Home

sedFlow - User manual (Version 1.00)

1. 13 6 3 ReducedWaterEmergyslope ln 13 6 4 ReturnBedslope and ReturnWaterEnergyslope 13 7 tauCalculationMethod realisations 13 8 thresholdCalculationMethod realisations 13 8 1 Constant ThresholdForInitiationOfBedloadMotion 13 8 2 LambEtAlCriticalTheta 22 l n 13 9 hidingFactorsCalculationMethod realisations 13 9 1 PowerLawHidingFunction lll 1329 2 NoHiding 2 0 8 24a kan RE Rond 13 9 3 WilcockCroweHidingFunction 2 2 20 13 10estimate ThicknessOfMovingSedimentLayer realisations 13 10 1 Constant ThicknessOfMovingSedimentLayer 13 10 2 MultipleDiameterOfCoarsestGrainMoved 13 10 3 MultipleReferenceGrainDiameter 13 11additionalMethods with Sternberg abrasion III Appendix Notation Lists of Tables and Figures References 56 56 58 60 Part I Quick start guide 1 General remarks The new model sedFlow has been developed to provide an efficient tool for the simulation of bedload transport in mountain streams The following elements were important for the development of sedFlow i provision of a sediment transport model together with its complete source code open and free of charge ii consideration of state of the art approaches for the calculation of bed load transport in steep channels accounting for macro roughness effects iii individual calculation for several grai
2. _realisationT ype LambEtAlCriticalTheta minimumCriticalDimensionlessShearStress 0 03 correctionForBedloadWeight AtSteepCounterSlopes true hidingFactorsCalculationMethod realisation Type is lt saa PowerLawHidingFunction referenceDiameterPercentile ee eee eee ee 50 0 exponents in sas Gets Eri ane seno viduae de dane Gees 0 8 strataSorting realisationType TwoLayerWithShearStressBasedUpdate dynamicLayerThickness 00 ce cece eee ee eens false uselInitialGrainSizesForConstant LayerThickness true layerThicknessFactor 1 75 referenceGrainSizePercentile 00 e eee eee ee 84 0 Lo laver Thickness dann ees only needed if neither dynamicLayerThickness nor useInitialGrainSizesForConstantLayerThickness dynamicBreakUpConditions 200 e eee eee false 56 Predehned Brea K UpCondiHOnS false referenceMedianDiameter only needed if usePredefinedBreakUpConditions _thetaCriticalFor ActiveLayer only needed if usePredefinedBreakUpConditions thetaCriticalForSublayer only needed if usePredefinedBreakUpConditions thresholdCalculationMethod method from bedloadTransportEquations or default displayed there additionalMethods 0 0 0 cece eee eee optional L outputMethods Figure 8 riverSystem
3. List CONDO dg WN Hm of Tables Structure of BranchTopology txt 00 6 Structure of BranchTopology txt for river network example of BiG 2a uc he eae ee A a a A a a 7 Structure of BranchTopology txt for main channel with stub tributaries example of Fig 2b leen 7 Minimum structure of BranchXProfileetxt 8 Structure of grain size distribution spreadsheets 9 Structure of BranchXDischarge txt 00 9 Complete structure of BranchXProfile txt part 1 13 Complete structure of BranchXProfile txt part 2 including default Values ood due e Baral ge RR oR p 14 Structure of SedimentInputs txt part 1 14 Structure of SedimentInputs txt part 2 including default values 15 Structure of sedigraph spreadsheets 2 222 220 15 Structure of InstantaneousSedimentInputs txt part 1 16 Structure of InstantaneousSedimentInputs txt part 2 includ ing default values 22e 16 Structure of Sills txt including default values 18 of Figures Minimum structure for simulation folder 6 Examples of potential BranchTopology 7 Minimum Sa 4 0 8 2 ey R ae Hs se gee dene Seis oV 10 Minimum xml with recommended values 11 Complete structure for simulation folder 12 overallParameters including default values 19 riverSystemMethods part 1 including default values 22 riverSystemMethods par
4. L 17d Dso OcgA o 14 pe for 034 gt L 17e In this 0 4 is 0 for Dg4 0g4 is 0 for Dg4 and L is a process break point bedloadTransportEquations realisation Type sss ReckingBedloadCapacityNonFractional Figure 23 ReckingBedloadCapacityNonFractional 13 4 bedload Velocity CalculationMethod realisations 13 4 1 VelocityAs TranpsportRatePer Unit CrossSectionalArea The VelocityAs TranpsportRatePer UnitCrossSectionalArea Fig 24 estimates bedload velocity vy according to the following simple relation in which hg is the thickness of the moving sediment layer qb Up 77 18 ha Al The method for the estimation of the thickness of the moving sediment layer is defined by the optional node estimate ThicknessOfMovingSediment Layer with its default realisation Constant ThicknessOfMovingSedimentLayer section 13 10 1 bedload VelocityCalculationMethod realisationTypeVelocityAsTranpsportRatePerUnitCrossSectionalArea estimateThicknessOfMovingSedimentLayer PB realisationType ConstantThicknessOfMovingSedimentLayer Figure 24 VelocityAsTranpsportRatePerUnitCrossSectionalArea including default values 13 4 2 JulienBounvilayRollingParticles Velocity The JulienBounvilayRollingParticles Velocity Fig 25 estimates bedload ve locity according to the following equations from Julien and Bounvilay 6 T B as 19a ps p g a Dz ww Vg rn S 3 3 1nd 17 7 19b In this 6 is 0 for the ro
