scholarly journals Uncertainty analysis of a European high-resolution emission inventory of CO<sub>2</sub> and CO to support inverse modelling and network design

2020 ◽  
Vol 20 (3) ◽  
pp. 1795-1816 ◽  
Author(s):  
Ingrid Super ◽  
Stijn N. C. Dellaert ◽  
Antoon J. H. Visschedijk ◽  
Hugo A. C. Denier van der Gon
Keyword(s):  
High Resolution ◽  
Network Design ◽  
Emission Inventory ◽  
Atmospheric Transport ◽  
Temporal Distribution ◽  
Inverse Modelling ◽  
Climate Mitigation ◽  
Emission Inventories ◽  
Mixing Ratios ◽  
Modelling Framework

Abstract. Quantification of greenhouse gas emissions is receiving a lot of attention because of its relevance for climate mitigation. Complementary to official reported bottom-up emission inventories, quantification can be done with an inverse modelling framework, combining atmospheric transport models, prior gridded emission inventories and a network of atmospheric observations to optimize the emission inventories. An important aspect of such a method is a correct quantification of the uncertainties in all aspects of the modelling framework. The uncertainties in gridded emission inventories are, however, not systematically analysed. In this work, a statistically coherent method is used to quantify the uncertainties in a high-resolution gridded emission inventory of CO2 and CO for Europe. We perform a range of Monte Carlo simulations to determine the effect of uncertainties in different inventory components, including the spatial and temporal distribution, on the uncertainty in total emissions and the resulting atmospheric mixing ratios. We find that the uncertainties in the total emissions for the selected domain are 1 % for CO2 and 6 % for CO. Introducing spatial disaggregation causes a significant increase in the uncertainty of up to 40 % for CO2 and 70 % for CO for specific grid cells. Using gridded uncertainties, specific regions can be defined that have the largest uncertainty in emissions and are thus an interesting target for inverse modellers. However, the largest sectors are usually the best-constrained ones (low relative uncertainty), so the absolute uncertainty is the best indicator for this. With this knowledge, areas can be identified that are most sensitive to the largest emission uncertainties, which supports network design.

scholarly journals Uncertainty analysis of a European high-resolution emission inventory of CO<sub>2</sub> and CO to support inverse modelling and network design

2019 ◽  
Author(s):  
Ingrid Super ◽  
Stijn N. C. Dellaert ◽  
Antoon J. H. Visschedijk ◽  
Hugo A. C. Denier van der Gon
Keyword(s):  
High Resolution ◽  
Network Design ◽  
Emission Inventory ◽  
Atmospheric Transport ◽  
Temporal Distribution ◽  
Inverse Modelling ◽  
Climate Mitigation ◽  
Emission Inventories ◽  
Mixing Ratios ◽  
Modelling Framework

Abstract. Quantification of greenhouse gas emissions is receiving a lot of attention, because of its relevance for climate mitigation. Quantification is often done with an inverse modelling framework, combining atmospheric transport models, prior gridded emission inventories and a network of atmospheric observations to optimize the emission inventories. An important aspect of such method is a correct quantification of the uncertainties in all aspects of the modelling framework. The uncertainties in gridded emission inventories are, however, not systematically analysed. In this work, a statistically coherent method is used to quantify the uncertainties in a high-resolution gridded emission inventory of CO2 and CO for Europe. We perform a range of Monte Carlo simulations to determine the effect of uncertainties in different inventory components, including the spatial and temporal distribution, on the uncertainty in total emissions and the resulting atmospheric mixing ratios. We find that the uncertainty in the total emissions for the selected domain are 1 % for CO2 and 6 % for CO. Introducing spatial disaggregation causes a significant increase in the uncertainty of up to 40 % for CO2 and 70 % for CO for specific grid cells. Using gridded uncertainties specific regions can be defined that have the largest uncertainty in emissions and are thus an interesting target for inverse modelers. However, the largest sectors are usually the best-constrained ones (low relative uncertainty), so the absolute uncertainty is the best indicator for this. With this knowledge areas can be identified that are most sensitive to the largest emission uncertainties, which supports network design.

scholarly journals Inverse modelling of European N<sub>2</sub>O emissions: assimilating observations from different networks

