Unraveling pathways of elevated ozone induced by the 2020 lockdown in Europe by an observationally constrained regional model using TROPOMI

Questions about how emissions are changing during the COVID-19 lockdown periods cannot be answered by observations of atmospheric trace gas concentrations alone, in part due to simultaneous changes in atmospheric transport, emissions, dynamics, photochemistry, and chemical feedback. A chemical transport model simulation benefiting from a multi-species inversion framework using well-characterized observations should differentiate those influences enabling to closely examine changes in emissions. Accordingly, we jointly constrain NOx and VOC emissions using well-characterized TROPOspheric Monitoring Instrument (TROPOMI) HCHO and NO2 columns during the months of March, April, and May 2020 (lockdown) and 2019 (baseline). We observe a noticeable decline in the magnitude of NOx emissions in March 2020 (14 %–31 %) in several major cities including Paris, London, Madrid, and Milan, expanding further to Rome, Brussels, Frankfurt, Warsaw, Belgrade, Kyiv, and Moscow (34 %–51 %) in April. However, NOx emissions remain at somewhat similar values or even higher in some portions of the UK, Poland, and Moscow in March 2020 compared to the baseline, possibly due to the timeline of restrictions. Comparisons against surface monitoring stations indicate that the constrained model underrepresents the reduction in surface NO2. This underrepresentation correlates with the TROPOMI frequency impacted by cloudiness. During the month of April, when ample TROPOMI samples are present, the surface NO2 reductions occurring in polluted areas are described fairly well by the model (model: −21 ± 17 %, observation: −29 ± 21 %). The observational constraint on VOC emissions is found to be generally weak except for lower latitudes. Results support an increase in surface ozone during the lockdown. In April, the constrained model features a reasonable agreement with maximum daily 8 h average (MDA8) ozone changes observed at the surface (r = 0.43), specifically over central Europe where ozone enhancements prevail (model: +3.73 ± 3.94 %, + 1.79 ppbv, observation: +7.35 ± 11.27 %, +3.76 ppbv). The model suggests that physical processes (dry deposition, advection, and diffusion) decrease MDA8 surface ozone in the same month on average by −4.83 ppbv, while ozone production rates dampened by largely negative JNO2[NO2]-kNO+O3[NO][O3] become less negative, leading ozone to increase by +5.89 ppbv. Experiments involving fixed anthropogenic emissions suggest that meteorology contributes to 42 % enhancement in MDA8 surface ozone over the same region with the remaining part (58 %) coming from changes in anthropogenic emissions. Results illustrate the capability of satellite data of major ozone precursors to help atmospheric models capture ozone changes induced by abrupt emission anomalies.


S2. Anomaly of HCHO columns and top
updates to the surface classification climatology and cloud products that might have some effects on the magnitude of HCHO in cloudy scenes. We again remove bad pixels based on qa_flag < 0.75 and recalculate shape factors using the simulated profiles derived from our regional model.

Validation efforts reported in the sixth Quarterly Validation Report of the Copernicus
Sentinel-5 Precursor Operational Data Products [Lambert et al., 2020] indicate varying biases depending on the magnitude of HCHO concentrations in comparison to ground-based observations. Locations with HCHO concentrations above 8×10 15 molec/cm 2 show a low bias of ~-31%. Conversely, clean sites with HCHO concentrations below 2.5×10 15 molec/cm 2 undergo a high bias of 26%. Vigouroux et al. [2020] expanded the validation suite by including more than 25 FTIR stations located over both pristine and polluted sites. Results from the comparison with FTIR measurements (over clean areas) also indicate a high bias, whereas those compared in polluted areas show a low bias. By compiling numbers quoted in Lambert et al. [2020] and Vigouroux et al. [2020], we correct the existing biases in TROPOMI HCHO by scaling 25% (<2.5×10 15 molec/cm 2 ) down columns in clean areas and 30% (>=8×10 15 molec/cm 2 ) up in polluted areas. We assume the constant term of errors (econst) to be equal to 4% of HCHO total columns based on Vigouroux et al. [2020]. The precision error (eprecision) is populated with the column uncertainty variable provided with the data.
We investigate the changes in HCHO total columns shown in Figure S1. Various VOCs with different sources contribute to the formation of HCHO (see Figure 2 in Chan Miller et al. [2016]). In theory, it is easier to single out anthropogenic-derived HCHO concentration by HCHO measurements made in wintertime, although temperature and photochemistry are always key influencers of oxidizing/photolyzing all types of VOCs. The inevitable trade-off for this is dealing with a weaker signal that is near to instrument detection limit. The TROPOMI HCHO retrieval offers a low detection limit for individual pixels (7×10 15 molec/cm 2 ) that can be further lowered down by co-adding measurements (roughly a factor of 1/√n). Accordingly, we observe a promising signal in March over eastern European countries that is not explainable by biogenic emissions; but the magnitudes of the difference over these areas (<1.5×10 15 molec/cm 2 ) are below the detection limit (~ 2.4×10 15 molec/cm 2 given the co-added measurements over time).
In April, results show elevated HCHO concentrations in high latitudes in 2019 (box I), mainly a result of biomass burning activities in eastern Europe [e.g., Karlsson et al. 2013; https://earthobservatory.nasa.gov/global-maps/MOD14A1_M_FIRE, accessed June 2020]. As temperature rises in May, the footprint of biogenic emissions become more visible. This signal is not only induced by the inherent temperature-dependency of biogenic emissions, but also stems from faster isoprene oxidation through higher levels of OH [Pusede et al. 2015]. The dipole anomaly of HCHO columns suggested by TROPOMI (box J and K) pertains largely to variations in ambient surface air temperature (discussed later).

S2. Anomaly of HCHO columns and top-down VOC emissions
As to VOC emissions, we observe improvements in the magnitude and spatial distribution of simulated HCHO columns after the inversion with respect to TROPOMI data over areas with a practical amount of information (e.g., AK>0.2) ( Figure S15 and S16). Very low averaging kernels over major European cities in this month are indicative of inadequacies of one-month averaged TROPOMI HCHO data in March. The inversion partly corrects for the large underrepresentation of biomass burning emissions in high latitudes occurring in April 2019 but due to large uncertainties of the retrieval over this area, averaging kernels are low. Vigouroux et al. [2020] showed FTIR HCHO columns to be around 4-6×10 15 molec/cm 2 in Saint Peterburgh (59. along with solar radiation to be higher than the norm. This is primarily due to a well-developed high-pressure system over the region ( Figure S17) resulting in elevated HCHO columns. The topdown estimate is indicative of too low prior VOC emission rates over this area in April 2020.

However, the reason behind the enhancement of VOCs over several urban areas such Paris and Po
Valley is not fully understood. This can be caused by the errors in the chemical mechanism or the limited VOC compounds provided by the CEDS emission inventory. Given the significant role of VOCs in the formation of ozone in urban settings, this correction with reasonable AK (~0.4) is crucial for precisely modeling the surface ozone anomalies (shown in the manuscript). We revisit the pronounced dipole anomaly of dominantly biogenic VOC emissions in May. In this month, the biogenic VOCs dominate. Our model suggests that ambient surface temperature differences between Russian and central Europe are more than 7 o C, possibly inducing a strong dipole anomaly in biogenic emissions. It is readily evident from the averaging kernels that more realistic information from TROPOMI HCHO is attainable in warmer months, contrary to the NO2 case.