The other approach to modeling elevated sources is to create an elevated-point-source input file for one of the UAM models or CAMX. To do this, SMOKE performs the following steps:
Uses the Smkinven program to import the annual, average-day, day-specific, and/or hour-specific emissions.
Optionally uses the Temporal program to calculate hourly emissions if emissions will be used as a criterion for selecting elevated sources or PinG sources. For example, you may wish to select facilities with NOx emissions greater than 100 tons/day.
Uses the Elevpoint program to select elevated and optionally PinG sources. Unlike processing for CMAQ, this step is always required.
Uses the Elevpoint program to create the STACK_GROUPS file, which is needed for creating the elevated-point-source file with Smkmerge.
Uses the Laypoint program to compute elevated plume rise for explicit plume rise sources, and store the layer fractions for the explicit sources only.
Uses the Smkmerge program to combine the STACK_GROUPS information, optional explicit plume rise information, and the hourly emissions to generate an ASCII elevated-point-source
file with optional PinG flags.
The primary difference between this approach is that the Elevpoint processing step is required. Traditionally, elevated point sources have been selected for the UAM or CAMX models using the analytical plume rise calculation. While Elevpoint can perform this calculation and use it exclusively to determine the elevated sources, you can also use the other selection
criteria if desired: emissions values, emissions rank, stack parameters, and plant numbers. As described in Section 2.14.1, “Computing layer fractions for CMAQ”, if the selection criteria include emissions, then the PTMPLIST file will be used to input all hourly emissions files for the entire modeling episode.
If explicit plume rise sources (e.g., wildfires with precomputed hourly plume rise) are included in the inventory, you must
run the Laypoint program to compute the layer fractions for these sources only. This is the only reason Laypoint would be run in the UAM/CAMX processing approach, and it requires that the EXPLICIT_PLUMES_YN and the HOUR_PLUMEDATA_YN settings be set to Y. These settings cause Laypoint to write the layer fractions to the PLAY_EX file instead of to the usual PLAY file. For these explicit sources, Laypoint will skip the plume rise calculation. Instead, it will use the hourly data from the SMOKE PHOUR intermediate file containing the fraction of emissions in layer 1 and the top and bottom of the plume. Laypoint will combine these data with the pressure weights used for all elevated point sources to compute the fraction of emissions
to go into each layer.
The Smkmerge program looks for the SMK_ASCIIELEV_YN setting to determine whether the ASCII output file should be created. When this is set to Y, the output files from Elevpoint (PELV and STACK_GROUPS) are read to determine which sources should not be included in the 2-D emissions output file for point sources. The emissions
from these sources are instead output to an ASCII elevated file along with stack parameters and locations so that the AQM
can compute the plume rise.
If EXPLICIT_PLUMES_YN is also set to Y, Smkmerge will read the PLAY_EX file for the explicit plume sources. Since the UAM-based approach assumes that the AQM will compute the plume rise, SMOKE
must manipulate the input file to trick the model into using precomputed plume rise. This is done by inserting fake stacks
into the ASCII elevated file that extend to the center of each of the model layers and setting the stack parameters so that
the plume rise calculation will keep the emissions associated with the fake stacks in the layer of the stack. As the emissions
move from layer to layer, Smkmerge moves the reported emissions in the ASCII elevated file from fake stack to fake stack to represent the same behavior.