A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Uncertainty analysis of modeled ozone changes due to anthropogenic emission reductions in Eastern Texas
https://orcid.org/0000-0002-8508-1422
Abstract We investigate whether an air quality model can predict a concentration change between two related scenarios with greater certainty than either scenario alone and also estimate the uncertainty in Relative Reduction Factors (RRFs). RRFs are ratios of future-year to base-year model results that are used by the U.S. Environmental Protection Agency and the States to predict future ozone (O3) and thus are critical to developing emission control strategies to reduce O3 and achieve compliance with the O3 standard. We conducted simulations for 2012 and 2020 using the Comprehensive Air Quality Model with Extensions and calculated the uncertainties for O3, O3 change (ΔO3), and RRFs using O3 sensitivities. These uncertainties were derived from estimated uncertainties in an extensive set of model inputs comprising the boundary concentrations, dry deposition velocities, emissions, and chemistry. We developed new equations to determine the uncertainties in ΔO3 and RRFs. We evaluated the RRFs obtained from our model by comparing them to recent ozone trends in Eastern Texas (the trends omit 2020 due to the effects of the pandemic.) Uncertainties in ΔO3 and RRFs depend strongly on the degree of correlation between errors in the anthropogenic emissions in the two years. We considered Case A, full correlation of the errors, and Case B, no correlation of errors, as limiting cases. The uncertainty in ΔO3 in Case A is 10–20% of the uncertainty in O3 over most of eastern Texas, and in Case B it is 45–75%. For monitoring sites in three Texas cities, the uncertainty in the RRFs is 0.2–1.0% of the RRFs (Case A) or 4.8–8.5% (Case B). Using ΔO3 or RRFs with base-year design values to predict O3 is likely to produce a more accurate prediction than the modeled O3 for the future year regardless of whether Case A or Case B holds. However, determining the degree of correlation of emission errors between years is necessary to refine further the uncertainty in predicted O3 and merits future study.
Graphical abstract Display Omitted
Highlights Modeled ozone changes have smaller uncertainties than modeled ozone concentrations. Relative Reduction Factors reduce the uncertainty in estimated future ozone. Correlation of emission uncertainties between years strongly affects the results. Uncertainties are estimated from sensitivities and alternative chemical mechanisms.
Uncertainty analysis of modeled ozone changes due to anthropogenic emission reductions in Eastern Texas
https://orcid.org/0000-0002-8508-1422
Abstract We investigate whether an air quality model can predict a concentration change between two related scenarios with greater certainty than either scenario alone and also estimate the uncertainty in Relative Reduction Factors (RRFs). RRFs are ratios of future-year to base-year model results that are used by the U.S. Environmental Protection Agency and the States to predict future ozone (O3) and thus are critical to developing emission control strategies to reduce O3 and achieve compliance with the O3 standard. We conducted simulations for 2012 and 2020 using the Comprehensive Air Quality Model with Extensions and calculated the uncertainties for O3, O3 change (ΔO3), and RRFs using O3 sensitivities. These uncertainties were derived from estimated uncertainties in an extensive set of model inputs comprising the boundary concentrations, dry deposition velocities, emissions, and chemistry. We developed new equations to determine the uncertainties in ΔO3 and RRFs. We evaluated the RRFs obtained from our model by comparing them to recent ozone trends in Eastern Texas (the trends omit 2020 due to the effects of the pandemic.) Uncertainties in ΔO3 and RRFs depend strongly on the degree of correlation between errors in the anthropogenic emissions in the two years. We considered Case A, full correlation of the errors, and Case B, no correlation of errors, as limiting cases. The uncertainty in ΔO3 in Case A is 10–20% of the uncertainty in O3 over most of eastern Texas, and in Case B it is 45–75%. For monitoring sites in three Texas cities, the uncertainty in the RRFs is 0.2–1.0% of the RRFs (Case A) or 4.8–8.5% (Case B). Using ΔO3 or RRFs with base-year design values to predict O3 is likely to produce a more accurate prediction than the modeled O3 for the future year regardless of whether Case A or Case B holds. However, determining the degree of correlation of emission errors between years is necessary to refine further the uncertainty in predicted O3 and merits future study.
Graphical abstract Display Omitted
Highlights Modeled ozone changes have smaller uncertainties than modeled ozone concentrations. Relative Reduction Factors reduce the uncertainty in estimated future ozone. Correlation of emission uncertainties between years strongly affects the results. Uncertainties are estimated from sensitivities and alternative chemical mechanisms.
Uncertainty analysis of modeled ozone changes due to anthropogenic emission reductions in Eastern Texas
https://orcid.org/0000-0002-8508-1422
Abstract We investigate whether an air quality model can predict a concentration change between two related scenarios with greater certainty than either scenario alone and also estimate the uncertainty in Relative Reduction Factors (RRFs). RRFs are ratios of future-year to base-year model results that are used by the U.S. Environmental Protection Agency and the States to predict future ozone (O3) and thus are critical to developing emission control strategies to reduce O3 and achieve compliance with the O3 standard. We conducted simulations for 2012 and 2020 using the Comprehensive Air Quality Model with Extensions and calculated the uncertainties for O3, O3 change (ΔO3), and RRFs using O3 sensitivities. These uncertainties were derived from estimated uncertainties in an extensive set of model inputs comprising the boundary concentrations, dry deposition velocities, emissions, and chemistry. We developed new equations to determine the uncertainties in ΔO3 and RRFs. We evaluated the RRFs obtained from our model by comparing them to recent ozone trends in Eastern Texas (the trends omit 2020 due to the effects of the pandemic.) Uncertainties in ΔO3 and RRFs depend strongly on the degree of correlation between errors in the anthropogenic emissions in the two years. We considered Case A, full correlation of the errors, and Case B, no correlation of errors, as limiting cases. The uncertainty in ΔO3 in Case A is 10–20% of the uncertainty in O3 over most of eastern Texas, and in Case B it is 45–75%. For monitoring sites in three Texas cities, the uncertainty in the RRFs is 0.2–1.0% of the RRFs (Case A) or 4.8–8.5% (Case B). Using ΔO3 or RRFs with base-year design values to predict O3 is likely to produce a more accurate prediction than the modeled O3 for the future year regardless of whether Case A or Case B holds. However, determining the degree of correlation of emission errors between years is necessary to refine further the uncertainty in predicted O3 and merits future study.
Graphical abstract Display Omitted
Highlights Modeled ozone changes have smaller uncertainties than modeled ozone concentrations. Relative Reduction Factors reduce the uncertainty in estimated future ozone. Correlation of emission uncertainties between years strongly affects the results. Uncertainties are estimated from sensitivities and alternative chemical mechanisms.
Uncertainty analysis of modeled ozone changes due to anthropogenic emission reductions in Eastern Texas
Dunker, Alan M. (author) / Nopmongcol, Uarporn (author) / Yarwood, Greg (author)
Atmospheric Environment ; 268
2021-10-17
Article (Journal)
Electronic Resource
English