coastal dispersion model reliability in accidental

Coastal dispersion model reliability in accidental situation (Fukushima, Japan), how to make the most of in-situ measurements?

Bailly du Bois P.1, Laguionie P.1, Garreau P.2, Theetten S.2

1 IRSN/PRP/SERIS/LRC Rue Max Pol Fouchet - BP10, Cherbourg-Octeville, 50130, France e-mail: [email protected]

2 IFREMER/DYNECO/PHYSED Technopôle de Brest-Iroise, Plouzané, 29280, France

e-mail: [email protected]

Coastal dispersion model reliability in accidental situation (Fukushima, Japan) - JONSMOD2012 2/21

Fukushima accident in marine environment conditions▌Accident in the Fukushima Dai-ichi nuclear power plant

represented the most important artificial radioactive release flux into the sea ever known.

▌Origins of the labelling:Atmospheric fallout onto the ocean, Direct release of contaminated water from the plantTransport of radioactive pollution by leaching through contaminated soil.

▌Principal radionuclides measured: 137Cs t1/2=30 years, 131I t1/2=8 days

▌ In the vicinity of the plant (less than 500 m), concentrations in early April reached 68 000 Bq.L-1 for 134Cs and 137Cs, and exceeded 100 000 Bq.L-1 for 131I

Measurement data sources: TEPCO, MEXT, Detection limits: more than 5Bq.l-1 (monitoring measurements)


Hoyashio

Kuroshio

General currents (Hycom simulation)

Hydrograhic domain

Bathymetry (JODC data )

Tsugaru strait


Atmospheric fallout

Estimation of total atmospheric release: 11.5 PBq of 137Cs.

IRSN estimation of cumulated 137Cs deposit at 80 km of distance from Fukushima Dai-ichi plant is 3 PBqInitial ratio (131I/137Cs) = 14)

Calculated cumulate 137Cs wet deposit, April 23, 2011

140.1 140.35 140.6 140.85 141.1 141.35 141.6 141.85 142.1 142.3536.2 36.3 36.4 36.5 36.6 36.7 36.8 36.9

37 37.1 37.2 37.3 37.4 37.5 37.6 37.7 37.8 37.9

38 38.1 38.2

137Cs wet deposit

Bq.m-2

Fukushima Dai-ichi

0 20 40 60 80 100 km

N0.0E+000

2.5E+005

5.0E+005

7.5E+005

1.0E+006

1.3E+006

1.5E+006

1.8E+006

2.0E+006

2.3E+006

2.5E+006

2.8E+006

3.0E+006

3.3E+006

3.5E+006

3.8E+006

4.0E+006

140.1 140.35 140.6 140.85 141.1 141.35 141.6 141.85 142.1 142.3536.2 36.3 36.4 36.5 36.6 36.7 36.8 36.9

37 37.1 37.2 37.3 37.4 37.5 37.6 37.7 37.8 37.9

38 38.1 38.2

137Cs dry deposit

Bq.m-2

Fukushima Dai-ichi

0 20 40 60 80 100 km

N0.0E+000

3.6E+003

7.2E+003

1.1E+004

1.4E+004

1.8E+004

2.2E+004

2.5E+004

2.9E+004

3.2E+004

3.6E+004

4.0E+004


Seawater 137Cs concentrations, 131I/137Cs ratio*

Seawater measurements at less than 2 km from the

Fukushima Dai-ichi

power plant.

Direct release average IR* around 20

Measurements at more than 5km from the coast.

IR* 5 -

30

1E0

1E1

1E2

1E3

Cae

sium

137

, Bq/

L

0

10

20

30

40

50

60

70

80

90

Iodi

ne 1

31 /

caes

ium

137

20/03 25/03 30/03 04/04 09/04 14/04 19/04 24/04 29/04 04/05 09/05 14/05 19/05 24/05 29/05 03/06 08/06 13/06 18/06 23/06 28/06

Caesium 137, Bq/L Ratio I131/Cs137, decay corrected

1E1

1E2

1E3

1E4

1E5

Cae

sium

137

, Bq/

L

0

10

20

30

40

50

60

70

80

Iodi

ne 1

31 /

caes

ium

137

20/03 27/03 03/04 10/04 17/04 24/04 01/05 08/05 15/05 22/05 29/05 05/06 12/06 19/06 26/06

Caesium 137 Ratio I-131/Cs-137, decay corrected

*initial ratio (IR): 131I/137Cs decay corrected to the 11 March 2011


Estimation of the direct release source-term into seawater137Cs quantities were estimated on the basis of individual measurements in a 50 x 100 km area around the plant (issues: depth, mixing layer, atmospheric fallout, rain water washout, …)

