AIR DISPERSION MODELING HIGHLIGHTS FROM 2012 ACEPresentation for the A&WMA UMS Board MeetingAugust 21, 2012

Sergio GuerraWenck Associates, Inc.

Outline• Introduction• EMVAP• Distance limitation for AERMOD use• Case studies

• North Dakota• Gibson Station

Why do we use a model?

What is a model?• A Model is a way of expressing the relationship between the different variables of a system in mathematical terms

What is an Air Quality ModelAn attempt to predict or simulate the ambient

concentrations of contaminants in an area of interest.

An Air Quality Model can be as simple as an algebraic equation or more complex

AERMOD• AERMOD is a steady-state plume model that incorporates

air dispersion based on planetary boundary layer turbulence structure and scaling concepts, including treatment of both surface and elevated sources, and both simple and complex terrain.

• AERMOD replaced the Industrial Source Complex (ISCST3) model as EPA’s regulatory model on December 9, 2006

• Preprocessors include: AERMET,AERMINUTE,AERSURFACE,AERMAP,BPIP

What are the inputs of a dispersion model?• Source data• Building data• Receptor data• Site data• Meteorological data• Terrain data

ACE 2012 Highlights

Emissions Variability Processor (EMVAP)EMVAP an Emission Variability Processor for Modeling ApplicationsPaper 2012-A-341-AWMARichard P. Hamel, Robert J. Paine, David W. Heinold (AECOM)Naresh Kumar and Eladio Knipping (EPRI)

EMVAP• Large variation possible over the course of a year• Intermittent sources (e.g., emergency backup engines or

bypass stacks) present modeling challenges• For these sources, assuming fixed peak 1‐hour emissions

on a continuous basis will result in unrealistic modeled results

• Better approach is to assume a prescribed distribution of emission rates

• EMVAP uses this information to develop alternative ways to indicate modeled compliance using a range of emission rates instead of just one value

Hourly emission profile

Cumulative frequency distribution

Distance limitations of AERMODLimitations of Steady-State Dispersion Models and Possible Advanced ApproachesPaper 2012-500-AWMAGary Moore, Robert Paine, and David Heinold (AECOM)Steve Hanna (Hanna Consultants)

Short range model distance applicability• Plumes are assumed to travel to infinite distances within 1

hour (“lighthouse beam” effect)• Each hour, the previous hour’s emissions are replaced

and forgotten• Worst‐case conditions, especially associated with low

winds, result in impossible distances• Currently, though, US EPA considers these models to be

applicable to a rather arbitrary distance of 50 km• Equivalence between ISC and CALPUFF for 2 met data

locations:• Salem, Oregon• Evansville, Indiana

Short range model distance applicability• 20‐30 km is the extent a single hour’s travel for most of

the hours• Even after 4‐5 hours, more than half of air parcels followed with a 10‐m wind are still on the 50‐km modeling domain• Results suggest that a 20‐km limit seems more

appropriate for steady‐state model (e.g., AERMOD) applicability rather than the current limit of 50 km

Case Study 1- North DakotaComparison of AERMOD Modeled 1-hour SO2 Concentrations to Observations at Multiple Monitoring Stations in North DakotaPaper 2012-A-353-AWMAMary M. Kaplan, Robert Paine (AECOM)

Evaluation Opportunity in North Dakota• Mercer County: Antelope Valley Station and Great Plains

Synfuels Plant• Electrical generating unit sources dominate SO2

emissions – hourly data available• Five SO2 monitors in area within about 10 km of two

nearby “central” sources• Site‐specific PSD quality meteorological data years

available (10‐m tower)• Major SO2 sources within 50 km were modeled• Five recent years of data were used

Case Study 1- Dakota Gasification Co.• Allowable emissions used for all sources, assumed to be

constantly at peak rates• Receptors placed at monitor sites only, using actual

terrain (even though slopes are < 2%), except to characterize the spatial concentration pattern

• Four of the five monitors were at elevations near local stack base, a fifth monitor was about 100 m higher

Test of Terrain Problem for Gentle Slope• Used generic tall stack buoyant source• Modeled both flat and very gentle terrain• Terrain case was uniformly sloped upward 1% in all

directions• Modeled entire year of meteorology• Obtained peak concentration on each ring of receptors

out to 50 km• Plots follow for flat and gently sloping terrain

Conclusions from Gentle Slope Test• AERMOD has unusual prediction result for very low wind,

stable conditions and low slope• Problem is, in part, caused by very low mixing height that

leads to very compact plume• Mixing height is below building obstacles, which the

model does not know about• Plume stays perfectly level; terrain should not be

considered in these cases• With terrain, result is an unexpected plume impact “bulge”

at point of terrain impact

Case Study 2-Gibson Generating Station• Review of IDEM’s AERMOD Evaluation for the Gibson

Generating Station• Robert Paine and Carlos Szembek (AECOM)

Case Study 2-Gibson Generating Station• The Indiana Department of Environmental Management

(IDEM) conducted an evaluation of AERMOD• Gibson is an isolated source with 4 stacks and 3 nearby

monitors• On-site met data and hourly SO2 emission data for 2010• Comparison of monitored versus predicted concentrations

Case Study 2-Gibson Generating Station

Case Study 2-Gibson Generating Station• Low winds produced highest concentrations (~0.5m/s)• Plume travel distance within an hour is short of the

distance needed to reach maximum receptors• Formulation problem or coding error related to sigma-z

(used to calculate effective mixing lid)

Questions?

Sergio A. GuerraEnvironmental EngineerPhone: (651) 395-5225sguerra@wenck.com