2011 ◽  
Vol 11 (5) ◽  
pp. 2381-2398 ◽  
Author(s):  
M. Corazza ◽  
P. Bergamaschi ◽  
A. T. Vermeulen ◽  
T. Aalto ◽  
L. Haszpra ◽  
...  
Keyword(s):  
Atmospheric Transport ◽  
A Priori ◽  
N2o Emission ◽  
Inverse Modelling ◽  
N2o Emissions ◽  
Emission Inventories ◽  
Uncertainty Range ◽  
Spatial Gradients ◽  
Mixing Ratios ◽  
Modelling System

Abstract. We describe the setup and first results of an inverse modelling system for atmospheric N2O, based on a four-dimensional variational (4DVAR) technique and the atmospheric transport zoom model TM5. We focus in this study on the European domain, utilizing a comprehensive set of quasi-continuous measurements over Europe, complemented by N2O measurements from the Earth System Research Laboratory of the National Oceanic and Atmospheric Administration (NOAA/ESRL) cooperative global air sampling network. Despite ongoing measurement comparisons among networks parallel measurements at a limited number of stations show that significant offsets exist among the different laboratories. Since the spatial gradients of N2O mixing ratios are of the same order of magnitude as these biases, the direct use of these biased datasets would lead to significant errors in the derived emissions. Therefore, in order to also use measurements with unknown offsets, a new bias correction scheme has been implemented within the TM5-4DVAR inverse modelling system, thus allowing the simultaneous assimilation of observations from different networks. The N2O bias corrections determined in the TM5-4DVAR system agree within ~0.1 ppb (dry-air mole fraction) with the bias derived from the measurements at monitoring stations where parallel NOAA discrete air samples are available. The N2O emissions derived for the northwest European and east European countries for 2006 show good agreement with the bottom-up emission inventories reported to the United Nations Framework Convention on Climate Change (UNFCCC). Moreover, the inverse model can significantly narrow the uncertainty range reported in N2O emission inventories for these countries, while the lack of measurements does not allow to reduce the uncertainties of emission estimates in southern Europe. Several sensitivity experiments were performed to test the robustness of the results. It is shown that also inversions without detailed a priori spatio-temporal emission distributions are capable to reproduce major regional emission patterns within the footprint of the existing atmospheric network, demonstrating the strong constraints of the atmospheric observations on the derived emissions.

scholarly journals Inverse modelling of European N<sub>2</sub>O emissions: assimilating observations from different networks

2010 ◽  
Vol 10 (11) ◽  
pp. 26319-26359 ◽  
Author(s):  
M. Corazza ◽  
P. Bergamaschi ◽  
A. T. Vermeulen ◽  
T. Aalto ◽  
L. Haszpra ◽  
...  
Keyword(s):  
Atmospheric Transport ◽  
A Priori ◽  
N2o Emission ◽  
Inverse Modelling ◽  
N2o Emissions ◽  
Emission Inventories ◽  
Uncertainty Range ◽  
Spatial Gradients ◽  
Mixing Ratios ◽  
Modelling System

Abstract. We describe the setup and first results of an inverse modelling system for atmospheric N2O, based on a four-dimensional variational (4DVAR) technique and the atmospheric transport zoom model TM5. We focus in this study on the European domain, utilizing a comprehensive set of quasi-continuous measurements over Europe, complemented by N2O measurements from the Earth System Research Laboratory of the National Oceanic and Atmospheric Administration (NOAA/ESRL) cooperative global air sampling network. Despite ongoing measurement comparisons among networks parallel measurements at a limited number of stations show that significant offsets exist among the different laboratories. Since the spatial gradients of N2O mixing ratios are of the same order of magnitude as these biases, the direct use of these biased datasets would lead to significant errors in the derived emissions. Therefore, in order to also use measurements with unknown offsets, a new bias correction scheme has been implemented within the TM5-4DVAR inverse modelling system, thus allowing the simultaneous assimilation of observations from different networks. The N2O bias corrections determined in the TM5-4DVAR system agree within 0.1 ppb (dry-air mole fraction) with the bias derived from the measurements at monitoring stations where parallel NOAA discrete air samples are available. The N2O emissions derived for the northwest European countries for 2006 show good agreement with the bottom-up emission inventories reported to the United Nations Framework Convention on Climate Change (UNFCCC). Moreover, the inverse model can significantly narrow the uncertainty range reported in N2O emission inventories, while the lack of measurements does not allow for better emission estimates in southern Europe. Several sensitivity experiments were performed to test the robustness of the results. It is shown that also inversions without detailed a priori spatio-temporal emission distributions are capable to reproduce major regional emission patterns within the footprint of the existing atmospheric network, demonstrating the strong constraints of the atmospheric observations on the derived emissions.

