ML110590928

Lr Hearing - FW: Modeling Review - Indian Point

ML110590928
Person / Time
Site:	Indian Point
Issue date:	01/31/2011
From:	Office of Nuclear Regulatory Research
To:	Division of License Renewal
References
Download: ML110590928 (15)

Text

1 IPRenewal NPEmails From:

Ghosh, Tina Sent:

Monday, January 31, 2011 10:48 AM To:

Stuyvenberg, Andrew

Subject:

FW: Modeling Review - Indian Point Attachments:

Modeling Review Document 6 14 10.doc Follow Up Flag:

Follow up Flag Status:

Flagged From: Jones, Joe A [1]

Sent: Tuesday, June 15, 2010 3:53 PM To: Palla, Robert; Ghosh, Tina Cc: Bixler, Nathan E

Subject:

Modeling Review - Indian Point Bob and Tina, We have completed our review of the AERMOD and CalPuff models comparing characteristics to the MACCS2 model.

Joe Schelling, one of our MACCS2 modelers, did the research on this and he and Nate iterated the document a couple of times.

Please let us know if you have any questions regarding this review.

Thanks Joe

Hearing Identifier:

IndianPointUnits2and3NonPublic_EX Email Number:

2295 Mail Envelope Properties (6F9E3C9DCAB9E448AAA49B8772A448C55EE2C75CCD)

Subject:

FW: Modeling Review - Indian Point Sent Date:

1/31/2011 10:48:04 AM Received Date:

1/31/2011 10:48:05 AM From:

Ghosh, Tina Created By:

Tina.Ghosh@nrc.gov Recipients:

"Stuyvenberg, Andrew" <Andrew.Stuyvenberg@nrc.gov>

Tracking Status: None Post Office:

HQCLSTR01.nrc.gov Files Size Date & Time MESSAGE 569 1/31/2011 10:48:05 AM Modeling Review Document 6 14 10.doc 109634 Options Priority:

Standard Return Notification:

No Reply Requested:

No Sensitivity:

Normal Expiration Date:

Recipients Received:

Follow up

SevereAccidentMitigationAnalysisModelingPlan 6/14/2010

Objective The objective of this plan is to determine the feasibility, and if appropriate, the level of effort required to use the AERMOD and CALPUFF models in the same capacity MACCS2 is used to perform consequence modeling in terms of dose, contamination, and economic cost for severe accident mitigation alternatives (SAMA) analyses. AERMOD and CALPUFF are preferred and recommended by the U.S. Environmental Protection Agency for modeling air dispersion and air quality [1,2]. As air dispersion and air quality models, AERMOD and CALPUFF include models of contaminant dispersion and deposition that produce contaminant concentrations at various receptor locations as a function of time. MACCS2 models dispersion and deposition of a radioactive plume released from a nuclear power plant, and includes models of radioactive decay and ingrowth, ground contamination, dose consequences to the public for defined emergency response, and economic impacts of the release.

MACCS2Overview The purpose of the MACCS2 code is to simulate the impact of severe accidents at nuclear power plants on the surrounding environment. The principal phenomena considered in MACCS2 are atmospheric transport, mitigative actions based on dose projection, dose accumulation by a number of pathways, including food and water ingestion, early and latent health effects, and economic costs [3].

MACCS2 models atmospheric dispersion and transport, and includes models for deposition, weathering, resuspension, and radioactive decay. A Gaussian plume model is used to model plume dispersion during downwind transport. Scaling factors are used in MACCS2 to model effects of surface roughness on plume dispersion. Statistical distributions of consequence measures that depict the range and probability of consequences are generated by MACCS2 in terms of complementary cumulative distribution functions (CCDFs).[4] These distributions are generated using stratified Monte Carlo sampling of combinations of representative sets of source terms, weather sequences, and exposed populations. Radioactive decay and ingrowth, and dry and wet aerosol deposition processes are modeled. Economic effects models in MACCS2 are intended to estimate the direct offsite costs resulting from a reactor accident and include costs resulting from both short-term and long-term protective actions [3].

