URI vs Rascal DOSE ASSESSMENT 4.2 vs Gaussian … 18_Dose Assessment... · TEDE CDE Thyroid TEDE CDE Thyroid SB (0.5 mi) 18.2 rem 297 rem 59rem 1600 rem 2 mile 1.44 rem 23.8 rem 4.7

URI vs Rascal URI vs Rascal URI vs Rascal URI vs Rascal

4.2 vs 4.2 vs 4.2 vs 4.2 vs

Gaussian vs Gaussian vs Gaussian vs Gaussian vs

MidasMidasMidasMidas

DOSE ASSESSMENTDOSE ASSESSMENTDOSE ASSESSMENTDOSE ASSESSMENT

�Unified RASCAL Interface

�RASCAL 4.2

�Simple Gaussian Model

�Midas

2016 National REP MI DEQ REP Unit

THE PLAYERS

�There are many assumptions made in dose

projections and the estimated range of

uncertainty between projected and actual

off-site doses for severe accidents is best

within a factor of 10, and at worst, off by

a factor of 100,000.

Sources:

�Nuclear Regulatory Commission

Regulation NUREG-1210

�FEMA IS 303


THE DISCLAIMER

�One of my NPPs finally started using Unified RASCAL Interface (URI) for their dose assessment (after much hype)

� I know from various meetings, trainings and conversations that URI uses the RASCAL engine

�But which one? Theoretically if you get the right one they should match exactly.

� I’ve got RASCAL 3.0.5, 4.1, 4.2… I think we’re up to 4.3.1 now


THE ASSUMPTION

� RASCAL 4.2

� URI/RASCAL Interface Description:

(RSCS “Unif ied RASCAL Interface – URI Ensuring Accuracy and Ease of

RASCAL Use for Dose Assessment Professionals” , Apri l 2012)

“This interface provides a specific overlay on the

existing RASCAL 4.2 meteorological, dispersion, and

dose assessment models…”

Leads to

Expectation: URI and RASCAL 4.2 should match closely


THE DISCOVERY


THE TEST

� Source Term: Clad

� TAS: 1:00

� Release Duration: 4:00

� NG: 1.00E+02 Ci/s

� Iodines: 7.45E+01 Ci/s

� Particulate: 3.87 Ci/s

� Wind Dir: 0 degrees

� Wind Speed: 5 mph

� Stability Class C

� Precipitation: None


A CLOSER LOOK

� We’ve exceeded PAGs out to 6.5 mi. It’s likely you’ll be receiving a PAR from the plant to evacuate a 5 mi radius and 10 mi downwind. Do you agree?


THE TYPICAL METHOD

� Typical Dose Assessor using RASCAL

� Effluent Releases – by Mixtures


THE TYPICAL METHOD

�Why?

�Because that’s what I got on my plant message form

� I got the release rates in Ci/s from the plant dose

assessor at the EOF

� I’m ahead of the game, I back-calculated the source

term release rate (Ci/s) from field team data

�Where do you get your data?


THE TYPICAL METHOD

�Putting in my stuff:


THE TYPICAL METHOD

�Ensuring matching components:


THE TYPICAL METHOD

�Same MET data:


THE RESULT

URIURIURIURI RASCAL 4.2RASCAL 4.2RASCAL 4.2RASCAL 4.2

TEDE CDE Thyroid TEDE CDE Thyroid

SB (0.5 mi) 18.2 rem 297 rem 59 rem 1600 rem

2 mile 1.44 rem 23.8 rem 4.7 rem 130 rem




LET’S SHADE THAT!








HOW FAR DOES IT GO?

� Somewhere in the 25 mi range. What now? Ad-hoc PARs?

URIURIURIURI RASCALRASCALRASCALRASCAL 4.24.24.24.2






15 mile 13 rem

20 mile 9.8 rem

30 mile 3.7 rem


WAIT, I HAVE A BACKUP METHOD!

�To the Gaussian!!

�Most dose assessors have an alternate

method available, either a homemade dose

modelling program or worksheet (aka

“napkin math”)


WAIT, I HAVE A BACKUP METHOD!

�Yikes! (I’m not even going to look how far that goes out)

� I think we exceeded the factor of 10 comparison

URIURIURIURI GaussianGaussianGaussianGaussian


SB (0.5 mi) 18.2 rem 297 rem 327 rem 1,085 rem



10 mile 0.187 rem 3.83 rem 1.08 rem 35.5 rem


SO WHAT WENT WRONG?

�Since URI uses RASCAL 4.2 dispersion and

dose engine, if everything is equal in the

inputs, the results should be nearly

identical.

�Suggestions?


THE SOLUTION

� Using “Effluent Releases - by Mixtures” is wrong

� Use “Effluent Release Rates – by Nuclide”


SURE, NO PROBLEM!

�Let me just get that data…

�Good thing this is a Clad accident (has the fewest isotopes)


INPUTTING, INPUTTING, INPUTTING


AND THE WINNER IS?








CLOSER, BUT NOT QUITE THERE YET

� One more thing to consider

� RASCAL defaults to a 10m “Release height”

� Many dose assessors either skip by this or

mistakenly think it’s the “stack height”

� The actual stack height for this plant is 58m

� Changing the “Release height” to 58m results in…

Note: RASCAL displays results in two significant

figures


THERE IT IS








NEXT STEPS

�How does this work with other dose

models (Midas, Rad Dose 5, older RASCAL

versions, newer RASCAL versions, etc?)

�Can you easily obtain the data necessary

to perform dose calculations this way?


MIDAS

Situation inputs for MIDAS:

TAS: 1 hour

Containment Bypass Release

Gap Release

Normal Filtration

Wind: From 270 deg

Wind Speed: 5 mph

Duration: 4 hour

NG: 1.0e+02 Ci/s

I-131 Equiv.: 7.2e-02 Ci/s

Particulate: 7.3e-03 Ci/s

Individual Isotope Release Rate Available µCi/s


MIDAS VS GAUSSIAN

MIDASMIDASMIDASMIDAS GaussianGaussianGaussianGaussian

Effluent Release Rates MethodEffluent Release Rates MethodEffluent Release Rates MethodEffluent Release Rates Method


SB (0.5 mi) 0.75 rem 3.2 rem 2.104 rem 10.48 rem





MIDAS VS RASCAL 4.2

MIDASMIDASMIDASMIDAS RASCAL 4.2RASCAL 4.2RASCAL 4.2RASCAL 4.2

Effluent Releases Effluent Releases Effluent Releases Effluent Releases ---- MixturesMixturesMixturesMixtures







MIDAS VS RASCAL 4.2

MIDASMIDASMIDASMIDAS RASCAL 4.2RASCAL 4.2RASCAL 4.2RASCAL 4.2

Effluent Releases Effluent Releases Effluent Releases Effluent Releases ---- NuclideNuclideNuclideNuclide






QUESTIONS?


� Document owner: Greg Gothard

� Contact: [email protected]

� Created: 12/22/2015

� Last Revision:

� Revision Number: 0

VERSION CONTROL


URI vs Rascal DOSE ASSESSMENT 4.2 vs Gaussian … 18_Dose Assessment... · TEDE CDE Thyroid TEDE CDE Thyroid SB (0.5 mi) 18.2 rem 297 rem 59rem 1600 rem 2 mile 1.44 rem 23.8 rem 4.7

Documents