Ryk Kłos, Aleksandria Sciences, Sheffield UK Anders Wörman, KTH, Stockholm. SSM, Stockholm 19 th November 2013

Transcription

1 Ryk Kłos, Aleksandria Sciences, Sheffield UK Anders Wörman, KTH, Stockholm SSM, Stockholm 19 th November 2013

2 Background SSM s Review of the SR-Site License Application Radionuclide transport model in the dose model Hydrological modelling MIKE-SHE (R-10-02) Parameterisation and parameters Ecosystem description (TR-10-01) Dose Assessment model Application (TR-10-06) This meeting Requests for information What we need to be sure we understand The most efficient way of communicating Thanks for coming Ecosystems TR-10-01, TR-10-02, TR Hydrological modelling R Dose assessment model TR Landscape TR Elementspecific TR-10-07

3 Our understanding Concept Hill with high head Lake / wetland balance Average object Dispersion Advective transport F AsubCatch 1 f P E flood F AsubCatch P E loss F AwaterShed P E loss V F loss res Assessment model Advective fluxes Ter. Rego upper Ter. Rego mid 20 FaquMid AsubCatch f P E aqumidup aquup 1 fmire A v obj LowMid 19 F AsubCatch f flood P E 14 aquup FaquUp AsubCatch faqumidup P E f Aobjv mire LowMid FterMid AsubCatch f P E Aqu. Rego termidup upper Aqu. Rego mid 21 F AsubCatch f P E aqumidup FaquUp AsubCatch faqumidup P E 9 aqumid 12 Downstream 1 FLow Aobj fmirevlowmid termid 2 FLow Aobj 1 f v mire LowMid aqumid Rego low

4 Modelling Typical Swedish Lake

5 Themes The average object Development of the radionuclide transport model Derivation of hydrological parameters that drive the radionuclide transport model Ecosystems TR-10-01, TR-10-02, TR Dose assessment model TR Landscape TR Elementspecific TR Hydrological modelling R-10-02

6 Where do the numbers come from? How are they used in the model? How are they justified?

7 MIKE-SHE - Basis for SR-Site Hydrology MIKE-SHE in R Network of independent basins results were extracted and delivered to the dose calculations Tool for defining mass balance in compartment models

8 balance Dose assessment model (TR-10-06) Radionuclide transport model Results from MIKE-SHE (R-10-02) interpreted in TR The Average Object Six lake/mire objects in present day terrestrial landscape Treated as a sample of future objects Snapshot at 5000 CE Other times are available Mass balance using advective velocities aka area normalised flows Does MIKE-SHE describe agriculture land?

9 The average object Six present day lake/mire objects because the MIKE-SHE model can be verified Treated as a statistical sample What is the output from MIKE-SHE? How are the numerical values derived/combined? Are mass balance schemes available for each basin in the landscape? Are they available for the different times? Obj 5000 CE Obj 5000 CE

10 Implicit mass balance Average object Does it add up? geosphere subcatchment Ter_regoLow Ter_regoMid Ter_ Aqu_regoLow Aqu_regoMid Aqu_ Atm Downstream geosphere sub-catchment Ter_regoLow Ter_regoMid Ter_ Aqu_regoLow 6 9 Aqu_regoMid Aqu_ Atm Upstream Inflow Outflow Balance % difference 100.0% 10.0% 0.5% 0.1% 20.0% 0.2% 0.3% 100.0% 100.0%

11 How do they relate to the average object? How are they used in the radionuclide transport model? How is object evolution accounted for?

12 Evolution of the hydrological description R TR TR Structures: R TR The parameterisation: TR TR What is the reasoning behind this? Implementation in TR There are lots of logic-switches Is the average object hydrology universally applicable? Can we see a full description of the code Appendix 1 of TR as pseudocode? if ( have_water AND time_ge_threshold_start ) area_subcatch runoff (1.0 + Flooding_coef ) / ( Ter_area_obj Ter_poro_regoUp Ter_z_regoUp Ter_R_regoUp ) else 0.0 end

13 1 F A f P E subcatch flood F A P E loss subcatch F A P E loss watershed F loss V res Ter. Rego upper F A f P E 7 1 fmire A v F A f P E termid subcatch flood subcatch termidup F A f P E aquup subcatch aqumidup obj LowMid Aqu. Rego upper 21 F A f P E subcatch aqumidup aquup Downstream Ter. Rego mid F A f P E aqumid subcatch aqumidup aquup 1 F A f v Low obj mire LowMid termid 1 f A v mire obj LowMid 9 12 Aqu. Rego mid 2 F A 1 f v F A f P E aquup subcatch aqumidup aqumid Low obj mire LowMid aqumid Rego low

