Quantifying Risks from Changing U.S. PM2.5 Distributions Due to Climate Variability and Warming with Large Multi-Model Ensembles and High-Resolution DownscalingEPA Grant Number: R835878
Title: Quantifying Risks from Changing U.S. PM2.5 Distributions Due to Climate Variability and Warming with Large Multi-Model Ensembles and High-Resolution Downscaling
Investigators: Fiore, Arlene M , West, Jason
Institution: Lamont Doherty Earth Observatory of Columbia University , University of North Carolina at Chapel Hill
EPA Project Officer: Chung, Serena
Project Period: January 1, 2016 through December 31, 2018 (Extended to December 31, 2020)
Project Amount: $788,625
RFA: Particulate Matter and Related Pollutants in a Changing World (2014) RFA Text | Recipients Lists
Research Category: Air , Climate Change
Description:With an unprecedented large ensemble of 21st century simulations from two global chemistry-climate models (CCMs) and high-resolution dynamical downscaling, we will assess air pollution risks due to climate warming and climate variability in the U.S.
We will quantify: (a) changing meteorological hazards conducive to air pollution episodes and their relationships with regional PM2.5, (b) impacts of climate variability versus warming on PM2.5, and (c) the time frame for emergence of an anthropogenic warming signal. We will examine regional vulnerabilities of PM2.5 to changing meteorological hazards, such as changing anthropogenic emissions and climate feedbacks from “natural” emissions (wildfires, biogenic secondary organic aerosol). For specific U.S. regions, we propose to estimate time-evolving regional PM2.5 risks (statistics relevant for policy and health impacts) from the changes in regional hazards and vulnerabilities (risk = hazard * vulnerability).
We propose to analyze daily PM2.5 and its co-occurrence with air pollution meteorology (e.g., air stagnation and precipitation events, ventilating storm passages) for 2005 to 2065 within several U.S. regions (e.g., NE, SE, SW, NW, Midwest, and Intermountain West), using a large ensemble (20-40 members) from the NCAR CCM with the RCP8.5 global change scenario, and simulations (3 members each) from the GFDL CCM with both RCP4.5 and RCP8.5 scenarios. GFDL CCM sensitivity simulations isolate the role of climate change from changing emissions of PM2.5 and precursors. Long (500+ years) control simulations enable us to establish baseline climate variability, and connections between PM2.5 and modes of variability (e.g., ENSO). Selected meteorological years for the 2050s will be dynamically downscaled at high-resolution (12 km) with the regional WRF and CMAQ models over the continental U.S.A. Within CMAQ, new sensitivity simulations will address vulnerabilities from feedbacks due to wildfires and biogenic secondary organic aerosol. By combining the downscaled PM2.5 and CCM distributions, we will develop probabilistic distributions of future PM2.5 visibility, and health impacts–and return periods for targeted PM2.5 levels– for several U.S. regions, and nation-wide gridded fields.
We will produce policy-relevant distributions for each of the next five decades that describe regional probabilities of future compliance with the national PM2.5 standard and regional haze rule, as well as health impacts in 2050, to inform decadal air quality planning. The distributions of time-evolving regional meteorological hazards may also inform U.S. planning in other areas (e.g., water resources). Our characterization of climate variability and change, and impacts on PM2.5, also contributes to building the knowledge base needed for seasonal-to-decadal prediction of PM2.5 and related meteorology.