Research ArticleCORONAVIRUS

Global tropospheric ozone responses to reduced NOx emissions linked to the COVID-19 worldwide lockdowns

See allHide authors and affiliations

Science Advances  09 Jun 2021:
Vol. 7, no. 24, eabf7460
DOI: 10.1126/sciadv.abf7460
  • Fig. 1 Schematic diagram of the methodology used in this study.

    (1) (A) The top-down 2010–2019 emissions obtained from the chemical data assimilation (green lines) were used to (B) evaluate relative temporal emission changes from the base date (February 1 and January 10 for China only) through July 31 each year. (C) The calculated relative temporal emission changes were averaged over the 10 years (2010–2019) to obtain climatological relative emission variations (solid blue line). (D) The climatological variations were applied to the 2020 emission (solid red line) values on the base date to obtain the BAU emissions for 2020 (solid blue line) and then compared with the 2020 emissions to estimate the COVID emission anomaly. (2) The COVID-19 ozone response through February to July 2020 and monthly ozone production efficiency (OPE) estimated from the beginning to end of each month were estimated from model simulations by replacing the BAU emissions with the 2020 emissions for each region or globally. TOB, tropospheric ozone burden. (3) The evaluated ozone response was compared with the observed changes from the Cross-Track Infrared Sounder (CrIS) satellite and surface observations.

  • Fig. 2 Spatial distributions of the monthly mean NOx emission reductions due to the COVID-19 lockdowns.

    The COVID NOx emission anomaly in February to June 2020 was estimated from differences between the 2020 and BAU emissions. Results are shown for the absolute changes in country total emissions (in TgN per year, top), relative changes in country total emissions (in %, middle), and absolute changes in grid-scale emissions (in 10−12kg Nm−2 s−1, bottom). The model grid points that were not analyzed because of unstable emission estimates and fire influences are shown in gray. N America, North America; S America, South America; ME + W Asia, Middle East and West Asia.

  • Fig. 3 Reductions in anthropogenic NOx emissions and global TOB.

    Monthly mean global and regional total changes in (A) NOx emissions (in TgN per year) due to the COVID-19 lockdowns and in (B) global TOB (in TgO3) due to regional NOx emission changes are shown for Africa, Europe, Australia, the Middle East and West Asia, the rest of Asia, South America, North America, China, and other regions.

  • Fig. 4 Time series of relative changes in country total NOx emissions (in %, black line) due to the COVID-19 lockdowns.

    The COVID-19 Government Response Stringency Index is shown by the dashed red line. The x axis represents days from 1 January 2020. The shaded area represents the 1σ uncertainty as measured from the SD of the BAU emissions.

  • Fig. 5 Monthly ozone changes due to the COVID NOx emission reductions in May 2020.

    Spatial distribution of the ozone anomaly (in ppb) at (A) the surface, (B) 500 hPa, and (C) zonal mean values in latitude-pressure coordinates. EQ, equator.

  • Fig. 6 Comparisons to ozone measurements from the CrIS satellite and surface networks.

    Time series of differences in monthly root mean square errors (RMSEs) (in ppb) of ozone against (A) the CrIS satellite retrievals at 700 hPa for the Northern Hemisphere (NH) (20°N to 90N) and Southern Hemisphere (SH) (90°S to 20S) and (B) the surface observations from the OpenAQ platform for Europe (light blue), the United States (blue), the Middle East (yellow), and China (red). The RMSE differences were estimated from two model simulations using the BAU and 2020 emissions, where the negative values show improved agreement against the observations using the 2020 emissions.

  • Fig. 7 Global map of the OPE.

    OPE is estimated from the change in the global TOB corresponding to the regional COVID NOx emission anomaly. The diameter of each circle represents the averaged OPE value during February to July 2020 in TgO3/TgN, while each sector of the circle represents the relative OPE magnitude for each month. The background map shows the defined areas used in this study.

  • Fig. 8 Monthly ozone changes observed from CrIS from 2019 to 2020.

