an approximation of the probability of occurrence of the error distributions. This approximation is deduced from the well-tested method of the maximum likelihood estimation. Thus, the marginalized inversion relies on an automatic objectified diagnosis of the error statistics, without any prior knowledge about the matrices. It robustly accounts for the uncertainties on the error distributions, contrary to what is classically done with frozen expert-knowledge error statistics. Some expert knowledge is still used in the method for the choice of an emission aggregation pattern and of a sampling protocol in order to reduce the computation cost. The relevance and the robustness of the method is tested on a case study: the inversion of methane surface fluxes at the mesoscale with virtual observations on a realistic network in Eurasia. Observing system simulation experiments are carried out with different transport patterns, flux distributions and total prior amounts of emitted methane. The method proves to consistently reproduce the known "truth" in most cases, with satisfactory tolerance intervals. Additionally, the method explicitly provides influence scores and posterior correlation matrices. An in-depth interpretation of the inversion results is then possible. The more objective quantification of the influence of the observations on the fluxes proposed here allows us to evaluate the impact of the observation network on the characterization of the surface fluxes. The explicit correlations between emission aggregates reveal the mis-separated regions, hence the typical temporal and spatial scales the inversion can analyse. These scales are consistent with the chosen aggregation patterns.

scholarly journals Arctic regional methane fluxes by ecotope as derived using eddy covariance from a low-flying aircraft

2017 ◽  
Vol 17 (13) ◽  
pp. 8619-8633 ◽  
Author(s):  
David S. Sayres ◽  
Ronald Dobosy ◽  
Claire Healy ◽  
Edward Dumas ◽  
John Kochendorfer ◽  
...  
Keyword(s):  
Eddy Covariance ◽  
Surface Flux ◽  
Methane Emissions ◽  
Surface Fluxes ◽  
The Arctic ◽  
Observation System ◽  
Permafrost Region ◽  
Aircraft Measurements ◽  
The North ◽  
Methane Fluxes

Abstract. The Arctic terrestrial and sub-sea permafrost region contains approximately 30 % of the global carbon stock, and therefore understanding Arctic methane emissions and how they might change with a changing climate is important for quantifying the global methane budget and understanding its growth in the atmosphere. Here we present measurements from a new in situ flux observation system designed for use on a small, low-flying aircraft that was deployed over the North Slope of Alaska during August 2013. The system combines a small methane instrument based on integrated cavity output spectroscopy (ICOS) with an air turbulence probe to calculate methane fluxes based on eddy covariance. We group surface fluxes by land class using a map based on LandSat Thematic Mapper (TM) data with 30 m resolution. We find that wet sedge areas dominate the methane fluxes with a mean flux of 2.1 µg m−2 s−1 during the first part of August. Methane emissions from the Sagavanirktok River have the second highest at almost 1 µg m−2 s−1. During the second half of August, after soil temperatures had cooled by 7 °C, methane emissions fell to between 0 and 0.5 µg m−2 s−1 for all areas measured. We compare the aircraft measurements with an eddy covariance flux tower located in a wet sedge area and show that the two measurements agree quantitatively when the footprints of both overlap. However, fluxes from sedge vary at times by a factor of 2 or more even within a few kilometers of the tower demonstrating the importance of making regional measurements to map out methane emissions spatial heterogeneity. Aircraft measurements of surface flux can play an important role in bridging the gap between ground-based measurements and regional measurements from remote sensing instruments and models.

scholarly journals Error characterization of CO<sub>2</sub> vertical mixing in the atmospheric transport model WRF-VPRM

2012 ◽  
Vol 12 (5) ◽  
pp. 2441-2458 ◽  
Author(s):  
R. Kretschmer ◽  
C. Gerbig ◽  
U. Karstens ◽  
F.-T. Koch
Keyword(s):  
Transport Model ◽  
Vertical Mixing ◽  
Atmospheric Model ◽  
Surface Fluxes ◽  
Model Errors ◽  
Simple Method ◽  
European Continent ◽  
Mixing Ratios ◽  
Error Characterization ◽  
The Impact

Abstract. One of the dominant uncertainties in inverse estimates of regional CO2 surface-atmosphere fluxes is related to model errors in vertical transport within the planetary boundary layer (PBL). In this study we present the results from a synthetic experiment using the atmospheric model WRF-VPRM to realistically simulate transport of CO2 for large parts of the European continent at 10 km spatial resolution. To elucidate the impact of vertical mixing error on modeled CO2 mixing ratios we simulated a month during the growing season (August 2006) with different commonly used parameterizations of the PBL (Mellor-Yamada-Janjić (MYJ) and Yonsei-University (YSU) scheme). To isolate the effect of transport errors we prescribed the same CO2 surface fluxes for both simulations. Differences in simulated CO2 mixing ratios (model bias) were on the order of 3 ppm during daytime with larger values at night. We present a simple method to reduce this bias by 70–80% when the true height of the mixed layer is known.

scholarly journals Using boundary layer equilibrium to reduce uncertainties in transport models and CO<sub>2</sub> flux inversions

2011 ◽  
Vol 11 (18) ◽  
pp. 9631-9641 ◽  
Author(s):  
I. N. Williams ◽  
W. J. Riley ◽  
M. S. Torn ◽  
J. A. Berry ◽  
S. C. Biraud
Keyword(s):  
Boundary Layer ◽  
Transport Model ◽  
Atmospheric Transport ◽  
Vertical Mixing ◽  
Systematic Errors ◽  
Surface Fluxes ◽  
Concentration Gradients ◽  
Layer Depth ◽  
Transport Models ◽  
Mixing Rates

Abstract. This paper reexamines evidence for systematic errors in atmospheric transport models, in terms of the diagnostics used to infer vertical mixing rates from models and observations. Different diagnostics support different conclusions about transport model errors that could imply either stronger or weaker northern terrestrial carbon sinks. Conventional mixing diagnostics are compared to analyzed vertical mixing rates using data from the US Southern Great Plains Atmospheric Radiation Measurement Climate Research Facility, the CarbonTracker data assimilation system based on Transport Model version 5 (TM5), and atmospheric reanalyses. The results demonstrate that diagnostics based on boundary layer depth and vertical concentration gradients do not always indicate the vertical mixing strength. Vertical mixing rates are anti-correlated with boundary layer depth at some sites, diminishing in summer when the boundary layer is deepest. Boundary layer equilibrium concepts predict an inverse proportionality between CO2 vertical gradients and vertical mixing strength, such that previously reported discrepancies between observations and models most likely reflect overestimated as opposed to underestimated vertical mixing. However, errors in seasonal concentration gradients can also result from errors in modeled surface fluxes. This study proposes using the timescale for approach to boundary layer equilibrium to diagnose vertical mixing independently of seasonal surface fluxes, with applications to observations and model simulations of CO2 or other conserved boundary layer tracers with surface sources and sinks. Results indicate that frequently cited discrepancies between observations and inverse estimates do not provide sufficient proof of systematic errors in atmospheric transport models. Some previously hypothesized transport model biases, if found and corrected, could cause inverse estimates to further diverge from carbon inventory estimates of terrestrial sinks.

Sign in / Sign up

Export Citation Format

Share Document