140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

April 11 - 18

Fukushima Dai-ichi

140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi

April 18 - 25140.8 141 141.2 141.4 141.6

April 25 - May 02

Fukushima Dai-ichi

140.8 141 141.2 141.4 141.6May 02 - 16

Fukushima Dai-ichi

a b c d

140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

Fukushima Dai-ichi

140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi

140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi

140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi

Blue frame: area for integration of 137Cs quantities : Sampling locations

e f g hMay 16 - 30 May 30 - June 13 June 13 - 27 June 27 - July 110 km 20 km 40 km 60 km

Interpolated137Cs

concentrations from April 11

to July 11


Estimation of the source-term from direct releases into seawater

Measurement period Number of measurements

accounted

137Cs quantity

Begin End Middle

11/04/11 18/04/11 14/04/11 92 11.6 PBq

18/04/11 25/04/11 21/04/11 77 4.75 PBq

25/04/11 02/05/11 28/04/11 118 3.38 PBq

02/05/11 16/05/11 09/05/11 293 0.67 PBq

16/05/11 30/05/11 23/05/11 233 0.26 PBq

30/05/11 13/06/11 06/06/11 227 0.16 PBq

13/06/11 27/06/11 20/06/11 250 0.04 PBq

27/06/11 12/07/11 04/07/11 202 0.002 PBq

Quantities of 137Cs deduced from interpolation of individual measurements in seawater.

Other estimations: 1 – 4PBq (TEPCO, Tsunume et al., Kawamura et al.)

Average mixing layer thickness deduced from hydrographic measurement: 32m

NERH, 2011. Report of Japanese Government to the IAEA Ministerial Conference on Nuclear Safety – The Accident at TEPCO’s Fukushima Nuclear Power Stations. NERH Report. www.kantei.go.jp/foreign/kan/topics/201106/iaea_houkokusho_e.htmlTsumune D., Tsubono T., Aoyama M., Hirose K., 2011. Distribution of oceanic 137Cs from the Fukushima Daiichi Nuclear Power Plant simulated numerically by a regional ocean model. Journal of Environmental Radioactivity Volume, Issues 0. DOI: 16/j.jenvrad.2011.10.007Kawamura H., Kobayashi T., Furuno A., IN T., Iishikawa Y., Nakayama T., Shima S., Awaji T., 2011. Preliminary Numerical Experiments on Oceanic Dispersion of 131I and 137Cs Discharged into the Ocean because of the Fukushima Daiichi Nuclear Power Plant Disaster. Journal of Nuclear Science and Technology Volume 48, Issues 11, Pages 1349-1356. DOI: 80/18811248.2011.9711826

PBq: 1015Bq


Estimation of the rate of seawater renewal

Environmental half-time

exponential decay

t1/2 = 6.9 d

Constant dilution by clean water through marine currents due to convergence of Kuroshio and Oyashio currents

Seasonal changes in the ocean circulation ?Return of contaminated water back in the area ?

If no more releases occur and dilution remains constant, back to prior situation could be expected in Nov 2011

Exponential decay back extrapolated to April 8: 22PBq

Main incertitude: depth of the mixing layer

Evolution of 137Cs quantities measured in seawater


Flux estimation of 137Cs from direct releases

This source-term could be used in numerical dispersion models

Evolution of concentrations and

corresponding fluxes of 137Cs in

seawater

Assumptions:• Measurements close to the plant are representative of the released flux.• Amount of 22 PBq corresponds to the quantity of 137Cs released from March 26 to April 8

(average concentration:15 716 Bq.L-1, number of values = 28, duration = 13.2 days).• Fluxes of 137Cs could be deduced from concentrations by applying the factor:

1.06E+11 Bq released / Bq.L-1 measured

Right Y axis in figure shows this conversion.

1E1

1E2

1E3

1E4

1E5

Cae

sium

137

, Bq/

L

1E12

1E13

1E14

1E15

1E16

Cae

sium

137

, Bq/

day

20/03 27/03 03/04 10/04 17/04 24/04 01/05 08/05 15/05 22/05 29/05 05/06 12/06 19/06 26/06 03/07 10/07 17/07

30 m North of Fukushima Dai-ichi

More details in: Bailly du Bois P., Laguionie P., Boust D., Korsakissok I., Didier D., Fiévet B. 2011. Estimation of marine source- term following Fukushima Dai-ichi accident. Journal of environmental Radioactivity. DOI:10.1016/j.jenvrad.2011.11.015

Total direct release until July 18 -> 27 PBq(12 PBq -> 41 PBq)


Hydrodynamic models in marine accidents

▌Coastal models are key components of operational forecast systems which could be deployed during crisis such as point- source pollution incidents.