5. the two profiles should have a distance as described before In the same way the lowermost reach of the complete system should also have an adequate length usually defined by the distance of its cross section to a kilometrage of zero ElevationInM defines the river bed elevation of the cross section at the upstream end of the reach Implicitly this also defines the slope of the reach When using a kinematic wave routing it is important to make sure that no adverse i e negative slopes occur as they would lead to kryptic artefacts The slope of the lowermost reach of the system is set equal to the slope of the next upstream reach in front of it ChannelWidthInM defines the width of the infinitely deep rectangular channel StrataGrainSizeDistribution defines the file names without txt exten sion of the grain size distributions of the alluvium to be used at the start of a simulation Of course individual file names may appear repeatedly within this column 6 Grain size distributions For each grain size distribution defined in the BranchXProfile tat files the code will try to find the corresponding file Tab 5 in the GrainSizeDistri butions folder The column GrainDiameterInCM usually gives the upper boundaries for the individual grain size fractions These values need to be sorted in increasing order and need to be identical in all grain size distribu tion files The column RelativeAbundance gives the relative abundance of the
6. which always has the same structure Fig 1 my WhateverSimulation myWhateverInput xml LongitudinalProfile BranchTopology txt BranchXProfile txt GrainSizeDistributions myWhateverGrainSizeDistribution txt DischargeAndOtherInputs BranchXDischarge txt Output Figure 1 Minimum structure for simulation folder 4 BranchTopology tzt First the code will go into the folder LongitudinalProfile and look for the file Branch Topology tat Within sedFlow the river network is organised in branches A branch is a section of a river without any tributaries or conflu ences Different branches may be connected by confluences The topology of the river network is defined by the Branch Topology tzt The file contains the two columns BranchIDs and DownstreamBranchIDs Tab 1 in which the IDs are given as integer numbers It is required that a branch can only flow into another branch with a higher ID Examples are given in Figure 2 and Tables 2 and 3 Table 1 Structure of BranchTopology txt BranchIDs DownstreamBranchIDs 22 22 Figure 2 Examples of potential BranchTopology of a a river network and b a main channel with stub tributaries Table 2 Structure of BranchTopol Table 3 Structure of BranchTopol ogy txt for river network example of ogy txt for main channel with stub Fig 2a tribut
7. which is used by default in the setup of Figure 3 11 Part II Further specifications The Quick start guide on the previous pages described the minimum amount of information which is needed to run a simulation In such a case many simulation parameters are set to their default values The user can override these default values simply by adding further files to the simulation folder or further columns to the spreadsheet files or further nodes to the input xml file It is important to note that complete columns need to be given Even if only one value differs from the default the complete column needs to be input with the defaults given explicitly and the single other value If default values are not given and the corresponding rows have therefore less entries than the header line these rows will be ignored completely my WhateverSimulation myWhateverInput xml LongitudinalProfile BranchTopology txt BranchXProfile txt Sills txt GrainSizeDistributions myWhateverGrainSizeDistribution txt DischargeAndOtherInputs L BranchXDischarge txt SedimentInputs txt myWhateverSedigraph txt InstantaneousSedimentInputs txt Output Figure 5 Complete structure for simulation folder The complete structure for the simulation folder including all possible input files is displayed in Figure 5 The complete structure of the BranchX Profile trt spreadsheet files including all possible columns is de
8. 14 is marimum TimeStep with its default value of 900 which defines the maximum time step length that will not be exceeded during this simulation It is recommended that this value should be smaller than the time step length of the input time series e g discharge waterFlowRouting realisation Types aaa cereo bx Duane Ple dh e N is UniformDischarge maximum TimMeStep s icc dseedid anna 900 Figure 14 UniformDischarge including default values 13 2 2 ExplicitKinematic Wave The ExzplicitKinematic Wave Fig 13 2 2 has no specific nodes waterFlowRouting realisationType 2 2222c220000 ExplicitKinematicWave Figure 15 Explicit KinematicWave 35 13 2 3 ImplicitKinematic Wave The ImplicitKinematicWave Fig 16 performs an implicit kinematic wave routing using the algorithms of Liu and Todini 8 Just like UniformDis charge the ImplicitKinematic Wave has the optional node mazimum TimeStep with its default value of 900 Additionally it has the optional node check ForCourantFriedrichsLewy with its default value false which can be used to switch on a test for the Courant Friedrichs Lewy criterion based on the water flow velocity waterFlowRouting realisation Ype ImplicitKinematicWave maximum Timestep x x 0 00 c cece cece n eee nne 900 checkForCourantFriedrichsLewy eese false Figure 16 ImplicitKinematicWave including default values 13 3 bedloadTransportEquations realisations All b
