The first 1-year-long estimate of the Paris region fossil fuel CO 2 emissions based on atmospheric inversion

Johannes Staufer, Grégoire Broquet, François‐Marie Bréon, V. Puygrenier, Frédéric Chevallier, Irène Xueref-Rémy, Elsa Dieudonné, Morgan Lopez, Martina Schmidt, Michel Ramonet, Olivier Perrussel, Christine Lac, Lin Wu, Philippe CiaisAtmospheric 2016 · DOI

TL;DR

This paper presents the first year-long estimate of fossil fuel CO2 emissions for the Paris region using an atmospheric inversion technique. By carefully selecting atmospheric CO2 measurements from three monitoring stations, the researchers significantly reduced uncertainty in monthly emission estimates, finding results consistent with independent inventories and showing a realistic seasonal cycle correlated with heating demand.

What Problem Does This Paper Address?

Accurately quantifying urban CO2 emissions is crucial for climate policy and understanding regional carbon budgets. Traditional emission inventories, while valuable, often have uncertainties in their temporal and spatial resolution. Atmospheric inversion offers a complementary approach by using atmospheric CO2 measurements to constrain emission estimates. However, applying this to complex urban environments like Paris presents challenges, including accurately modeling atmospheric transport, accounting for biogenic fluxes, and dealing with the influence of emissions from surrounding areas. Previous work had provided initial estimates for Paris but only for a short period and with less stringent data selection. This study aims to extend this to a full year and refine the methodology for more robust monthly emission estimates.

Methods

The core of the study is a Bayesian atmospheric inversion framework, building upon previous work by Bréon et al. (2015). This method uses a transport model (CHIMERE) to simulate CO2 concentrations and compares these simulations to observed CO2 mole fraction gradients between pairs of monitoring sites located on the edges of the Paris urban area. The inversion then adjusts prior emission estimates to minimize the difference between simulated and observed gradients.

Key methodological advancements in this paper include:

  • Stringent Data Selection: A critical aspect is the rigorous filtering of CO2 gradient data. This involves selecting only mid-afternoon measurements when wind speeds exceed 3 m/s and the wind direction is tightly aligned with the transect between the monitoring sites (within ±15°). This strict selection, while discarding 92% of the data, is deemed necessary to overcome significant modeling uncertainties and ensure gradients are primarily influenced by urban emissions.
  • Revised Gradient Selection: The study discards gradients between two specific sites (GON-MON) that were found to be less representative of the entire urban area and more influenced by local sources like the airport.
  • One-Year Time Series: The inversion is applied to a full year of data (August 2010–July 2011), allowing for the analysis of seasonal variations and annual budgets.
  • Sensitivity Analyses: The authors perform extensive sensitivity tests to evaluate the impact of prior emission estimates, spatial distribution of emissions, and meteorological forcing on the inversion results.

Datasets

  • CO2 Mole Fraction Data: Hourly averaged CO2 measurements from three surface monitoring stations (MON, GON, GIF) operated as part of the CO2-MEGAPARIS experiment (August 2010–July 2011).
  • Fossil Fuel CO2 Emission Inventory: The AIRPARIF 2008 inventory for the Île-de-France region, used for prior emission estimates and spatial distribution. A revised AIRPARIF 2010 inventory was used for independent evaluation.
  • Biogenic CO2 Fluxes: Simulated by the C-TESSEL land surface model.
  • Meteorological Data: Operational analyses from ECMWF and simulations from Meso-NH/TEB were used for atmospheric transport modeling and gradient selection.

Key Insights and Ideas

  • The Power of Strict Data Filtering: The study strongly emphasizes that even with a drastic reduction in data (92% discarded), a stringent selection of CO2 gradients is essential for reliable inversion results in complex urban environments. This highlights the significant impact of modeling uncertainties on raw atmospheric data.
  • Gradients as a Constraint: Using CO2 gradients between monitoring sites is more effective than using individual site measurements for constraining urban emissions, as it helps to filter out the influence of remote and natural fluxes.
  • Seasonal Cycle Matters: The inverted emissions show a much more realistic seasonal cycle, peaking in winter, which strongly correlates with heating degrees. This suggests the inversion is capturing the dominant drivers of urban CO2 emissions.
  • Local Inventories are Crucial: The results demonstrate a significant sensitivity to the spatial distribution of emissions, underscoring the need for high-resolution local inventories like AIRPARIF.

Key Results

  • Uncertainty Reduction: The assimilation of CO2 gradients significantly reduced the uncertainty in monthly emission estimates, ranging from 9% in November 2010 to 50% in October 2010.
  • Annual Emission Estimate: The inverted annual mean emission for the Paris region was estimated at 40.9 MtCO2 yr⁻¹, which is close to the independent AIRPARIF 2010 inventory estimate of 41.8 MtCO2 yr⁻¹.
  • Seasonal Pattern: The inverted monthly emissions showed a clear seasonal cycle, with higher emissions in winter months (November-February), correlating well with heating degrees (r² = 0.45 for Nov-Feb).
  • Robustness to Sensitivity Tests: The reference inversion results were generally robust to variations in prior emission estimates and meteorological forcing, indicating the strength of the measurement constraint.