The principal phenomena considered in MACCS2 are atmospheric transport using a Gaussian plume model, short-term and long-term dose accumulation through several pathways (including cloudshine, groundshine, inhalation, deposition onto the skin, and food and water ingestion), mitigative actions based on dose projection, early and latent health effects, and economic costs. The following phenomena can be incorporated within a single calculation:

Release characteristics;

Meteorological sampling; Atmospheric dispersion and deposition; Exposure pathways and duration; Protective actions and dose mitigation; Movement of populations as cohorts; Individual and population doses; and Health and economic consequences.

MACCS2 has three major components, ATMOS, EARLY, and CHRONC. Of these components, only ATMOS, which treats dispersion and deposition using a Gaussian plume model, has similarities to AERMOD and CALPUFF. The EARLY and CHRONC modules are used to model emergency phase events and radiological consequences and economic costs of response actions, which are not considered in the functionality of AERMOD and CALPUFF.

ATMOS calculates the dispersion and deposition of material released to the atmosphere as a function of downwind distance, and uses a Gaussian plume model with Pasquill-Gifford dispersion parameters. ATMOS treats the following phenomena:

building wake effects, buoyant plume rise, plume dispersion during transport, wet and dry deposition, and radioactive decay and ingrowth [4].

EARLY models the time period immediately following a radioactive release, referred to as the emergency phase, which may extend up to one week after the first plume reaches any downwind location. User-specified scenarios may include evacuation, sheltering, and dose-dependent relocation. Dose calculations from early exposure consider five pathways: cloudshine, cloud inhalation, groundshine, resuspension inhalation, and skin dose from deposition onto skin. Acute doses and lifetime dose commitments are calculated [4].

CHRONC simulates events following the emergency-phase time period.

CHRONC calculates the number of health effects resulting from direct exposure to contaminated materials as well as health effects caused by subsequent consumption of contaminants. It calculates mitigative actions taken to reduce doses to the public, including decontamination, interdiction, and condemnation of property. CHRONC calculates the economic costs of these long-term protective actions as well as the costs from the emergency response actions modeled during the emergency phase [4].

EPARegulatoryAtmosphericModelingCodes Two code systems recommended by the U.S. Environmental Protection Agency (EPA) for use in regulatory atmospheric modeling, and required for use for several programs are the AERMOD and CALPUFF modeling systems [1,2]. The EPA Support Center for Regulatory Atmospheric Modeling website provides the following basic descriptions of these modeling systems [1]:

AERMOD Modeling System - 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.

CALPUFF Modeling System - A non-steady-state puff dispersion model that simulates the effects of time-and space-varying meteorological conditions on pollution transport, transformation, and removal. CALPUFF can be applied for long-range transport and for complex terrain.

AERMODOverview AERMOD was designed for short range (up to 50 km) dispersion from stationary sources, and includes modeling of surface terrain on the behavior of air pollution plumes and of building downwash effects. A Gaussian plume model is used [5]. AERMOD can model exponential decay. (The default regulatory option forces use of a 4 hr half life for SO2 in an urban source, and does not allow for exponential decay for other applications

[6].) Multiple sources from a single point can be modeled, and there are several options for specifying an array of receptor locations. Only a single value of halflife can be used in a given run [7]. The original version of the AERMOD model did not include algorithms to handle the gravitational settling and removal by dry deposition of particulates, or scavenging and removal by wet deposition of gases and particulates [8].

However, the AERMOD User Guide Addendum 09292 describes the later inclusion of dry and wet deposition algorithms for both particulate and gaseous emissions [10]. No dose pathway, dose mitigation (KI ingestion, sheltering, or evacuation), or economic models appear to be included in AERMOD.

CALPUFFOverview CALPUFF is proposed by the EPA for applications involving long-range transport, which is typically defined as transport over distances beyond 50 km, and requires approval by the relevant reviewing authorities for EPA regulatory applications involving transport distances less than 50 km [11]. A puff model may be used for most applications where a plume model is appropriate, but the technical decision on using a puff model should be based on whether the straight-line, steady-state assumptions on which a plume model is based are valid [11]. The EPA approved version of the CALPUFF System includes Version 5.8 - Level 070623 of CALPUFF and includes changes through MCB-D. Wet and dry deposition, and chemical transformation are modeled. No radioactive decay and ingrowth, dose pathway, dose mitigation (e.g.,

evacuation), or economic models are included in CALPUFF.