9. 48 49 50 50 50 51 51 51 52 52 53 53 53 54 54 55 55 55 References 1 2 3 4 wo 5 6 7 8 9 10 A Badoux and D Rickenmann Berechnungen zum Geschiebetransport wahrend der Hochwasser 1993 und 2000 im Wallis Wasser Energie Luft 100 3 217 226 2008 R I Ferguson Flow resistance equations for gravel and boulder bed streams Water Resour Res 43 W05427 2007 doi 10 1029 2006WR005422 F U M Heimann J M Turowski D Rickenmann and J W Kirchner sedflow an efficient tool for simulating bedload transport bed rough ness and longitudinal profile evolution in mountain streams Earth Surface Dynamics submitted F U M Heimann D Rickenmann M B ckli A Badoux J M Tur owski and J W Kirchner Recalculation of bedload transport observa tions in swiss mountain rivers using the model sedflow Earth Surface Dynamics submitted M N R J ggi Sedimenthaushalt und Stabilit t von Flussbauten In D Vischer editor Mitteilung der Versuchsanstalt f r Wasserbau Hydrologie und Glaziologie number 119 pages 1 105 ETH Zurich Switzerland 1992 P Y Julien and B Bounvilay Velocity of rolling bed load particles J Hydr Eng 139 2 177 186 2013 doi 10 1061 ASCE HY 1943 7900 0000657 M P Lamb W E Dietrich and J G Venditti Is the critical Shields stress for incipient sediment motion dependent on channel bed sl
10. Alternative methods may be implemented in future versions of sedFlow The waterFlowRouting node defines the way in which water is trans ferred through the channel system and how the flowResistance relation is applied to calculate flow depths and velocities Different realisations are described in section 13 2 The ExplicitKinematic Wave section 13 2 2 and ImplicitKinematic Wave section 13 2 3 depend on positive bed slopes The ImplicitKinematic Wave section 13 2 3 further depends on the FiredPow erLawFlowResistance flow resistance section 13 1 1 the approximation of the hydraulic radius by flow depth section 13 1 and on an infinitely deep rectangular channel which is given anyway The UniformDischarge sec tion 13 2 1 has no dependencies and replaces the discharge routing approach by a uniform discharge approach The node bedloadTransportEquations defines the way in which the bed load transport capacity is estimated Different realisations are described in section 13 3 Fractional transport as well as representative single grain size approaches can be selected However single grain size approaches do not imply an increase in calculation speed as the model framework is optimised for fractional approaches The node strataSorting defines the interaction between the surface active layer and the subsurface alluvium Different realisations are described in section 13 5 If the node strataSorting is not given its realisation is set to T
11. In this S is the energy slope corrected for the effects of bedload weight at steep bed slopes S is the uncorrected slope is the angle of repose and Sp is the bed slope For the fractional versions Figs 18 amp 20 the node hidingFactorsCal culationMethod defines the hiding function to be used The default hiding function is the PowerLawHidingFunction section 13 9 1 For the NonFractional versions Figs 19 amp 21 the switch takeArmourLay erIntoAccount selects whether the following correction of the discharge thresh old for the initiation of bedload motion according to Badoux and Rickenmann 1 should be applied or not default Sy 14 38 By ME QcA Qc 2 15 DmArith In this qc is the threshold discharge per unit flow width for the initiation of bedload motion q A is qe corrected for the effect of bed armouring and DymArith the arithmetic mean diameter of the local grain size distribution bedload Transport Equations realisationType RickenmannBedloadCapacityBasedOnTheta useOnePointOneAsExponentForFroudeNumber false simplifiedEquation nen true thresholdCalculationMethod realisation Ype LambEtAlCriticalTheta hidingFactorsCalculationMethod ee Em PowerLawHidingFunction thetaCriticalBasedOnConstantSred 0 0 ccc ee eee false Figure 18 RickenmannBedloadCapacityBasedOnTheta including default values bedload Transport Equations realisat
12. StrataGrainSize Distribution for the definition of the initial SurfaceLayerGrainSize Distribution Bedrock RoughnessEquivalentRepresentativeGrainDiameterInCM defines the flow resistance and hiding properties of a reach when the river runs over bedrock and there is no grain size distribution left to define the prop erties of the reach For example if a flow resistance relation is based on the 84 percentile grain diameter Dg4 the values in this column will be used as Dg for this flow resistance relation It is advisable to select values close to the measured grain diameters in this area If the alluvium cover gets thin the local grain sizes will tendentially approach the value of the bedrock roughness If the column BedrockRoughnessEquivalentRepresenta tiveGrainDiameterInCM is not given the code will use the Dga of the grain size distributions given in the SurfaceLayerGrainSizeDistribution Table 7 Complete structure of BranchXProfile txt part 1 KilometrageUpstreamDirected ElevationInM ChannelWidthInM AlluviumThicknessInM 22 22 TU 4000 0 13 Table 8 Complete structure of BranchXProfile txt part 2 including default values Strata SurfaceLayer BedrockRoughnessEquivalent GrainSizeDistribution GrainSizeDistribution RepresentativeGrain DiameterInCM 292 file name of Da of StrataGrainSizeDistribution SurfaceLayerGrainSizeDistribution 10 Sediment input 10 1 Tr
