Modeling the air quality and public health benefits of increased residential insulation in the United States

Abstract

According to the Residential Energy Consumption Survey (RECS), homes in the United States consume approximately 10 quadrillion BTUs of energy each year, including electricity consumption for cooling, various fuels utilized for space heating, and other end uses. Electricity consumption will influence emissions from power plants, and these along with direct residential fuel combustion will also contribute to emissions of multiple key pollutants, with corresponding air quality and health impacts. We have developed models to quantify the energy savings associated with increased residential insulation and to estimate in monetary terms the environmental and public health benefits. We are considering both retrofits to existing housing and new construction, focusing on the 2012 International Energy Conservation Code (IECC), which specifies R-values and U-factors by climate zone and a number of other structural components and design specifications. We are applying EnergyPlus to a series of template files to estimate energy savings by fuel type and state, for both retrofits and new construction. To determine the emissions reductions related to reduced electricity generation, we used EPA’s AVERT tool. AVERT uses the basic attributes of electricity dispatch modeling to determine the power plants most likely influenced by energy efficiency programs, and provides the direct nitrogen oxide (NOx), sulfur dioxide (SO2), and carbon dioxide (CO2) emissions reductions on a plant-by-plant basis. For residential combustion, we used EPA’s AP-42 database and other resources to quantify direct emissions by fuel type (including natural gas, fuel oil, and wood). To model the health benefits of the criteria pollutant emissions, we linked the emissions reductions due to increased energy efficiency with the Community Multiscale Air Quality (CMAQ) model. We developed a series of simulations using CMAQ v4.7.1 instrumented with the Decoupled Direct Method (DDM), an advanced sensitivity analysis technique that allows us to estimate the influence of individual pollutants from individual sources or regions. We considered direct residential combustion by state, leveraging Census and housing start data to determine spatial patterns of emissions within states, and modeled individual power plants in geographic groupings using a design of experiments that allow us to estimate the impacts for all major power plants on the grid. We focused on fine particulate matter and ozone concentrations, as the key drivers of monetized health impacts. As CMAQ provides concentration estimates by grid cell, we are able to determine total public health benefits in terms of avoided mortality and morbidity as well as the distribution of those benefits for directly modeled facilities and locations. We estimate 19,000 premature deaths per year associated with EGU emissions, with more than half of the EGU-related health impacts attributable to emissions from seven states with significant coal combustion. We also estimate 10,000 premature deaths per year associated with residential combustion emissions, driven by primary PM2.5 emissions. In general, primary PM2.5 health damage functions are an order of magnitude larger than those of secondary PM2.5 precursors. Our findings reinforce the significance of source-specific assessment of air quality and health impacts for developing public health policies.