Air Emissions Modeling

State-Level Hourly Sulfur Dioxide (SO2) Data

EPA’s final 1-hour SO2 National Ambient Air Quality Standard (NAAQS) provides the opportunity for States to use ambient air quality monitoring and/or emissions dispersion modeling to satisfy the requirements of the Data Requirements Rule. To assist states that choose to use emissions dispersion modelling, EPA is providing emissions data that is collected under the Part 75 continuous emission monitoring regulation. These data include stack characteristics and hourly emissions data provided by coal-fired sources that participate in the Acid Rain Program, Cross-State Air Pollution Rule, or former NOX Budget Trading Program. The data are primarily from electric generating units (EGUs) with some industrial boilers included. All of the sources collect emissions data using continuous emission monitoring systems (CEMS) and electronically report this information to the EPA on a quarterly basis. The data being provided cover calendar years 2012 through 2014.

A ZIP file for each State is provided below. Each ZIP file contains two Comma Separated Value (CSV) files. “Location Information” contains the characteristics of the facility (including the facility name, location, SO2 emission controls (if any), and characteristics of the stack (shape, height, construction, etc.). “Emissions Information” contains the hourly emissions data for each location described in the Location Information dataset. A data dictionary for these two files is in the fact sheet below.

Part 75 requires that facilities report emissions and operations data for every hour that an affected unit is operating (i.e., combusting fuel). If for some reason (e.g. CEMS malfunction) quality-assured data are not available for a particular operating hour, the Part 75 missing data routines must be used to provide substitute data for that hour. The longer the period of missing quality-assured data or the more frequently quality-assured data are not available, as measured by a CEMS’ percent monitor availability (PMA), the more conservative (i.e., higher) the substitute data. Substitute data in the dataset can be identified by the method of determination code (MODC). Substitute data will be “flagged” with a MODC of 06, 07, 08, 09, 10, 11, 12, 13, 15, 21, 23, or 55. Prior to modeling, States should determine which (if any) of the substituted data is appropriate to include in their dispersion modeling.

Sources that report under Part 75 are required to periodically perform RATAs to test their CEMS against a reference method. If the CEMS readings are found to be statistically biased low compared to the reference method, a bias adjustment factor (BAF) is calculated and applied to any subsequent hourly emissions data. BAFs are used for SO2 concentration and stack gas flow rate. The EPA-Provided data includes both the unadjusted, quality assured SO2 concentration for the hour, as measured by the SO2 monitor and the bias-adjusted SO2 concentration calculated by multiplying the SO2 unadjusted value by the bias adjustment factor (BAF) determined during the relative accuracy test audit (RATA). Prior to modeling, States should determine whether it is appropriate to use the unadjusted SO2 concentration or the bias adjusted SO2 concentration.