13. course in this realisation the real layer thickness corresponds to the current thickness of the alluvium The value of the node layerThickness de termines the alluvium thickness below which the bedrock starts to influence flow resistance and hiding processes The realisation SingleLayerNoSorting is intended for non fractional bedloadTransportEquations section 13 3 strataSorting realisation ID SingleLayerNoSorting layerT hicknegs2 eec eter Rey REP edd ER E C TE Dein 0 4 Figure 29 SingleLayerNoSorting including default values 13 6 Gradient calculation method realisations In section 8 the realisation types NoFlowResistancePartitioning and With FlowResistancePartitioning were introduced These pseudo realisation types are short hands for realisations which are consistent with the current water FlowRouting section 13 2 If a kinematic wave flow routing is selected the NoFlowResistancePartitioning will expand to ReturnBedslope section 13 6 4 46 and the WithFlowResistancePartitioning will expand to Reduced WaterEner gyslope section 13 6 3 If UniformDischarge section 13 2 1 is selected as waterFlowRouting the NoFlowResistancePartitioning will expand to Simple Downstream TwoCellGradient section 13 6 1 with hydraulicHead as prop ertyOfInterest and the WithFlowResistancePartitioning wil expand to Re duced WaterEnergyslope Not Using WaterEnergyslope Variable section 13 6 3 with SimpleDownstream TwoCellGradient as simple
14. flow resistance relation with 1 6 5 g D 13 1 2 VariablePowerLawFlowResistance The VariablePowerLawFlowResistance Fig 13 implements the following equation from Ferguson 2 with the parameter values by Rickenmann and Recking 14 and has no specific additional nodes ids 6 5 2 5 fe 6 V 5 2 9 3 6 52 2 52 Fu 34 flowResistance realisation Type VariablePowerLawFlowResistance Figure 13 VariablePowerLawFlowResistance 13 2 waterFlowRouting realisations It has to be noted that the UniformDischarge approach section 13 2 1 in its default combination with some sediment energy slope based on hydraulic Head section 13 6 is based on the assumption that the simulated system only consists of the two extreme cases of pondages friction slope approxi mately zero on the one hand and situations of parallel slopes friction slope approximately equal bedslope on the other For details please refer to Heimann et al 3 It will produce large errors when intermediate cases of moderate backwater effects are part of the simulated system In such systems a KinematicWave approach which uses bedslope both as friction slope for the hydraulic and as energy slope for the sediment transport calcu lations will produce better estimates of the transported sediment volumes but requires the absence of adverse channel gradients 13 2 1 UniformDischarge The only optional node of UniformDischarge Fig
15. individual fractions not cumulative The abundances are normalised by the code Therefore they do not need to sum up to 1 or 100 Table 5 Structure of grain size distribution spreadsheets GrainDiameterInCM RelativeAbundance 77 77 7 BranchXDischarge txt The branch topology defines those branches for which no upstream branch is available For these headwater branches the code tries to find the BranchX Discharge txt files Tab 6 within the DischargeAndOtherInputs folder with X substituted by the corresponding branch ID The column ElapsedSeconds defines the points in time at which a discharge value is given Of course these values should increase The column DischargeInM3PerS defines the corresponding discharge values in m Between the points of this discrete time series the discharge values are interpolated linearly Beyond the range of this time series discharge is constantly set equal to the first or last discharge value respectively If one wants to feed water and sediment at a certain point to the system it is recommended to introduce a stub branch consisting of just one reach and give a Branch X Discharge tzt for it Table 6 Structure of BranchXDischarge txt ElapsedSeconds DischargeInM3PerS LT TT 8 The main input xml Most of the simulation data is defined in the spreadsheet files Most of the simulation parameters will be defined in the main input xml the location of whi