A summary comparison of the dispersion and deposition models of the ATMOS component of MACCS2, AERMOD, and CALPUFF are shown in Table 1. More detailed summary information on the latter two codes from Appendix A to Appendix W of 40 CFR Part 51 is shown in Appendix A.

Table1.SummaryComparisonofAtmosphericDispersionandDepositionModels Parameter MACCS2/ATMOS AERMOD CALPUFF Plume Model Non-steady-state Gaussian plume with multiple plume segments; Plume expansion factor for meander; Briggs model for buoyant plume rise Steady-state Gaussian plume model Non-steady-state Lagrangian Gaussian puff model; Wind shear puff split; Briggs model; Partial penetration; Buoyant and momentum rise; Stack tip effects; Vertical wind shear Dispersion Two models:

Pasquill-Gifford stability class sigma-y and sigma-z; Time-based dispersion function No user input Five models, including P-G Building Wake and Terrain Effects Building wake entrainment; Surface terrain effects modeled with scaling factor Stack-tip downwash; Surface terrain effects (via surface roughness length)

Building downwash by two methods; Applicable to rough or complex terrain; Simulates changes in the flow and dispersion rate induced by terrain features Wet Deposition Exponential function of rain duration and intensity Wet particulate and gaseous deposition Empirical scavenging model as function of pollutant and precipitation type Dry Deposition Source depletion by particle size dependent deposition velocity Dry particulate and gaseous deposition Resistance model Meteorological Sampling Weather bin or random sampling of annual data; Specification of mixing layer height; 15-min, 30-min, or hourly time periods Surface boundary layer and profile variables; Hourly time periods Hourly wind/temp on 3D grid, w/2D parameters Material Sources Point or area source w/multiple plume segments, aerosol sizes, and isotopic components Multiple sources; point, volume, area, and line; Variable emission rates Time varying point, line, volume, and area sources; Variable emission rates

Parameter MACCS2/ATMOS AERMOD CALPUFF Receptor Locations Polar grid centered on facility; Up to 35 radial intervals; Up to 64 compass directions; Up to 7 receptors per grid Multiple receptors; Discrete, polar or Cartesian grid networks; Variable receptor height Essentially unlimited gridded or discrete receptors Range 0.05 to 9999 km; Up to 50 km 10s of m to 100s of km Decay and Ingrowth Radioactive decay and daughter ingrowth over six generations Single exponential decay Linear removal and chemical conversion All three codes use a type of Gaussian dispersion model. AERMOD uses the simplest Gaussian representation, which is a steady-state plume; MACCS2 also uses a plume model, but the wind speed, stability class, and precipitation rate can change as the plume moves through the grid. CALPUFF uses a Gaussian puff model, which allows wind direction to change along the course of the plume.

An important distinction between these codes is that AERMOD and CALPUFF were specifically designed to treat chemical air pollutants. Because of this, they treat removal of the pollutants by chemical reaction but do not handle radioactive decay with daughter ingrowth. On the other hand, MACCS2 was specifically designed to treat radioactive air pollutants. It treats radioactive decay and daughter ingrowth for decay chains up to a length of six.

An approach to addressing the objective of this plan is to evaluate whether or not output generated by AERMOD and CALPUFF could be used as input to a post-processing routine to model radioactive decay and dose pathway processes, emergency phase relocation actions, and economic costs. That is, AERMOD and CALPUFF account for plume dispersion and deposition, and provide material concentrations at receptor locations. Approximation would be needed to model radioactive decay with daughter ingrowth during atmospheric transport for a large set of parents and daughters, and the corresponding adjustment made to deposited concentrations. An approach to modeling cloudshine effects would also be needed to ensure that predictions are not underestimated. From the deposited material concentrations, it may be possible to develop a set of dose conversion factors to account for various uptake pathways. The most difficult implementation would be to account for the interaction between time-dependent plume concentrations and evacuating public. Deposited material concentrations could be used to estimate economic costs for mitigative actions.

