2.9. Inventory import

2.9.1. Check the correctness and consistency of input file formats
2.9.2. Assign pollutant names to data input by code numbers
2.9.3. Select pollutants from the input files for further SMOKE processing
2.9.4. Check for duplicate records
2.9.5. Combine toxics and criteria inventories
2.9.6. Sort the inventory
2.9.7. Aggregate or disaggregate toxics emissions
2.9.8. Assign point-source locations to area sources
2.9.9. Fill in and check point-source stack parameters
2.9.10. Convert coordinates from UTM to lat-lon
2.9.11. Optionally ensure that lat-lon coordinates are in the Western Hemisphere
2.9.12. Convert units of emissions and activities
2.9.13. Set the weekday averaging approach
2.9.14. Assign country codes, years, and time zones
2.9.15. Handle inventories that have data for multiple years
2.9.16. Set the base year
2.9.17. Report results of import step
2.9.18. Import day-specific and hour-specific data
2.9.19. Processing hour-specific CEM data

The importing of emission inventory and related data is the first processing step needed for any emissions processing effort. The Smkinven program imports data for anthropogenic sources, the Rawbio program imports BEIS2 land use data for biogenic sources, and the Normbeis3 program imports BEIS3 land use data for biogenic sources. In this section, we focus on the import of the anthropogenic inventories using Smkinven. The biogenic import is further described in Section 2.17, “Biogenic processing”.

Smkinven performs many types of activities during import of the anthropogenic inventories. Though the primary purpose is reading the data from ASCII formats and outputting and I/O API SMOKE intermediate inventory, there are many other actions that need to be performed duringthe inventory import stage of processing. These actions are the following:

  1. Check that the formats of the input files are correct and consistent, and ensure that all data can be read properly.

  2. Assign pollutant names to data input by code numbers.

  3. Select pollutants from the input files to keep for further SMOKE processing.

  4. When multiple files are provided, combine all annual and/or average-day data into aconsistent inventory. This includes checking for duplicates and possibly aborting, depending on program options set by the user.

  5. Combine toxics and criteria inventories, and eliminate duplicate mass using either an integrate or no-integrate approach.

  6. Sort the inventory records into the order expected by other SMOKE programs.

  7. Aggregate or disaggregate toxics emissions data as specified by user inputs.

  8. Assign point-source locations to area sources, when available.

  9. Fill in and check point-source stack parameters.

  10. Convert stack locations from UTM to lat-lon.

  11. Optionally ensure that lat-lon coordinates are in the Western Hemisphere.

  12. Convert units of emissions and activities to the units used in the SMOKE intermediate inventory.

  13. Set the weekday averaging approach.

  14. Assign country codes, years, and time zones.

  15. Handle inventories that have data for multiple years.

  16. Set the base year.

  17. Report results of import including pollutant totals for toxics data and other information needed for quality assurance.

  18. Import day-specific and hour-specific data, if available, and ensure that records in these files match inventory records provided in the annual or average-day inputs.

In the following subsections, we describe what SMOKE does for each of these activities.

2.9.1. Check the correctness and consistency of input file formats

Smkinven can read the following ASCII formats for annual and average-day inventory data:

  • Inventory Data Analyzer format: The IDA format is the SMOKE input format that is most similar to the NEI “flat file” format. EPA typically prepares the IDA format for officially released versions of the NEI. There is a different IDA format for each source category: area (including nonroad mobile), point, and on-road mobile.

  • SMOKE ORL format: This set of input formats is used for inputting point, nonpoint, on-road, and nonroad HAP emissions inventories, also called toxics emission inventories. There is a different SMOKE ORL format for nonpoint, point, nonroad mobile, and on-road mobile sources.

  • EMS-95 format: These are input formats used for EMS-95. They are provided to allow EMS-95 users to input their data to SMOKE. There are different formats for area(including nonroad mobile), point, and on-road mobile sources. The EMS-95 formats must be used in conjunction with the list format (below).

  • List format: This is the input format used to provide multiple files of the previous three types to Smkinven in a single run. This format is simply an ASCII file that contains a list of other files.

  • Gridded I/O API format: This format is a gridded I/O API file for allowing the import of pregridded data. This approach is described more in Section 2.8.2.2, “Pregridded data route”.

Smkinven ensures that all file formats provided to SMOKE are correct and include the required data fields. The formats and their required fields are provided in Chapter 8, SMOKE Input Files.

Smkinven also ensures that the files listed in list format input files do not include incompatible file formats. Usually, only one file format can be used for one Smkinven run. The major exception to this rule is that Smkinven can import the IDA and SMOKE ORL formats in the same run. The only other minor exception to the input rule is that the EMS-95 point-source format actually includes five separate formats, which can be input at the same time, when users follow the instructions about the order of these files provided in Section 8.2.8, “PTINV: Point source annual or average day emissions”.