16. reaches or the value for each reach is calculated according to the local grain size distribution at the beginning of the simulation The mandatory node dynamicLayerThickness selects whether the active layer thickness should be constant in time or dynamically updated If the value of dynamicLayerThickness is false the node uselnitialGrainSizesFor ConstantLayer Thickness is needed to select whether the constant thickness should be the same for all reaches or defined according to the initial local grain size distribution If both the dynamicLayerThickness and the useIni tialGrainSizesForConstantLayer Thickness are false the active layer thick ness is defined by the node layerThickness If only one of the two nodes is true the local active layer thickness is defined as b D with b defined by the node layerThicknessFactor and x defined by the node referenceGrain SizePercentile 44 strataSorting _realisationType TwoLayerWithCont inuousUpdate dynamicLayerThickness 2222ccsnessesnernenne false uselnitialGrainSizesForConstantLayerThickness true layerThicknessFactor 0 cece eee eee ee eee 1 75 referenceGrainSizePercentile 0 84 0 _layerThickness only needed if neither dynamicLayerThickness nor uselnitialGrainSizesForConstantLayerThickness Figure 27 TwoLayerWithContinuousUpdate including default values 13 5 3 TwoLayerWithShearStressBased Update The TwoL
17. resistance in gravel bed rivers through a large field data set Water Resour Res AT W07538 2011 doi 10 1029 2010WR009793 H Sternberg Untersuchungen ber langen und querprofil geschiebe f hrender fl sse Zeitschrift f r Bauwesen 25 483 506 1875 P R Wilcock and J C Crowe Surface based transport model for mixed size sediment J Hydr Eng 129 2 120 128 2003 doi 10 1061 ASCE 0733 9429 2003 129 2 120 61
18. the corresponding sill entry is ignored The SillTopEdgeElevationInM gives the absolute elevation of the top edge of the sill It should equal the elevation of the corresponding reach plus the elevation drop of the sill Alternatively the column SillDropHeight InM can be given In this case the elevation of the sill top edge is calculated as the sum of the drop height and the elevation of the corresponding reach If both columns are given the values from the SillTopEdgeElevationInM are used Please note that the top edge elevation is fixed while the drop height may change in the course of a simulation as the crosssection at the sill may experience erosion or accumulation of sediment The sill hydraulics are calculated according to the so called Poleni equation 2 Q 5 H Dg whe 1 In this Q is discharge g is gravity acceleration w is flow width and h is hydraulic head at the sill The weir coefficient u is defined by the column PoleniFactor If this column is not given its values are set to i by default The way in which the bed elevation upstream of a sill is updated may be defined by node upstreamOfSills WedgeShapedInsteadOfParallelUpdate in the main input xml For details see section 12 2 12 Complete structure of the main input xml file The structure of the main input xml as described in section 8 represents only a minimum to run a sedFlow simulation The following sections describe all 17 Table 14 Structure of Sills txt i
19. the way in which the bed elevation upstream of a sill section 11 is updated If the switch is false the model assumes that erosion and deposition are equally distributed within the completete reach which is also assumed in any other reach not affected by a sill If the switch is true the model assumes that erosion and deposition are distributed in a wedge shaped way within the reach This will cause the bed elevation upstream of a sill to adjust twice as fast as in reaches not affected by sills If the node is not given its value is set to false by default The nodes bedSlopeCalculationMethod waterEnergySlopeCalculationM ethod and sedimentEnergySlope CalculationMethod define the way in which the bed slope and the energy slope for the hydraulic and sediment transport capac ity calculations is determined Different realisations are described in sec tion 13 6 However it is not recommended to change the default values of bedSlopeCalculationMethod and waterEnergySlopeCalculationMethod If these nodes are not given their realisation types are set to the SimpleDown stream Two CellGradient of elevation section 13 6 1 and the ReturnBedslope section 13 6 4 by default For sedimentEnergySlopeCalculationMethod the realisations WithFlowResistancePartitioning or NoFlowResistancePartition ing both section 13 6 are recommended The flowResistance node defines the way in which average flow velocity and wetted cross section area are determined for a g
20. values for propertyOfInterest and weightingProperty are the same as for reg ularRiverReachPropertiesForOutput described in section 12 3 1 The use of simpleThreeCellGradient is not recommended 47 bed or waterEnergy or sedimentEnergySlopeCalculation Method realisationType 22202220 SimpleThreeCellGradient propertyOfinterest 0c cece cece eee nee e eee ee eeee weightingProperty 0 cece eee eect hh n 277 Figure 31 SimpleThreeCellGradient 13 6 3 Reduced Water Energyslope The ReducedWaterEnergyslope Fig 32 corrects the water energy slope ac cording to the following equations from Rickenmann and Recking 14 in order to use it as energy slope for sediment transport calculations consider ing the shear stress partitioning 6 5 2 5 pe Utot rn wV9gTh S E A 22a 3 TE 2 52 fe 1 vo rn vg PR 6 5 E 22b Dg4 fo _ Vtot Th 22c us MES Utot w Sred I 22d g vo rn In this fo is the base level flow resistance and ftot is the total flow re sistance vo and vtot are virtual velocities corresponding to fo or ftot respec tively and S eq is the energy slope corrected for the effects of shear stress partitioning The value of the empiric exponent e is defined by the node stressParti tioningExponent with its default value of 1 5 The code will make sure that vtot will not result in a Froude number which exceeds the threshold defined by the node