However, the mitigative actions, such as decontamination of property, are coupled with public doses. As a result, a coupled model is required to properly treat these effects.

Unlike MACCS2, neither AERMOD nor CALPUFF appear to have options for generating statistical distributions of consequences for variations in source terms, meteorology, or populations. Alternatively, it may be possible to incorporate these modeling functions in the codes directly.

Developing, testing, producing quality assurance (QA) documentation, and obtaining EPA approval for either AERMOD or CALPUFF in the context of nuclear reactor accidents would likely require considerable effort over a period of several years.

Some of the steps involved are shown in the following graphic.

AERMODOutput For each averaging period (1, 3, 8, or 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> or 1 month), concurrent averages for all receptors for each day of data processed are generated. Receptor networks are printed first, followed by concurrent averages for each source group. Optional output files include POSTFILE, a file of concurrent (raw) results at each receptor. Appendix D of the AERMOD User Guide shows for the POSTFILE option a file including X and Y coordinates of the receptor location, receptor height, averaging period, source group ID, and either the date variable for the end of the averaging period or the number of hours in the period for period averages [9].

CALPUFFOutput CALPUFF output files include one-hour averaged concentrations at gridded and discrete receptor locations for selected species and of wet and dry deposition fluxes; hourly reports of mass fluxes into and out of regions, and of mass changes for all species, are also produced. The CALPOST postprocessor program averages and reports concentrations or wet/dry fluxes based on hourly data in the CALPUFF output file [12].

Summary The objective of this plan is to determine the feasibility, and if appropriate, the level of effort required to use the AERMOD and CALPUFF models in the same capacity as MACCS2 for performing consequence modeling in terms of dose, health effects, land contamination, and economic cost for SAMA analyses. Because AERMOD and CALPUFF do not include models for emergency response, radioactive decay and ingrowth, radiological health effects, mitigative actions, or economic costs, code revision or development of post-processing utilities would be needed to utilize either of these codes to model nuclear reactor accident consequences in a manner similar to MACCS2.

Statistical sampling of parameter distributions is also a feature of MACCS2 that is not available in AERMOD or CALPUFF. Developing, testing, and acquiring acceptance of the results produced by these approaches would involve significant effort. Model assumptions and constraints needed for these utilities may limit the accuracy/fidelity of the results obtained.

References

1. Preferred/Recommended Models, Technology Transfer Network, Support Center for Regulatory Atmospheric Modeling, U.S. Environmental Protection Agency, http://www.epa.gov/scram001/dispersion_prefrec.htm, accessed May 6, 2010.

2. Appendix A to Appendix W of Part 51 - Summaries of Preferred Air Quality Models, Federal Register, Vol. 70, No. 216, Nov. 9, 2005.

3. Jow, H-N, J.L. Sprung, J.A. Rollstin, L.T. Ritchie, D.I. Chanin, MELCOR Accident Consequence Code System (MACCS) Model Description, NUREG/CR-4691, SAND86-1562, Vol. 2, February, 1990.

4. Chanin, D, M.L. Young, J. Randall, and K. Jamali, Code Manual for MACCS2:

Volume 1, Users Guide, NUREG/CR-6613, SAND97-0594, Vol. 1, May, 1998.

5. Table 1, AERMOD: Latest Features and Evaluation Results, EPA-454/R-03-003, June 2003. (aermod_mep.pdf)

6. Users Guide for the AMS/EPA Regulatory Model - AERMOD (Under Revision),

EPA-454/B-03-001, U.S.

Environmental Protection

Agency, September 2004, p. 3-3.

7. Users Guide for the AMS/EPA Regulatory Model - AERMOD (Under Revision),

EPA-454/B-03-001, U.S.

Environmental Protection

Agency, September 2004, p. 3-8.

8. Users Guide for the AMS/EPA Regulatory Model - AERMOD (Under Revision),

EPA-454/B-03-001, U.S.

Environmental Protection

Agency, September 2004, p. 3-37.

9. Users Guide for the AMS/EPA Regulatory Model - AERMOD (Under Revision),

EPA-454/B-03-001, U.S.

Environmental Protection

Agency, September 2004, Appendix D.

