SMOKE is primarily an emissions processing system and not an emission inventory preparation system. This means that its main purpose is to provide an efficient tool for converting emission inventory data into the formatted emission input files required by an AQM. However, for mobile and biogenic sources, SMOKE does offer emission inventory preparation functions. For mobile sources, SMOKE computes an emission inventory from mobile-source activity data, using emission factors from the MOBILE6 model. Previous versions of SMOKE (version 1.4 and below) supported the MOBILE5b model. For biogenic sources, SMOKE includes both the BEIS2 and BEIS3 models for computation of hour-specific, meteorology-based biogenic emissions from vegetation and soils.
The SMOKE modeling system can be run using either of two approaches. The most common method is through the use of UNIX scripts, which includes the Linux versions of the UNIX operating system. The second approach is to run SMOKE using the MIMS interface, which is a general-purpose graphical user interface (GUI).
Below, we summarize the major features of SMOKE v2.3.2.
The latest Biogenic Emission Inventory System (BEIS) model version 3.13 is integrated to the SMOKE v2.3.2 for generating gridded, hourly emissions in a format consistent for air quality modeling. BEIS 3.13 revises the parameterizations that account for the effects of leaf temperature and solar radiation on emissions.
Compatible with the Federal Aviation Administration (FAA)'s Emission Modeling and Dispersion Modeling System (EDMS) used to estimate emissions from airports. A new utility program EDMS2Inv creates SMOKE-ready inventory from EDMS output.
Enhancement on combining aggregated criteria VOC and hazardous air pollutant (HAP) inventories were made throughout the SMOKE.
The updated Smkinven can optionally support upto 16 character size for the Chemical Abstracts Services (CAS) code, also known as the pollutant code, to allow for all HAPs which have more than 5 characters for the pollutant name in day-specific (PTDAY) and hour-specific (PTHOUR) emission inventories.
Support fully extended ORL format for Smkinven and Grwinven.
Below, we also summarize the major features from the previous SMOKE v2.3 release.
Supports new wild and Rx fires emission inventory in new ORL-formatted inventories. The fire emissions are either estimated by the EPM/CONSUME model in the Bluesky modeling framework or they come from the 2002 NEI.
Supports the extended ORL format. It accepts source type codes for area, nonroad mobile and onroad mobile sources, as well as ORIS and boiler identification codes for point sources.
Supports new CEM format with additional variables including flow rate for the boiler unit, and pollutant status flags ( measured/substituted/not used ).
Allows multiple plant descriptions and FIPS codes for an ORIS ID when matching hourly point data to annual data and also supports matching hourly point source data to annual data in either IDA or ORL format.
Supports 16 characters for the CAS code, also known as the pollutant code, to allow criteria pollutants to be input using the ORL format.
Treats duplicate sources as separate sources with a warning message instead of dropping the emission.
Supports only new INVTABLE
format that has emissions MODE in a new column after the SMOKE NAME. These columns will be used together to create the internal
SMOKE name.
Improves the SMKINVEN_FORMULA
option to support multiple formula.
Outputs ORL-formatted file as a more condensed format in comma delimited format with quotes around all string fields.
Supports an optional special header in GSPRO file that defines the meaning of NONHAPVOC to support of combining criteria and toxic VOC using the NONHAPVOC calculation feature of SMOKE.
Supports both new and old GSCNV
file. New GSCNV
file includes additonal infomation, such as name of pollutant or emission types converting from and to.
Accepts source type codes of 03 and 04 for nonroad and onroad mobile respectively, and accept twelve new cases in the projection packet for plant-specific source matching. Fixed a bug in the plant-specific cross reference assignment.
Read in surrogates from either one file or multiple files and now requires the SRGDESC file. As a result of this update, Grdmat requires far less memory to run than it did in previous versions.
Perform multiple non-sequential processing dates.
Re-normalize the assigned temporal profile based on the duration of fires.
Below, we also list the major features from the previous release SMOKE v2.2.
Supports Inventory Data Analyzer (IDA) formats, Emissions Modeling System, ‘95 (EMS-95) format, and a SMOKE one-record-per-line (ORL) format. The criteria and PM National Emissions Inventory (NEI) data are most readily converted to the IDA formats or ORL format. The toxics inventory must be converted to the ORL format.
Supports both gridded and county-total land use for biogenic emissions modeling.
Includes the BEIS2 or BEIS3 system (and the user can choose which he or she wants).
Includes a driver for MOBILE6 runs and features to improve run time when using MOBILE6 for large domains and long time periods (can support annual, national runs using MOBILE6).
Multicountry capability, up to 10 countries.
Lower disk space requirements over other emissions processing software for the same level of detail in the emissions.
Any pollutant can be processed by the system.
Any chemical mechanism can be used to partition pollutants to model species, as long as the appropriate input data are supplied.
Supports Lambert conformal, lat-lon, UTM, and polar stereographic output grid projections with any number of grid cells.
Control strategies can include changes in the reactivity of emitted pollutants. This is useful, for example, when a solvent is changed in an industrial process.
Run-time memory allocation, eliminating any need to recompile the source code for different inventories, grids, or chemical mechanisms.
No third-party software is required to run SMOKE, although some input file preparation, such as gridded land use or spatial allocation surrogates, does require other software.
Supports plume in grid (PinG) processing.
Integrated use of Continuous Emissions Monitoring (CEM) data, an hour-specific dataset of emissions from electric generating facilities. Also supports other day-specific and hour-specific point-source data.
Supports hour-specific point-source stack parameters (exit temperature, exit velocity, and exit flow rate).
Supports externally computed hourly plume rise (used for processing wildfires as point sources with a different plume rise algorithm than is available in SMOKE).
Supports emissions data input formats for the following AQMs and chemical mechanisms:
Community Multiscale Air Quality (CMAQ) model: part of the Models-3 system. SMOKE v2.0 and later supports a research toxics version of CMAQ for non-point (stationary area) sources, nonroad mobile sources, on-road mobile sources, point sources, and biogenic sources.
Multiscale Air Quality Simulation Platform (MAQSIP): for both standard and real-time air quality modeling. Also MAQSIP-PM as an alternative system for modeling regional haze.
Regulatory Modeling System for Aerosols and Deposition (REMSAD)
Comprehensive Air Quality Model, with Extensions (CAMX)
Urban Airshed Model (UAM), versions IV and V