ABSTRACT
In recent years, the United States power sector emissions of CO2 and NOx have decreased due to declining coal and increasing natural gas and renewables in the fuel-mix. In April 2020, the COVID-19 social restrictions in the United States led to a decline in electricity demand from the commercial and industrial sectors. In this study, we estimate the changes in the emissions of CO2 and NOx from the U.S. power sector due to three factors: 1) weather, 2) the fuel-mix change in the past five years, and 3) the COVID-19 social restrictions. We use a multivariate adaptive regression splines (MARS) model to separate the impacts of outdoor temperature and type-of-day from the COVID-19 on power generation, and the daily operation status of 3013 power units to account for the fuel-mix change. We find that electricity demand changes due to COVID occurred mostly from March to June 2020, with electricity demand generally returning to 2015–2019 levels starting in July 2020. We find the U.S. power sector CO2 emissions, reported by EPA, dropped by 29.8 MTCO2 (−26%) in April 2020, relative to the average April emissions between 2015 and 2019. Of that reduction, we attribute declines of 18.3 ± 4.0 MTCO2 (−18 ± 4%) to the COVID-19 lockdowns, declines of 13.7 ± 4.2 MTCO2 (−12 ± 4%) to a fuel-mix change, and increases of 2.3 ± 1.1 MTCO2 (+2 ± 1%) due to weather variability compared to the five prior years. For the same month, the power sector NOx emissions dropped by 27.6 thousand metric tons (−42%) in April 2020, relative to the past five-year monthly average. Of that reduction, we attribute declines of 10.5 ± 2.4 thousand metric tons (−22 ± 5%) to the COVID-19 lockdowns, declines of 18.5 ± 2.5 thousand metric tons (−28 ± 4%) to a fuel-mix change, and increases of 1.4 ± 0.6 thousand metric tons (+2 ± 1%) due to weather variability. This result highlights the importance of accounting for weather and fuel-mix changes when estimating the impact of COVID-19 on the power sector emissions.
ABSTRACT
The worldwide lockdown in response to the COVID-19 pandemic in year 2020 led to an economic slowdown and a large reduction in fossil fuel CO2 emissions (Le Quéré 2020 Nat. Clim. Change 10 647-53, Liu 2020 Nat. Commun. 11);however, it is unclear how much it would slow the increasing trend of atmospheric CO2 concentration, the main driver of climate change, and whether this impact can be observed considering the large biosphere and weather variabilities. We used a state-of-the-art atmospheric transport model to simulate CO2, and the model was driven by a new daily fossil fuel emissions dataset and hourly biospheric fluxes from a carbon cycle model forced with observed climate variability. Our results show a 0.21 ppm decrease in the atmospheric column CO2 anomaly in the Northern Hemisphere latitude band 0-45 N in March 2020, and an average of 0.14 ppm for the period of February-April 2020, which is the largest decrease in the last 10 years. A similar decrease was observed by the carbon observing satellite GOSAT (Yokota et al 2009 Sola 5 160-3). Using model sensitivity experiments, we further found that the COVID and weather variability are the major contributors to this CO2 drawdown, and the biosphere showed a small positive anomaly. Measurements at marine boundary layer stations, such as Hawaii, exhibit 1-2 ppm anomalies, mostly due to weather and the biosphere. At the city scale, the on-road CO2 enhancement measured in Beijing shows a reduction by 20-30 ppm, which is consistent with the drastically reduced traffic during the COVID lockdown. A stepwise drop of 20 ppm during the city-wide lockdown was observed in the city of Chengdu. The ability of our current carbon monitoring systems in detecting the small and short-lasting COVID signals at different policy relevant scales (country and city) against the background of fossil fuel CO2 accumulated over the last two centuries is encouraging. The COVID-19 pandemic is an unintended experiment. Its impact suggests that to keep atmospheric CO2 at a climate-safe level will require sustained effort of similar magnitude and improved accuracy, as well as expanded spatiotemporal coverage of our monitoring systems. © 2021 The Author(s). Published by IOP Publishing Ltd.