10. Addendum, Users Guide for the AMS/EPA Regulatory Model - AERMOD, (EPA-454/B-03-001, September 2004), U.S. Environmental Protection Agency, October 2009, p. 20.

11. Question 1.1.1., CALPUFF FAQs Answers, CALPUFF Modeling, The Atmospheric Studies Group at

TRC, http://www.src.com/calpuff/FAQ-answers.htm#1.1.1, accessed 4/29/2010.

12. Scire, J.S. D.G. Strimaitis, and R.J. Yamartino, A Users Guide for the CALPUFF Dispersion Model (Version 5), Earth Tech, Inc., January 2000.

AppendixA.KeyFeaturesofRefinedAirQualityModels Parameter AERMOD1 CALPUFF1 TypeofModel AERMODisasteadystateplumemodel, usingGaussiandistributionsinthe verticalandhorizontalforstable conditions,andinthehorizontalfor convectiveconditions.Thevertical concentrationdistributionforconvective conditionsresultsfromanassumedbi Gaussianprobabilitydensityfunctionof theverticalvelocity.

(1)CALPUFFisanonsteadystatetime andspacedependentGaussianpuff model.CALPUFFtreatsprimary pollutantsandsimulatessecondary pollutantformationusinga parameterized,quasilinearchemical conversionmechanism.Pollutants treatedincludeSO2,SO4 2,NOX(i.e.,NO+

NO2),HNO3,NO3

,NH3,PM-10,PM-2.5, toxicpollutantsandotherspollutant speciesthatareeitherinertorsubjectto quasilinearchemicalreactions.The modelincludesaresistancebaseddry depositionmodelforbothgaseous pollutantsandparticulatematter.Wet depositionistreatedusingascavenging coefficientapproach.Themodelhas detailedparameterizationsofcomplex terraineffects,includingterrain impingement,sidewallscraping,and steepwalledterraininfluencesonlateral plumegrowth.Asubgridscalecomplex terrainmodulebasedonadividing streamlineconceptdividestheflowinto aliftcomponenttravelingoverthe obstacleandawrapcomponent deflectedaroundtheobstacle.

(2)Themeteorologicalfieldsusedby CALPUFFareproducedbytheCALMET meteorologicalmodel.CALMETincludes adiagnosticwindfieldmodelcontaining parameterizedtreatmentsofslopeflows, valleyflows,terrainblockingeffects,and kinematicterraineffects,lakeandsea breezecirculations,adivergence minimizationprocedure,andobjective analysisofobservationaldata.An energybalanceschemeisusedto computesensibleandlatentheatfluxes andturbulenceparametersoverland surfaces.Aprofilemethodisusedover water.CALMETcontainsinterfacesto prognosticmeteorologicalmodelssuch asthePennState/NCARMesoscale Model(e.g.,MM5;Section12.0,ref.86),

aswellastheRAMS,RucandEtamodels.

Parameter AERMOD1 CALPUFF1 PollutantTypes AERMODisapplicabletoprimary pollutantsandcontinuousreleasesof toxicandhazardouswastepollutants.

Chemicaltransformationistreatedby simpleexponentialdecay.

CALPUFFmaybeusedtomodelgaseous pollutantsorparticulatematterthatare inertorwhichundergoquasilinear chemicalreactions,suchasSO2,SO4 2,

NOX(i.e.,NO+NO2),HNO3,NO3

,NH3, PM-10,PM-2.5andtoxicpollutants.For regionalhazeanalyses,sulfateand nitrateparticulatecomponentsare explicitlytreated.

SourceReceptor Relationships AERMODappliesuserspecifiedlocations forsourcesandreceptors.Actual separationbetweeneachsource receptorpairisused.Sourceand receptorelevationsareuserinputorare determinedbyAERMAPusingUSGSDEM terraindata.Receptorsmaybelocated atuserspecifiedheightsaboveground level.

CALPUFFcontainsnofundamental limitationsonthenumberofsourcesor receptors.Parameterfilesareprovided thatallowtheusertospecifythe maximumnumberofsources, receptors,puffs,species,gridcells, verticallayers,andothermodel parameters.Itsalgorithmsaredesigned tobesuitableforsourcereceptor distancesfromtensofmetersto hundredsofkilometers.