14 Evolution of objects in the landscape generic Flux from to Expression F Low termid F Low aqumid F termid F Low termid Aobj fmire v LowMid Low aqumid 1 A f v obj mire LowMid termid A f P E subcatch termidup A 1f P E subcatch flood Date CE parameter Units A m 2 ter A aqu Aobj Ater Aaqu m m A m 2 subcatch A m 2 watershed v Low Mid Object 116 m a F F aqumid aquup F aquup aqumid F aquup F aquup F loss A f P E aqumid aquup subcatch flood A f P E subcatch aqumidup 1 f A v mire obj LowMid aquup aqumid A f P E aquup subcatch aqumidup A f P E subcatch aqumidup 1 f A v mire obj LowMid aquup A f P E subcatch aqumidup Downstream A P E watershed f termid f aqumid aquup P E m a f mire f flood Evolution via areas Relative fluxes constant for all objects

15 Numerical values ~ TR The six flows: Upwards velocity out of lower regolith: Adv_low_mid = v Low Mid Fraction of flow from lower regolith directed to mire: fract_mire = f mire Net precipitation: Runoff = P E Fraction of infiltration to catchment moving laterally in terrestrial subsystem: Ter_adv_midup_norm = f termid Fraction of infiltration to catchment moving laterally in aquatic subsystem: Aqu_adv_midup_norm = f aqumid aquup Fractional lateral flux from subcatchment to wetland: flooding_coef = f flood Object specific or not? Ter. Rego upper Ter. Rego mid Hill with high head Dispersion F AsubCatch 1 f flood P E F AsubCatch f flood P E 1 F A f v Rego low FaquUp AsubCatch faqumidup P E 7 1 FterMid AsubCatch ftermidup P E Low obj mire LowMid termid fmire AobjvLowMid Aqu. Rego upper F A f P E aqumid subcatch aqumidup aquup 1 f A v mire obj LowMid 9 12 Aqu. Rego mid 2 FLow Aobj 1 fmire vlowmid aqumid Advective transport 21 F A f P E subcatch aqumidup aquup F A f P E aquup subcatch aqumidup aqumid Lake / wetland F AsubCatch P E loss F A P E loss watershed Downstream V F loss res

16 Adv_low_mid = vlow Upward flux from the lower regolith Vertical flux (internal) v Low Mid External to the object v Low Mid Mid outflow from Inflowto lower regolith lower regolith vterlow vaqulow vtermid v aqumid termid aqumid terlow aqulow External inflowto Downstreamloss lower regolith fromlower regolith vsubcatch vgeo vgeo v terlow terlow terlow aqulow Downstream geosphere geosphere subcatchment Ter_regoLow Ter_regoMid Ter_ Aqu_regoLow Aqu_regoMid Aqu_ Atm sub-catchment Ter_regoLow Ter_regoMid Ter_ Aqu_regoLow 6 9 Aqu_regoMid Aqu_ Upstream Atm Inflow Outflow Balance % difference 100.0% 10.0% 0.5% 0.1% 20.0% 0.2% 0.3% 100.0% 100.0% Downstream

17 The normalised fluxes ~ the drivers Ter_adv_mid_up_norm = f termid f termid net flux through termid total water flux in basin jtermid v termid j itermid v i termid total water fluxout of basin v v v v v v v v termid termid termid ter aqumid ter aqumid terlow termid termid ter termid Low Loss Loss Loss ? Aqu_adv_mid_up_norm = f aqumid f v v v aqu termid aqumid aqumid aqumid aquup vter vtermid downstream downstream v aqulow aqumid terlow downstream aquup

18

19 RFI summary - 1 MIKE-SHE MIKE-SHE results with SDM-Site, pre-modelling and regional model areas are they the same as far as the average object is concerned? How were the mass balance schemes to the six lake/mire objects at 5000 CE combined to give the average object fluxes? Can we have access to the mass balance schemes for the six objects at the three times? Deeper access to flow fields in SICADA? What is the normalising area? Does the input from the bedrock change on transition from aquatic to terrestrial conditions?

20 RFI summary - 2 Translating the average object into the dose model When was the structure of the hydrological fluxes in the radionuclide transport model decided? Can a rationale for the changes in structure be presented? Can we see a detailed, step-by-step derivation of the of the numerical values for the six constant hydrological parameters? Implementation in the dose modelling Can we see the coding of the dose model as used? Our interest is the translation to the average object to situations like those in Object 121_03 Why was the average object approach used rather than using the output from MIKE-SHE for each of the basins?

21