Anybody knowledgeable or experienced in using CALPUFF for particulate monitoring, or other air quality issues? Second, will the EPA accept CALPUFF models of particulate dispersion?
Recently, Gretchen G. posted that the EPA's AERMOD modeling, which is used in air quality permits, was not good because it ignored the conservation of mass. Gretchen, please correct anything that I got wrong or left out here.
I'm concerned because I think that air quality permits for sand mines, processing, and trans-loads are given using a bad model. The results of a bad model may be people breathing pollution.
The Wisconsin DNR uses some version of AERMOD, & appears to leave out the AERMAP step, which is the terrain modeling that's supposed to be included.
I keep coming across CALPUFF as an alternate. to AERMOD.
I wonder if its modelling is a better data fit than AERMOD. I also wonder if anybody tests models with measurements, some kind of reliable evidence.
I know nothing about these models, but thought I'd add some links for reference
AEROMOD: https://www3.epa.gov/scram001/dispersion_prefrec.htm#aermod
CALPUFF http://src.com/calpuff/calpuff1.htm
going by the name alone, I'd say CALPUFF is clearly better.
Reply to this comment...
Log in to comment
The question if "which model is best" is not easy to answer as it is always dependent on what the results are used for. There have been a few studies comparing the performance of these models ... one from 2013 is: (full text download) https://www.researchgate.net/publication/262922397_Performance_evaluation_of_AERMOD_CALPUFF_and_legacy_air_dispersion_models_using_the_Winter_Validation_Tracer_Study_dataset
The basic message is that no single model is always better than the others but they do state that steady state models (e.g. AERMOD) are preferable for regulatory use as they require less user judgement while lagrangian models (e.g. CALPUFF) require that the user makes lots of decisions that affect the quality of the results and those decisions can only be made with expert knowledge.
I've used CALPUFF many times in the past and I know that it can be over- and mis-used quite easily. On the other hand I haven't used AIRMOD but I suspect that if you're not careful it can also be used outside its purpose.
I can't comment on the particulars of this case as I don't know the details (and I'm half a world away in NZ ;-) ) but in general, the choice of model is defined by the data available to run it and the intended use of the results. Whether that means that CALPUFF is better suited than AIRMOD it's not something I can answer.
Reply to this comment...
Log in to comment
Hi Mary, This is a great question. I want to clarify that I don't necessarily think AERMOD is a bad model -- it just has draw-backs, and those draw-backs could explain differences between observed and modeled air quality indicators. Anyway, I've been reading a bit about the models (information hosted here), especially since it looks like CALPUFF is the model of choice for modeling emissions' impacts on visibility and regional haze. I agree with @guolivar that the appropriateness of a model is dependent on its intended use though. It looks like a combination of AERMOD and CMAQ has been deemed the best available modeling for primary pollutants (the NAAQS pollutants) in the continental US.
Reply to this comment...
Log in to comment
Hi again,
This document may be useful in deciphering situations in which CALPUFF would be more appropriate than AERMOD: https://www3.epa.gov/ttn/scram/7thconf/calpuff/calpuff_near-field_technical_issues_092608.pdf. It looks like wind-channeling valleys are the sorts of terrain in which CALPUFF can be significantly more accurate than AERMOD at predicting contaminant dispersion.
G
Reply to this comment...
Log in to comment