Parameter AERMOD1 CALPUFF1 PlumeBehavior (1)Intheconvectiveboundarylayer (CBL),thetransportanddispersionofa plumeischaracterizedasthe superpositionofthreemodeledplumes:

Thedirectplume(fromthestack),the indirectplume,andthepenetrated plume,wheretheindirectplume accountsfortheloftingofabuoyant plumenearthetopoftheboundary layer,andthepenetratedplume accountsfortheportionofaplumethat, duetoitsbuoyancy,penetratesabove themixedlayer,butcandisperse downwardandreenterthemixedlayer.

IntheCBL,plumeriseissuperposedon thedisplacementsbyrandomconvective velocities(Weiletal.,1997).

(2)Inthestableboundarylayer,plume riseisestimatedusinganiterative approach,similartothatinthe CTDMPLUSmodel(seeA.5inthis appendix).

(3)Stacktipdownwashandbuoyancy induceddispersioneffectsaremodeled.

Buildingwakeeffectsaresimulatedfor stackslessthangoodengineering practiceheightusingthemethods containedinthePRIMEdownwash algorithms(Schulman,etal.,2000).For plumeriseaffectedbythepresenceofa building,thePRIMEdownwashalgorithm usesanumericalsolutionofthemass, energyandmomentumconservation laws(ZhangandGhoniem,1993).

Streamlinedeflectionandthepositionof thestackrelativetothebuildingaffect plumetrajectoryanddispersion.

Enhanceddispersionisbasedonthe approachofWeil(1996).Plumemass capturedbythecavityiswellmixed withinthecavity.Thecapturedplume massisreemittedtothefarwakeasa volumesource.

(4)Forelevatedterrain,AERMOD incorporatestheconceptofthecritical dividingstreamlineheight,inwhichflow belowthisheightremainshorizontal, andflowabovethisheighttendstorise upandoverterrain(Snyderetal.,1985).

Plumeconcentrationestimatesarethe weightedsumofthesetwolimiting plumestates.

Momentumandbuoyantplumeriseis treatedaccordingtotheplumerise equationsofBriggs(1975)fornon downwashingpointsources,Schulman andScire(1980)forlinesourcesand pointsourcessubjecttobuilding downwasheffectsusingtheSchulman Sciredownwashalgorithm,andZhang (1993)forbuoyantareasourcesand pointsourcesaffectedbybuilding downwashwhenusingthePRIME buildingdownwashmethod.Stacktip downwasheffectsandpartialplume penetrationintoelevatedtemperature inversionsareincluded.Analgorithmto treathorizontallyorientedventsand stackswithraincapsisincluded.

Parameter AERMOD1 CALPUFF1

However,consistentwiththesteady stateassumptionofuniformhorizontal winddirectionoverthemodeling domain,straightlineplumetrajectories areassumed,withadjustmentinthe plume/receptorgeometryusedto accountfortheterraineffects.

HorizontalWinds Verticalprofilesofwindarecalculated foreachhourbasedonmeasurements andsurfacelayersimilarity(scaling) relationships.Atagivenheightabove ground,foragivenhour,windsare assumedconstantoverthemodeling domain.Theeffectofthevertical variationinhorizontalwindspeedon dispersionisaccountedforthrough simpleaveragingovertheplumedepth.

Athreedimensionalwindfieldis computedbytheCALMET meteorologicalmodel.CALMET combinesanobjectiveanalysis procedureusingwindobservationswith parameterizedtreatmentsofslopeflows, valleyflows,terrainkinematiceffects, terrainblockingeffects,andsea/lake breezecirculations.CALPUFFmay optionallyusesinglestation (horizontallyconstant)windfieldsinthe CTDMPLUS,AERMODorISCST3data formats.

VerticalWind Speed Inconvectiveconditions,theeffectsof randomverticalupdraftanddowndraft velocitiesaresimulatedwithabi Gaussianprobabilitydensityfunction.In bothconvectiveandstableconditions, themeanverticalwindspeedisassumed equaltozero.