21. CalculationMethod defines the method by which the threshold for 37 the initiation of bedload motion 6 is calculated The default method is Lam bEtAlCriticalTheta section 13 8 2 The switch simplifiedEquation defines whether the following simplified version of equation 8 based on Rickenmann 12 is to be used default or not y 2 5 V0 0 6 Fr 10 For consistency the following e is used in bedload transport calcula tions if 0 is based on S 4 from equation 22d der be 7 11 The optional switch thetaCriticalBasedOnConstantSred is used to select one of the following two alternatives for the calculation of the correction fac tor y If thetaCriticalBasedOnConstantSred is false default equation 12a is used If it is true equation 12b is used 12a 12b In the first approach equation 12a 0 varies with discharge as it de pends on S eq equation 22d which in turn is a function of ra In the second approach equation 12b suggested by Nitsche et al 9 0er is independent of discharge The value of Se is calculated using equations 22a to 22d with the value of r replaced by the critical hydraulic radius ry ps 1 er Rio Neo 1 The v 5 50 7 13 For the BasedOng versions Figs 20 amp 21 the switch correctionForBed load WeightAtSteepSlopes selects whether the following correction of energy slopes according to Rickenmann 13 is to be applied default or not sin dr sin Br Sp
22. Methods part 2 including default values 12 3 outputMethods 12 3 1 Standard and regular outputs General remarks Within sedFlow the format of the output files can be defined globally by the standard output properties Alternatively single out put files can be defined individually by the regular output nodes In both cases the nodes are the same For the standard output properties the term Standard is added to the node name and the nodes are direct child nodes of the OutputMethods node For individual files the node regularOutput or the nodes regularOutputX may be added to the OutputMethods In this X is replaced by an integer number starting at 1 and going up to the desired number of files If there is a break in the sequence of X values the follow ing nodes will be ignored The nodes for the formatting of output files are described in the following Output file formatting The node forVisualInterpretation is used to switch between the usual output format for numeric postprocessing using other pro grammes and a version for a more visual and by hand postprocessing work flow The usual format has a single line header and no additional formatting symbols The forVisualInterpretation version has a header which consists of several rows and has extra columns with formatting symbols which separate the individual properties and reaches This format is useful when different properties for only a few reaches are written to a single file which sha
23. WaterEnergyslope Calculation Method and with hydraulicHead as propertyOfInterest The output of the simulation setup section 12 3 3 shows in which way the short hand has been expanded 13 6 1 SimpleDownstream TwoCellGradient The simpleDownstreamTwoCellGradient Fig 30 is the recommended stan dard gradient calculation method within sedFlow It determines the local gradient as the difference between the local and downstream property value divided by the local reach length The only mandatory child node proper tyOfInterest defines for which property the gradient is calculated Potential values are the same as for regularRiverReachPropertiesForOutput described in section 12 3 1 bed or waterEnergy or sedimentEnergySlopeCalculation Method realisationType SimpleDownstreamTwoCellGradient propertyOflnterest 0c cece cece een eee nn Figure 30 SimpleDownstream TwoCellGradient 13 6 2 SimpleThreeCellGradient The simple ThreeCellGradient Fig 31 determines a local gradient as the difference between the upstream and downstream property value divided by the sum of the local and upstream reach length The propertyOfInter est defines for which property the gradient is calculated As there may be several upstream reaches the local gradient is calculated as the weighted average of the individual gradients The child node weightingProperty de fines which property is used to weight the individual gradients Potential
24. a version based on dimensionless shear stress 6 12 aai 2a vo o t re 8 Dao Ps _1 p and a version based on discharge per unit flow width q 12 p 1 5 Din 0 2 dy 3 1 2 1 32 9 96 8 9a 1 67 qe 0 065 L i sg Die gts 9b In these equations Fr is the Froude number q is the discharge per unit flow width qe is the discharge per unit flow width threshold for the initiation of bedload motion S is slope and Dsg is the median diameter of the local grain size distribution Additionally non fractional versions of these equations are implemented In these versions transport capacity is estimated based on the median di ameter Dso and the local grain size distributions stay constant in the course of a simulation Please note that sedFlow is optimised for fractional trans port Therefore the use of the non fractional version does not speed up simulations So in total there are four implementations of the Rickenmann equa tion RickenmannBedloadCapacityBasedOnTheta RickenmannBedloadCa pacityBasedOnThetaNonFractional RickenmannBedloadCapacityBasedOng RickenmannBedloadCapacityBasedOnqNonFractional For the BasedOn Theta versions Figs 18 amp 19 the node useOnePointOneA sExponentForFroudeNumber defines whether 1 0 default or 1 1 should be used as exponent for the Froude number The value of 1 1 corresponds to the initial version of the equation but is commonly not used any more The node threshold