    Global distributions of monthly mean ozone concentration difference between 2020 and 2019 (2020 minus 2019) observed from the CrIS satellite measurements at 700 hPa for March to June 2020. Negative values (blue) represent lower ozone concentrations in 2020 than in 2019.

  • Table 1 Monthly mean values of global and regional total surface NOx emission changes (in %) due to the COVID-19 restrictions.

    The 1-sigma uncertainties, estimated from the SD of the multiyear BAU emissions, are also shown. W Asia, West Asia; S America, South America; N America, North America.

    RegionFebruaryMarchAprilMayJune
    Globe−9.0 ± 1.5−12.7 ± 1.5−14.8 ± 2.3−15.0 ± 1.8−13.9 ± 1.8
    Africa−1.8 ± 3.7−2.1 ± 4.2−9.9 ± 4.4−10.3 ± 4.0−6.7 ± 4.1
    Europe−10.3 ± 4.1−16.5 ± 4.6−19.3 ± 5.8−18.7 ± 5.6−13.8 ± 3.6
    Australia−10.2 ± 4.0−12.8 ± 5.3−14.6 ± 5.7−15.7 ± 6.2−15.9 ± 7.4
    Middle East + W Asia−8.3 ± 4.8−14.8 ± 6.8−24.1 ± 9.7−24.8 ± 9.6−21.7 ± 10.6
    Rest of Asia−4.0 ± 1.3−7.4 ± 1.6−9.4 ± 2.6−10.6 ± 2.1−14.4 ± 2.1
    S America−3.3 ± 1.5−7.0 ± 1.8−10.2 ± 2.5−10.2 ± 2.4−10.3 ± 2.9
    N America−9.6 ± 2.6−16.1 ± 4.3−20.7 ± 6.2−20.1 ± 5.5−17.5 ± 4.6
    China−18.3 ±3.8−16.4 ± 3.1−6.2 ± 2.2−6.3 ± 2.4−6.9 ± 2.5
  • Table 2 Monthly values of the regional OPE (in TgO3/TgN).

    The OPE was estimated for the global TOB, using the regional COVID-19 NOx emission anomalies. The 1-sigma uncertainties, estimated from the SD (i.e., temporal changes) of the estimated TOB during the analysis period, are also shown.

    RegionFebruaryMarchAprilMayJune
    Africa2.15 ± 0.082.61 ± 0.231.51 ± 0.171.56 ± 0.081.53 ± 0.12
    Europe0.09 ± 0.010.13 ± 0.010.20 ± 0.060.23 ± 0.030.23 ± 0.03
    Australia2.68 ± 0.104.01 ± 0.103.16 ± 0.042.54 ± 0.052.40 ± 0.08
    Middle East + W Asia0.25 ± 0.010.40 ± 0.050.45 ± 0.050.47 ± 0.050.57 ± 0.05
    Rest of Asia1.11 ± 0.041.35 ± 0.031.54 ± 0.151.65 ± 0.071.44 ± 0.04
    S America3.65 ± 0.113.55 ± 0.092.75 ± 0.042.47 ± 0.042.21 ± 0.04
    N America0.23 ± 0.010.33 ± 0.040.45 ± 0.080.50 ± 0.060.45 ± 0.06
    China0.08 ± 0.000.17 ± 0.010.37 ± 0.050.44 ± 0.040.25 ± 0.02

Supplementary Materials

  • Supplementary Materials

    Global tropospheric ozone responses to reduced NOx emissions linked to the COVID-19 worldwide lockdowns

    Kazuyuki Miyazaki, Kevin Bowman, Takashi Sekiya, Masayuki Takigawa, Jessica L. Neu, Kengo Sudo, Greg Osterman, Henk Eskes

    Download Supplement

    The PDF file includes:

    • Supplementary Materials and Methods
    • Sections S1 to S4
    • Figs. S1 to S9
    • Table S1

    Other Supplementary Material for this manuscript includes the following:

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article