▌Model reliability can be assessed with field-based data.

▌ In this context, artificial radionuclides are exceptional validation tools because they can be measured at all scales and source-terms are generally well known.

▌The MARS (Model for Applications at Regional Scale) model was used After Fukushima Dai-ichi accident to reproduce and forecast the behaviour of contaminated waters for direct releases and deposit in Pacific ocean.

▌Dissolved 137Cs measurements allowed assessing model reliability in this low energetic tidal environment (currents of 1 m.s-1), but strong general circulation (Kurushio and Oiashio currents).

Coastal dispersion model reliability in accidental situation (Fukushima, Japan) - JONSMOD2012

Crisis modelling for Fukushima marine AccidentConstraints :

High resolution to follow the radionuclide releases

Taking into account the regional an local circulation- Kuroshio current (strong density unstable current)- The flux trough the Tsugaru Strait- The wind forcing (possibility of typhoons)- The tides (relatively low)

Use the data available in the framework of a crisis

The oceanic forecast and hindcast of global model of MERCATOR-Ocean

- resolution about 8-10 km (1/12°) - one field per day (T, S, U, V, ssh)

The atmospheric forecast and hindcast of global model NCEP- resolution about 50 km(1/2°)

Tidal Harmonic component (16) atlas FES2004 (1/8°)

Coastal dispersion model reliability in accidental situation (Fukushima, Japan) - JONSMOD2012

The FUKU0 configuration of MARS3D code

Animation of the sea surface temperature:

ftp://ftp.ifremer.fr/ifremer/dyneco/pgarreau/FUKU/film_temp.gif

Animation of current (zoom):ftp://ftp.ifremer.fr/ifremer/dyneco/pgarreau/FUKU/film_tsugaru.gif

10 000 m depth

Tsugaru strait

Fukushima power plant

Extension : 31°N - 43.2°N , 137°E - 152°E (1000km*1200km)Resolution : about 1 nautical mile (1/60°)

: 40 sigma levels refined near the surface

OBC and IC : temperature, salinity, currents, sea level from Mercator-Ocean,combined with the tidal harmonics

Run : on 256 MPI ranks

ftp://ftp.ifremer.fr/ifremer/dyneco/pgarreau/FUKU/film_temp.gif

ftp://ftp.ifremer.fr/ifremer/dyneco/pgarreau/FUKU/film_tsugaru.gif


Model calculation of the mixing layer

Quantities of 137Cs simulated in the inventory area confirms the 32m estimation of the mixing layer for the computation of inventories.

-60

-50

-40

-30

-20

-10

0

Dep

th (m

eter

s)

0% 20% 40% 60% 80% 100%Cesium 137 inventory

Cs137 inventory in the surface layer Average thickness for calculation


Environmental half time

1E12

1E13

1E14

1E15

1E16

1E17

Dis

solv

ed c

esiu

m 1

37 in

the

area

20/03 30/03 09/04 19/04 29/04 09/05 19/05 29/05 08/06 18/06 28/06 Date

Simulated quantities Environmental half time of 17 daysMeasured quantities Environmental half time of 6.9 days

Model renewing time is lower than deduced from in-situ measurements

Differences between model and measurements concentrations will increase with time.It could explain why other model/measurement comparisons gives lower estimations of direct releases (Tsunume et al., Kawamura et al.). But estimation from Buesseler et al. based on other measurements confirms our order of magnitude for direct releases. Buesseler, K. O., Jayne S. R., Fisher N. S., Rypina I. I., Baumann H., Baumann Z. et al. 2012. Fukushima-derived radionuclides in the ocean and biota off Japan. PNAS 2012 109 (16) 5984-5988, doi:10.1073/pnas.1120794109.