25. ansport capacity input The transport capacity input is the default way to feed sediment to the system At the start of a simulation the reaches at the upstream ends of the simulated river system are copied to create virtual margin reaches just outside the boundaries of the simulated system The elevation grain size distribution and alluvium thickness of these margin cells are kept constant for the complete simulation Within these margin cells bedload transport capacity is calculated corresponding to the grain size distribution of its active layer its width amp slope and the current discharge The calculated bedload tranport capacity is then fed to the simulated river system Therefore it is recommended to define the most upstream reaches in a way to be representa tive for the transport system of the headwaters which are not simulated If no sediment is to be fed to the system it is recommended to set the alluvium thickness to zero in the corresponding topmost reach A value close to zero will not do the job as alluvium thickness is not eroded in the virtual margin reaches 10 2 Sedigraph input In order to input sediment independent of the discharge one can define sedigraphs For this the optional file SedimentInputs txt is put into the DischargeAndOtherInputs folder The file Tabs 9 amp 10 contains at least the columns BranchIDs KilometrageUpstreamDirected GrainSizeDistribu tion and SedimentInput TimeSeries Table 9 Struct
26. aries example of Fig 2b Downstream Downstream BranchIDs BranchIDs BranchIDs BranchIDs 1 7 1 3 2 7 2 3 3 8 3 5 4 8 4 5 5 9 5 7 6 9 6 7 7 10 7 9 8 10 8 9 9 12 9 11 10 12 10 11 11 14 12 14 13 14 5 BranchXProfile txt For each branch defined in the Branch Topology txt the code will look for a Branch XProfile tzt file within the LongitudinalProfile folder with X sub stituted by the current branch ID Within sedFlow every branch is dis cretised into river reaches Every reach is described by the cross section at its upstream end Each such cross section is defined by one row in Branch X Profile tzt The file contains at least the following columns Tab 4 Table 4 Minimum structure of BranchXProfile txt KilometrageUpstreamDirected ElevationInM ChannelWidthInM StrataGrainSizeDistribution 22 22 22 Kilometrage UpstreamDirected gives an along channel kilometrage which may be e g the distance to the river mouth Within the file reaches should be sorted from upstream to downstream i e the kilometrage values should decrease It is advisable to define cross sections at roughly equal distances which should not be smaller than about 50 m In this context it is important to note that the lowermost cross section of a feeding branch and the upper most cross section of the recieving branch should not be the same This would define a reach of length zero and lead to kryptic artifacts Rather
27. asion To this date the only additionalMethods are Sternberg Abrasion WithoutFining Fig 47 and SternbergAbrasionIncludingFining Fig 48 These methods correct the estimated bedload flux for gravel abrasion according to the classic equation of Sternberg 15 in which q is bedload flux per unit flow width corrected for abrasion is an empiric abrasion coefficient and AX is the travel distance of the grains given in kilometres In this the material loss due to erosion is regarded as suspension throughput load dban dp exp CA AX 30 The value of lambda is defined by the mandatory child node sternber gAbrasionCoefficient The method SternbergAbrasionWithoutFining simply reduces the transported volumes according to equation 30 The method Sternberg AbrasionIncludingFining reduces the transported volumes and shifts material from coarser to finer grain size fractions assuming equally dis tributed grain sizes within each fraction If both nodes are given only the Sternberg AbrasionIncludingFining is used additionalMethods SternbergAbrasionWithoutFining _sternb ergAbrasionCoefficient cece eee eee ee Figure 47 SternbergAbrasionWithoutFining additionalMethods Sternb ergAbrasionIncludingFining sternb ergAbrasionCoefficient cece eee eens Figure 48 SternbergAbrasionIncludingFining 55 Part III Appendix Notation The following symbols are used in this manual B an empiric constant fact
