Data for: Projecting changes in the frequency and magnitude of ozone pollution events under uncertain climate sensitivity

East, James D.1 ; Monier, Erwan2; Saari, Rebecca K.3; Garcia-Menendez, Fernando 1

Published Apr 23, 2024 on Dryad. https://doi.org/10.5061/dryad.nk98sf7z2

Data files

Apr 23, 2024 version files 440.41 MB

East_et_al_2024_data.tar.gz

440.40 MB
README.md

4.72 KB

Abstract

Climate change is projected to worsen ozone pollution over many populated regions, with larger impacts at higher concentrations. More intense and frequent ozone episodes risk setbacks to human health and environmental policy achievements. However, assessing these changes is complicated by uncertain climate sensitivity, closely related to climate model response, and internal variability in simulations projecting climate’s influence on air quality. Here, leveraging a global modeling framework that one-way couples a human activity model, an Earth system model of intermediate complexity, and an atmospheric chemistry model, we investigate the role of climate sensitivity in climate-induced changes to high ozone pollution episodes in the United States using multiple greenhouse gas emissions scenarios, representations of climate sensitivity, and initial condition members. We bias correct and evaluate historical model simulations, identifying modeled and observed O₃ episodes using extreme value theory, and extend the approach to projections of mid- and end-century climate impacts. Results show that the influence of climate sensitivity can be as significant as that of greenhouse gas emissions scenario absent precursor emissions changes. Climate change is projected to increase the magnitude of the highest annually occurring O₃ concentrations by over 2.3 ppb on average across the U.S. at mid-century under a high climate sensitivity and moderate emissions scenario, but the increase is limited to less than 0.3 ppb under lower climate sensitivity. Further, we show that areas in the U.S. currently meeting air quality standards risk being pushed into non-compliance due to a climate-induced increase in frequency of high ozone days.

Contact: f_garcia@ncsu.edu

Data included in this folder are described below

Description of the data and file structure

GRIDDED DATA

Files are provided in NetCDF format on a global grid, with data masked outside the U.S. The following types of data are included:

Bias-corrected MDA8 ozone concentrations (folder camchem_adj)
- Annual files of daily MDA8 ozone-season concentrations are included for each available climate sensitivity, emissions scenario, and initial condition ensemble member at beginning, middle, and end-century. Each file contains 153 days of gridded ozone-season bias-corrected concentrations.
- File name convention: POL.CS.IC.YYYY-YYYY.yyyy.AQ.nc where POL is the emissions scenario (REF, P45, or P37), CS is the climate sensitivity scenario (CS45, CS30, CS20), YYYY-YYYY is the time span indicating beginning, middle, or end-century (1981-2010, 2036-2065, 2086-2115), and yyyy is the specific year for that file.
Ozone 1-year return levels (folder retvals)
- Probabilistic 1-year return levels are included for each available climate sensitivity, emissions scenario, and initial condition ensemble member at beginning, middle, and end-century. Each file contains gridded 1-year ozone return level (units of ppb) and gridded values for Kolmogorov-Smirnov (K-S) test results.
- File name convention: POL.CS.IC.YYYY-YYYY.O3X.1YR.nc where POL is the emissions scenario (REF, P45, or P37), CS is the climate sensitivity scenario (CS45, CS30, CS20), and YYYY-YYYY is the time span indicating beginning, middle, or end-century (1981-2010, 2036-2065, 2086-2115).
Ozone 70-ppb return intervals (folder retperiods)
- Probabilistic return intervals for 70-ppb MDA8 ozone are included for each available climate sensitivity, emissions scenario, and initial condition ensemble member at beginning, middle, and end-century. Each file contains gridded 1-year ozone return level (units of ppb) and gridded values for Kolmogorov-Smirnov (K-S) test results.
- File name convention: POL.CS.IC.YYYY-YYYY.O3X.1YR.nc where POL is the emissions scenario (REF, P45, or P37), CS is the climate sensitivity scenario (CS45, CS30, CS20), and YYYY-YYYY is the time span indicating beginning, middle, or end-century (1981-2010, 2036-2065, 2086-2115).

RETURN VALUES

Probabilistic return values calculated from observed and modeled concentrations at CASTNET sites (folder return_values_at_sites)
- CSV files for 1, 2, 5, 10, 25-year return values at CASTNET sites and for each IC member in the historical simulation. Data columns included are site (CASTNET site code), return period (units of years), return value (units of ppb), lower ci (return value confidence interval lower bound, units of ppb), upper ci (return value confidence interval upper bound, units of ppb), nextremes (count of MDA8 ozone exceedances of 70 ppb threshold)
Probabilistic return values calculated from observed and modeled concentrations at CASTNET sites, for the Pre-SIP period only (folder return_values_at_sites_preSIP)
- CSV files for 1, 2, 5, 10, 25-year return values at CASTNET sites and for each IC member in the historical simulation. Data columns included are site (CASTNET site code), return period (units of years), return value (units of ppb), lower ci (return value confidence interval lower bound, units of ppb), upper ci (return value confidence interval upper bound, units of ppb), nextremes (count of MDA8 ozone exceedances of 70 ppb threshold)
Probabilistic return values calculated from observed and modeled concentrations at CASTNET sites, for the uncorrected model concentrations (folder return_values_at_sites_uncorrected)
- CSV files for 1, 2, 5, 10, 25-year return values at CASTNET sites and for each IC member in the historical simulation. Data columns included are site (CASTNET site code), return period (units of years), return value (units of ppb), lower ci (return value confidence interval lower bound, units of ppb), upper ci (return value confidence interval upper bound, units of ppb), nextremes (count of MDA8 ozone exceedances of 70 ppb threshold)
Null values ("NaN") in the return value csv datasets mean the specified return value at that site cannot be calculated because there are too few values above the exceedance threshold of 70ppb.

Sharing/Access information

CASTNET data

CASTNET data used in this study are publicly available from the U.S. Environmental Protection Agency at https://www.epa.gov/castnet