scholarly journals The sensitivity of aerosol in Europe to two different emission inventories and temporal distribution of emissions

2006 ◽  
Vol 6 (12) ◽  
pp. 4287-4309 ◽  
Author(s):  
A. de Meij ◽  
M. Krol ◽  
F. Dentener ◽  
E. Vignati ◽  
C. Cuvelier ◽  
...  
Keyword(s):  
Emission Inventory ◽  
Dominant Role ◽  
Temporal Distribution ◽  
Temporal Variations ◽  
Anthropogenic Emissions ◽  
Emission Inventories ◽  
Global Transport ◽  
Modelling Uncertainties ◽  
The Vertical Distribution ◽  
Nitrate Aerosol

Abstract. The sensitivity to two different emission inventories, injection altitude and temporal variations of anthropogenic emissions in aerosol modelling is studied, using the two way nested global transport chemistry model TM5 focussing on Europe in June and December 2000. The simulations of gas and aerosol concentrations and aerosol optical depth (AOD) with the EMEP and AEROCOM emission inventories are compared with EMEP gas and aerosol surface based measurements, AERONET sun photometers retrievals and MODIS satellite data. For the aerosol precursor gases SO2 and NOx in both months the model results calculated with the EMEP inventory agree better (overestimated by a factor 1.3 for both SO2 and NOx) with the EMEP measurements than the simulation with the AEROCOM inventory (overestimated by a factor 2.4 and 1.9, respectively). Besides the differences in total emissions between the two inventories, an important role is also played by the vertical distribution of SO2 and NOx emissions in understanding the differences between the EMEP and AEROCOM inventories. In December NOx and SO2 from both simulations agree within 50% with observations. In June SO4= evaluated with the EMEP emission inventory agrees slightly better with surface observations than the AEROCOM simulation, whereas in December the use of both inventories results in an underestimate of SO4 with a factor 2. Nitrate aerosol measured in summer is not reliable, however in December nitrate aerosol calculations with the EMEP and AEROCOM emissions agree with 30%, and 60%, respectively with the filter measurements. Differences are caused by the total emissions and the temporal distribution of the aerosol precursor gases NOx and NH3. Despite these differences, we show that the column integrated AOD is less sensitive to the underlying emission inventories. Calculated AOD values with both emission inventories underestimate the observed AERONET AOD values by 20–30%, whereas a case study using MODIS data shows a high spatial agreement. Our evaluation of the role of temporal distribution of anthropogenic emissions on aerosol calculations shows that the daily and weekly temporal distributions of the emissions are only important for NOx, NH3 and aerosol nitrate. However, for all aerosol species SO4=, NH4+, POM, BC, as well as for AOD, the seasonal temporal variations used in the emission inventory are important. Our study shows the value of including at least seasonal information on anthropogenic emissions, although from a comparison with a range of measurements it is often difficult to firmly identify the superiority of specific emission inventories, since other modelling uncertainties, e.g. related to transport, aerosol removal, water uptake, and model resolution, play a dominant role.

scholarly journals Mapping pan-Arctic methane emissions at high spatial resolution using an adjoint atmospheric transport and inversion method and process-based wetland and lake biogeochemical models

2015 ◽  
Vol 15 (22) ◽  
pp. 32469-32518 ◽  
Author(s):  
Z. Tan ◽  
Q. Zhuang ◽  
D. K. Henze ◽  
C. Frankenberg ◽  
E. Dlugokencky ◽  
...  
Keyword(s):  
High Resolution ◽  
Atmospheric Transport ◽  
Methane Emissions ◽  
The Arctic ◽  
Inversion Method ◽  
Methane Cycle ◽  
Mixing Ratios ◽  
Biogeochemical Models ◽  
Northern Russia ◽  
And Inversion

