Motor Vehicle Emissions Simulator Input Data

Abstract

Motor Vehicle Emissions Simulator (MOVES) input data submitted for the 2011 National Emissions Inventory for more than 1,400 U.S. counties were evaluated against the U.S. Environmental Protection Agency's best practice guidance and analyzed to determine the distributions of the data provided. A MOVES sensitivity analysis was then performed on the basis of the levels determined from the spread of the submitted data. The work found that the adherence to best practice guidance varied depending on the input and that the submitted data showed large variations for major inputs. Sensitivity analysis of MOVES emissions confirmed a strong influence of this variation on model predictions; MOVES total daily emissions varied up to 56% for hydrocarbons (HCs), 70% for carbon monoxide (CO), 111% for oxides of nitrogen (NOx), and 149% for particulate matter (PM) on the basis of changes in a single input within the range of data submitted by the states. The importance of each input and source type varied by pollutant; for HC and CO, which predominantly come from the gasoline-fueled light-duty vehicles, the age distribution and population of passenger cars and trucks were highly influential. For NOx and PM, a large variation in the fraction of vehicle miles traveled for combination trucks led to large differences in emissions. This work presents the first comparison of the relative influence of emissions by input and source type on the basis of data submitted by the states and will help MOVES users target areas in which focused data collection will lead to the most improvement in emissions inventory estimates. Overall, this work underscores the need for good local data in the development of regional emissions inventories.