Limitations and Future Work

  • Data Scarcity: The limited number of monitoring sites and the strict data selection criteria result in a small number of assimilated data points for certain months, leading to poorer constraints for those periods (e.g., September, November 2010).
  • Diurnal Cycle Limitations: Assimilating only afternoon data limits the constraint on night-time emissions, which are important for understanding the full diurnal cycle.
  • Influence of Remote/Urban Fluxes: Despite efforts, the influence of remote CO2 fluxes and local urban emissions (like those from airports) can still impact results, especially when air masses originate from specific directions (e.g., NE).
  • Posterior Uncertainty Estimates: The authors note that posterior uncertainty estimates from the inversion system can sometimes appear unrealistically low, suggesting that the statistical assumptions (Gaussian distributions) may not fully capture the complexities.
  • Future Work:
    • Expand the observation network to encircle the city with more sites to improve data availability and gradient selection.
  • Develop new methods to exploit urban measurements for better spatial distribution estimates.
  • Develop techniques to incorporate morning, evening, and night-time data to better constrain night-time fluxes.
  • Further investigate the impact of urban ecosystems on gradients and explore urban ecosystem modeling.

Summary updated

Abstract

The ability of a Bayesian atmospheric inversion to quantify the Paris region's fossil fuel CO2 emissions on a monthly basis, based on a network of three surface stations operated for 1 year as part of the CO2-MEGAPARIS experiment (August 2010–July 2011), is analysed. Differences in hourly CO2 atmospheric mole fractions between the near-ground monitoring sites (CO2 gradients), located at the north-eastern and south-western edges of the urban area, are used to estimate the 6 h mean fossil fuel CO2 emission. The inversion relies on the CHIMERE transport model run at 2 km × 2 km horizontal resolution, on the spatial distribution of fossil fuel CO2 emissions in 2008 from a local inventory established at 1 km × 1 km horizontal resolution by the AIRPARIF air quality agency, and on the spatial distribution of the biogenic CO2 fluxes from the C-TESSEL land surface model. It corrects a prior estimate of the 6 h mean budgets of the fossil fuel CO2 emissions given by the AIRPARIF 2008 inventory. We found that a stringent selection of CO2 gradients is necessary for reliable inversion results, due to large modelling uncertainties. In particular, the most robust data selection analysed in this study uses only mid-afternoon gradients if wind speeds are larger than 3 m s−1 and if the modelled wind at the upwind site is within ±15° of the transect between downwind and upwind sites. This stringent data selection removes 92 % of the hourly observations. Even though this leaves few remaining data to constrain the emissions, the inversion system diagnoses that their assimilation significantly reduces the uncertainty in monthly emissions: by 9 % in November 2010 to 50 % in October 2010. The inverted monthly mean emissions correlate well with independent monthly mean air temperature. Furthermore, the inverted annual mean emission is consistent with the independent revision of the AIRPARIF inventory for the year 2010, which better corresponds to the measurement period than the 2008 inventory. Several tests of the inversion's sensitivity to prior emission estimates, to the assumed spatial distribution of the emissions, and to the atmospheric transport modelling demonstrate the robustness of the measurement constraint on inverted fossil fuel CO2 emissions. The results, however, show significant sensitivity to the description of the emissions' spatial distribution in the inversion system, demonstrating the need to rely on high-resolution local inventories such as that from AIRPARIF. Although the inversion constrains emissions through the assimilation of CO2 gradients, the results are hampered by the improperly modelled influence of remote CO2 fluxes when air masses originate from urbanised and industrialised areas north-east of Paris. The drastic data selection used in this study limits the ability to continuously monitor Paris fossil fuel CO2 emissions: the inversion results for specific months such as September or November 2010 are poorly constrained by too few CO2 measurements. The high sensitivity of the inverted emissions to the prior emissions' diurnal variations highlights the limitations induced by assimilating data only during the afternoon. Furthermore, even though the inversion improves the seasonal variation and the annual budget of the city's emissions, the assimilation of data during a limited number of suitable days does not necessarily yield robust estimates for individual months. These limitations could be overcome through a refinement of the data processing for a wider data selection, and through the expansion of the observation network.
BibTeX
@inproceedings{staufer2016first,
  author = {Johannes Staufer and Grégoire Broquet and François‐Marie Bréon and V. Puygrenier and Frédéric Chevallier and Irène Xueref-Rémy and Elsa Dieudonné and Morgan Lopez and Martina Schmidt and Michel Ramonet and Olivier Perrussel and Christine Lac and Lin Wu and Philippe Ciais},
  title = {{The first 1-year-long estimate of the Paris region fossil fuel CO <sub>2</sub> emissions based on atmospheric inversion}},
  booktitle = {Atmospheric},
  year = {2016},
  doi = {10.5194/acp-16-14703-2016},
}
Figure