Abstract. Understanding methane emissions from the Arctic, a fast warming carbon reservoir, is important for projecting changes in the global methane cycle under future climate scenarios. Here we optimize Arctic methane emissions with a nested-grid high-resolution inverse model by assimilating both high-precision surface measurements and column-average SCIAMACHY satellite retrievals of methane mole fraction. For the first time, methane emissions from lakes are integrated into an atmospheric transport and inversion estimate, together with prior wetland emissions estimated by six different biogeochemical models. We find that, the global methane emissions during July 2004–June 2005 ranged from 496.4 to 511.5 Tg yr−1, with wetland methane emissions ranging from 130.0 to 203.3 Tg yr−1. The Arctic methane emissions during July 2004–June 2005 were in the range of 14.6–30.4 Tg yr−1, with wetland and lake emissions ranging from 8.8 to 20.4 Tg yr−1 and from 5.4 to 7.9 Tg yr−1 respectively. Canadian and Siberian lakes contributed most of the estimated lake emissions. Due to insufficient measurements in the region, Arctic methane emissions are less constrained in northern Russia than in Alaska, northern Canada and Scandinavia. Comparison of different inversions indicates that the distribution of global and Arctic methane emissions is sensitive to prior wetland emissions. Evaluation with independent datasets shows that the global and Arctic inversions improve estimates of methane mixing ratios in boundary layer and free troposphere. The high-resolution inversions provide more details about the spatial distribution of methane emissions in the Arctic.

scholarly journals Designing optimal greenhouse gas monitoring networks for Australia

2016 ◽  
Vol 5 (1) ◽  
pp. 1-15 ◽  
Author(s):  
T. Ziehn ◽  
R. M. Law ◽  
P. J. Rayner ◽  
G. Roff
Keyword(s):  
Network Design ◽  
Greenhouse Gas ◽  
Atmospheric Transport ◽  
Dispersion Model ◽  
Network Size ◽  
Particle Dispersion ◽  
Data Uncertainty ◽  
Monitoring Networks ◽  
Modelling Framework ◽  
Reverse Mode

Abstract. Atmospheric transport inversion is commonly used to infer greenhouse gas (GHG) flux estimates from concentration measurements. The optimal location of ground-based observing stations that supply these measurements can be determined by network design. Here, we use a Lagrangian particle dispersion model (LPDM) in reverse mode together with a Bayesian inverse modelling framework to derive optimal GHG observing networks for Australia. This extends the network design for carbon dioxide (CO2) performed by Ziehn et al. (2014) to also minimise the uncertainty on the flux estimates for methane (CH4) and nitrous oxide (N2O), both individually and in a combined network using multiple objectives. Optimal networks are generated by adding up to five new stations to the base network, which is defined as two existing stations, Cape Grim and Gunn Point, in southern and northern Australia respectively. The individual networks for CO2, CH4 and N2O and the combined observing network show large similarities because the flux uncertainties for each GHG are dominated by regions of biologically productive land. There is little penalty, in terms of flux uncertainty reduction, for the combined network compared to individually designed networks. The location of the stations in the combined network is sensitive to variations in the assumed data uncertainty across locations. A simple assessment of economic costs has been included in our network design approach, considering both establishment and maintenance costs. Our results suggest that, while site logistics change the optimal network, there is only a small impact on the flux uncertainty reductions achieved with increasing network size.

Satellite-derived Indian methane emission sources with TROPOMI retrievals and a high-resolution modelling framework: Initial comparison with WRF-GHG model results

2020 ◽  
Author(s):  
Dhanyalekshmi Pillai ◽  
Monish Deshpande ◽  
Julia Marshall ◽  
Christoph Gerbig ◽  
Oliver Schneising ◽  
...  
Keyword(s):  
High Resolution ◽  
Greenhouse Gas ◽  
Methane Emission ◽  
Weather Prediction ◽  
Horizontal Resolution ◽  
International Standards ◽  
Inverse Modelling ◽  
Optical Absorption Spectroscopy ◽  
Retrieval Algorithm ◽  
Modelling Framework