Environmental half-times

6.9 days

17 days


Matching calculations / measurements

1E0

1E1

1E2

1E3

1E4

1E5

Bq/

L

28/03 04/04 11/04 18/04 25/04 02/05 09/05 16/05 23/05 30/05 06/06 13/06 20/06 27/06 04/07

Measurements Mars Simulation

Cs137

1E0

1E1

1E2

1E3

Bq/

L

28/03 02/04 07/04 12/04 17/04 22/04 27/04 02/05 07/05 12/05 17/05 22/05 27/05 01/06 06/06 11/06 16/06 21/06 26/06 01/07 06/07

Measurements Mars simulation

Cs137

Less than 1 km off the facility

More than 5 km off the facility


140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

137Cs in surface seawater, April 11 - 18

Fukushima Dai-ichi

Measurements : Sampling location

140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

Fukushima Dai-ichi

Simulation : Sampling location

140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

137Cs in surface seawater, April 25 - May 02

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

137Cs in surface seawater, May 02 - 16

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

137Cs in surface seawater, May 30 - June 13

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

Fukushima Dai-ichi


Matching calculations / measurementsApril, 14 April, 28

May, 9 June, 6


Atmospheric wet deposit

A large incertitude remain concerning parameters of wet deposit onto the sea. Results confirms the orders of magnitude used in simulation of deposit (vd=5.10-5 m/s, lambda=5.10-5 hr.mm-1.s-1)

137Cs in surface seawater, March 22

140.8 141 141.2 141.4 141.6

37

37.2

37.4

37.6

37.8

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi

02468101215192329364556698610013016020025031039048060074092011001400170021002700

Bq.L-1

137Cs in surface seawater, April 7

140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi


140.8 141 141.2 141.4 141.6

Fukushima Dai-ichi

Coloured spots: measurementsShaded background: simulation

: Sampling locationColoured spots: measurementsShaded background: simulation



: Sampling location

End of the atmospheric deposit: 16 March

After 28 March: influence of the direct release


Model animation

Color-scale changed by a factor 10


Discussion

• Association between simulations and measurements and early measurements are essential to estimate source terms.

• Why simulated renewing time is too long: wind shear stress or open condition limits?

• Studies in the English Channel and North Sea show that the wind drag coefficient is a key parameter, depending of the meteorological data applied. This question will be examined in more details.

• Before March 11 2011, direct releases to the sea was not accounted for possible nuclear power plant accidents. Generally marine systems are not the first emergency in such situations, but answers must be prepared and other situations could end up to dispersion of radionuclides (ship wreck).

• Stochastic turbulence in the Fukushima area, measurements available and poor knowledge of the source term made difficult to investigate exact location of the plume following releases.


Conclusions: radionuclides

• Contamination of the marine environment following the accident in the Fukushima Dai-ichi nuclear power plant represents the most important artificial radioactive direct liquid release into the sea ever known (arround 27 PBq for 137Cs only).

• Available measurements made possible to estimate the environmental half-time of the release area, and the radionuclide fluxes from direct releases.

• Due to the exceptional dilution conditions, concentrations in seawater decrease strongly at the end of 2011.

• It will be different for benthic species living close to the plant or concerned by transfer through suspended matter coming from deposited sediments, rivers or desorption in seawater of radionuclides fixed onto sediments.

• Caesium 137 and 134 from Fukushima will remain detectable for several years in the North Pacific. 137Cs/134Cs ratio could be used as a tracer for studies on water masses pathways and transit times.


Conclusions: hydrodynamic models

• Hydrodynamic models are essential tools to approach consequences of accidents in marine environment. It is our duty to apply them, even very far from Europe.

• Simulations during accidents are useful to : organize in-situ intervention and sampling, manage living resources and human uses of marine system, information of stakeholders and populations.

• Models made possible dynamic calculation of transfers to living species by using biological half-lives of radionuclides (if available).

• Measurements of dissolved radionuclides made possible to approach Mid- and long-term precision of hydrodynamic models.

• Measured and simulated process are comparable, some key parameters as renewing time could be improved.

• Even if few early data are available, useful orders of magnitude for atmospheric deposit parameters are provided by simulations.


Special thanks to:

MEXT and TEPCO for the Internet publication of the concentration data collected in the environment.

Celine Duffa, Mireille Arnaud, Olivier Connan, Vanessa Parache, Tina Odaka and “Centre Technique de Crise” staff of the French Institute of Radioprotection and Nuclear Safety for searching compiling and checking thousands of measurements.

We want to express all our sympathy to Japanese people.


Inventory of main 137Cs releases in the oceans

Source PBq (1E+15Bq)

Fallout from nuclear weapon tests 704*

Nuclear fuel reprocessing 1951 - 2010

BNFL Sellafield 41.21

Areva-NC La Hague 1.04

Tchernobyl accident 16*

Solid waste dumping 41.4 (total beta + gamma)

Fukushima accident

Direct liquid release 27

Atmospheric fallout 3 (80 km radius circle)

* Aarkrog, 2003. Input of anthropogenic radionuclides into the World Ocean. Deep Sea Research Part II: Topical Studies in Oceanography Volume 50, Issues 17-21, Pages 2597-2606

coastal dispersion model reliability in accidental

Documents