Verticalwindspeedsarenotused explicitlybyCALPUFF.Verticalwinds areusedinthedevelopmentofthe horizontalwindcomponentsby CALMET.

Horizontal Dispersion Gaussianhorizontaldispersion coefficientsareestimatedascontinuous functionsoftheparameterized(or measured)ambientlateralturbulence andalsoaccountforbuoyancyinduced andbuildingwakeinducedturbulence.

Verticalprofilesoflateralturbulenceare developedfrommeasurementsand similarity(scaling)relationships.

Effectiveturbulencevaluesare determinedfromtheportionofthe verticalprofileoflateralturbulence betweentheplumeheightandthe receptorheight.Theeffectivelateral turbulenceisthenusedtoestimate horizontaldispersion.

Turbulencebaseddispersioncoefficients provideestimatesofhorizontalplume dispersionbasedonmeasuredor computedvaluesofv.Theeffectsof buildingdownwashandbuoyancy induceddispersionareincluded.The effectsofverticalwindshearare includedthroughthepuffsplitting algorithm.Optionsareprovidedtouse PasquillGifford(rural)andMcElroy Pooler(urban)dispersioncoefficients.

Initialplumesizefromareaorvolume sourcesisallowed.

Parameter AERMOD1 CALPUFF1 Vertical Dispersion Inthestableboundarylayer,Gaussian verticaldispersioncoefficientsare estimatedascontinuousfunctionsof parameterizedverticalturbulence.In theconvectiveboundarylayer,vertical dispersionischaracterizedbyabi Gaussianprobabilitydensityfunction, andisalsoestimatedasacontinuous functionofparameterizedvertical turbulence.Verticalturbulenceprofiles aredevelopedfrommeasurementsand similarity(scaling)relationships.These turbulenceprofilesaccountforboth convectiveandmechanicalturbulence.

Effectiveturbulencevaluesare determinedfromtheportionofthe verticalprofileofverticalturbulence betweentheplumeheightandthe receptorheight.Theeffectivevertical turbulenceisthenusedtoestimate verticaldispersion.

Turbulencebaseddispersion coefficientsprovideestimatesofvertical plumedispersionbasedonmeasuredor computedvaluesofw.Theeffectsof buildingdownwashandbuoyancy induceddispersionareincluded.Vertical dispersionduringconvectiveconditions issimulatedwithaprobabilitydensity function(PDF)modelbasedonWeilet al.(1997).Optionsareprovidedtouse PasquillGifford(rural)andMcElroy Pooler(urban)dispersioncoefficients.

Initialplumesizefromareaorvolume sourcesisallowed.

Chemical Transformation Chemicaltransformationsaregenerally nottreatedbyAERMOD.However, AERMODdoescontainanoptiontotreat chemicaltransformationusingsimple exponentialdecay,althoughthisoption istypicallynotusedinregulatory applications,exceptforsourcesofsulfur dioxideinurbanareas.Eitheradecay coefficientorahalflifeisinputbythe user.NotealsothatthePlumeVolume MolarRatioMethod(subsection5.1)and theOzoneLimitingMethod(subsection 5.2.4)andforpointsourceNO2analyses areavailableasnonregulatoryoptions.

Gasphasechemicaltransformationsare treatedusingparameterizedmodelsof SO2conversiontoSO4 2andNO conversiontoNO3

,HNO3.Organic aerosolformationistreated.The POSTUTILprogramcontainsanoptionto repartitionHNO3andNO3 inorderto treattheeffectsofammonialimitation.

PhysicalRemoval AERMOD can be used to treat dry and wet deposition for both gases and particles.

Drydepositionofgaseouspollutantsand particulatematterisparameterizedin termsofaresistancebaseddeposition model.Gravitationalsettling,inertial impaction,andBrownianmotioneffects ondepositionofparticulatematteris included.CALPUFFcontainsanoptionto evaluatetheeffectsofplumetilt resultingfromgravitationalsettling.Wet depositionofgasesandparticulate matterisparameterizedintermsofa scavengingcoefficientapproach.

1AppendixAtoAppendixWofPart51-SummariesofPreferredAirQualityModels,FederalRegister, Vol.70,No.216,November9,2005.