&lt;p&gt;In accordance with the Global stocktake under Article 14 of Paris Agreement, each county estimates its own greenhouse gas (GHG) emissions based on standardised bottom-up management methods. However, the accuracy of these methods along with the standards differs from country to country, resulting in large uncertainties that make it difficult to implement effective climate change mitigation strategies. India plays an important role in global methane emission scenario, necessitating the accurate quantification of its sources at the regional and the local levels. However, the country lacks sufficient long term, continuous and accurate observations of the atmospheric methane which are required to quantify its source, to understand changes in the carbon cycle and the climate system. Recent technological advancements in the use of satellite remote-sensing dedicated to the greenhouse gases enforce international standards for the observation methods; hence enabling those high-resolution-high-density observations to be utilised for this quantification purpose. This study focuses on exploring the use of such dedicated observations of the column-averaged dry-air mixing ratio of methane (XCH&lt;sub&gt;4&lt;/sub&gt;) retrieved from TROPOMI onboard Sentinel-5 Precursor to quantify the major CH&lt;sub&gt;4&lt;/sub&gt; anthropogenic and natural emission fluxes over India.&lt;/p&gt;&lt;p&gt;Our inverse modelling approach at the mesoscale includes a high-resolution atmospheric modelling framework consisting of the Weather Research and Forecasting model with greenhouse gas module (WRF-GHG) and a set of prior emission inventory model data. We use TROPOMI retrievals derived using the Weighting Function Modified Differential Optical Absorption Spectroscopy (WFM-DOAS) retrieval algorithm. WRF-GHG simulations are performed in hourly time intervals at a horizontal resolution of&amp;#160;10 km &amp;#215;10 km for a month. In order to compare our CH&lt;sub&gt;4 &lt;/sub&gt;simulations with the satellite column data, we have also taken into account the different vertical sensitivities of the instrument by applying the averaging kernel to the model simulations. To demonstrate the model performance, our simulations are also compared with the CAMS reanalysis product based on ECMWF (European Centre for Medium-Range Weather Forecasts) numerical weather prediction reanalysis data available at a horizontal resolution of&amp;#160;0.25&lt;sup&gt;o&lt;/sup&gt; &amp;#215; 0.25&lt;sup&gt;o&lt;/sup&gt;. Our comparison of these modelling results against unique satellite dataset indicates high potential of using TROPOMI retrievals in distinguishing the major CH&lt;sub&gt;4&lt;/sub&gt; anthropogenic and natural sources over India via inverse modelling. The results will help to objectively investigate the claims of emission reductions and the efficiency of reduction countermeasures, as well as the establishment of standards and advancement of technology. The details about our approach and preliminary results based on our analysis using above satellite measurements and WRF-GHG simulations over India will be presented.&amp;#160;&amp;#160;&lt;/p&gt;&lt;p&gt;&amp;#160;&lt;/p&gt;

scholarly journals Designing optimal greenhouse gas monitoring networks for Australia

2015 ◽  
Vol 5 (2) ◽  
pp. 247-283
Author(s):  
T. Ziehn ◽  
R. M. Law ◽  
P. J. Rayner ◽  
G. Roff
Keyword(s):  
Network Design ◽  
Greenhouse Gas ◽  
Atmospheric Transport ◽  
Dispersion Model ◽  
Network Size ◽  
Particle Dispersion ◽  
Data Uncertainty ◽  
Monitoring Networks ◽  
Modelling Framework ◽  
Reverse Mode

Abstract. Atmospheric transport inversion is commonly used to infer greenhouse gas (GHG) flux estimates from concentration measurements. The optimal location of ground based observing stations that supply these measurements can be determined by network design. Here, we use a Lagrangian particle dispersion model (LPDM) in reverse mode together with a Bayesian inverse modelling framework to derive optimal GHG observing networks for Australia. This extends the network design for carbon dioxide (CO2) performed by Ziehn et al. (2014) to also minimize the uncertainty on the flux estimates for methane (CH4) and nitrous oxide (N2O), both individually and in a combined network using multiple objectives. Optimal networks are generated by adding up to 5 new stations to the base network, which is defined as two existing stations, Cape Grim and Gunn Point, in southern and northern Australia respectively. The individual networks for CO2, CH4 and N2O and the combined observing network show large similarities because the flux uncertainties for each GHG are dominated by regions of biologically productive land. There is little penalty, in terms of flux uncertainty reduction, for the combined network compared to individually designed networks. The location of the stations in the combined network is sensitive to variations in the assumed data uncertainty across locations. A simple assessment of economic costs has been included in our network design approach, considering both establishment and maintenance costs. Our results suggest that while site logistics change the optimal network, there is only a small impact on the flux uncertainty reductions achieved with increasing network size.

Sign in / Sign up

Export Citation Format

Share Document