28. ations which implies long time steps executable versions of the code are provided which will give a warning if the length of the time step falls below a threshold of 1 second or 100 seconds respectively A version which does not give a warning for short time steps is provided as well Most of the simulation data is fed to the model by tabulator delimited text files which can be created and edited with regular spreadsheet applica tions such as Microsoft Excel Each such file contains one header line which defines the number of columns for this file together with the column names The code selects the data based on the column names Thus there is no fixed order for the columns and additional extra columns will be ignored Empty cells are ignored by the code as well Within rows containing more elements than the header line the first elements up to the number of header line entries are used as data entries Any further entries are ignored Rows containing less entries than the header line are ignored as well By this it is easy to include comments in the spreadsheet files Most of the simulation parameters are fed to the model by an extensible markup language xml input file For the creation and editing of the input xml file on the Microsoft Windows platform the use of the freeware program Microsoft XML Editor 2007 is recommended The code will find any infor mation within the xml file based on the position in the node hierarchy and the node
29. aximumFroudeNumber value from flowResistance minimumInputSlope 0 0 cece ee eee nee ne eees ensureMinimumInputSlope if minimumInputSlope node exists Figure 32 ReducedWaterEnergyslope including default values 49 bed or waterEnergy or sedimentEnergySlopeCalculation Method realisationTypeReducedWaterEnergyslopeNotUsingWaterEnergyslopeVariable stressPartitioningExponent 0 cece eee eee eee ee 1 5 _ calculationBasedOnglInsteadOfh 22 2222 false maximumFroudeNumber value from flowResistance minimumInputSlope 0 000 e cece eee nnne ensureMinimumInputSlope if minimumInputSlope node exists simpleWaterEnergyslopeCalculationMethod Do DET SimpleDownstreamTwoCellGradient propertyOflnterest 0 cece cence eee hydraulicHead Figure 33 Reduced WaterEnergyslopeNot UsingWaterEnergyslopeVariable including default values 13 6 4 ReturnBedslope and Return WaterEnergyslope The realisations ReturnBedslope Fig 34 and Return WaterEnergyslope Fig 35 do not calculate a new gradient but use the value of the bedslope or water energy slope respectively If the child node minimumSlope is given the code will use this value whenever the original bedslope or water energyslope are smaller than this value bed or waterEnergy or sedimentEnergySlopeCalculation Method realisationType 2 cc cece eee eee e eee nees ReturnBed
30. ayerWithShearStressBasedUpdate Fig 28 basically works in the same way as T woLayerWith Continuous Update Therefore TwoLayer WithS hearStressBasedUpdate contains all nodes of TwoLayer WithContinuous Up date However in case of erosion only some fraction is of the material added to the active layer from below carries grain size information from the sub surface 0 m Oes is gp 9 with 0 lt i lt 1 20 and with q calculated according to the following relation from Jaggi 5 D 3 Oca des marine 21 pesce In these equations 0 and Oca are representative 6 values for the sub surface alluvium or the active layer respectively Analogously DmAriths and DmAritha are DmAritn Values for the subsurface alluvium or the active layer respectively To determine is the TwoLayer WithShearStressBasedUpdate has the fol lowing specific nodes The method for the calculation of 0 is defined by the node thresholdCalculationMethod which is by default the same method as used in the bedloadTransportEquations section 13 3 Usually the break up conditions including the fcs and Oca are defined using the initial local con ditions and then kept constant for the complete simulations To constantly update the break up condition according to the changing local situation one simply sets the switch dynamicBreakUpConditions to true If some global constant break up condition shall be used for all reaches independent of the local conditions one
31. calDimensionlessShearStress 0 03 Figure 40 LambEtAlCriticalTheta including default values 13 9 hidingFactorsCalculationMethod realisations The hidingFactorsCalculationMethod modifies the threshold for the initiation of bedload motion e for the different grain size fractions to account for the effects of grain exposure and hiding 13 9 1 PowerLawHidingFunction The PowerLawHidingFunction Fig 41 modifies 0 according to the follow ing equation e g Parker 10 D m ies In this Qei is the 6 for the grain size fraction 7 D is the representative grain diameter of this fraction D is a reference grain diameter percentile with 52 x defined by the optional child node referenceDiameterPercentile with its default value of 50 0 and m an empiric hiding exponent which is defined by the optional child node exponent with its default value of 0 8 hidingFactorsCalculationMethod realisation Type ere tr asa daas at ER Ed PowerLawHidingFunction reference Diameter Percent 50 0 qu EUM 0 8 Figure 41 PowerLawHidingFunction including default values 13 9 2 NoHiding The NoHiding Fig 42 does not modify 6 at all In this way it corresponds to a PowerLawHidingFunction with an exponent of 0 0 hidingFactorsCalculationMethod es TE ae spe re NoHiding Figure 42 NoHiding 13 9 3 WilcockCroweHidingFunction The WilcockCroweHidingFunction Fig 43 modifies 6 according to the fol lowing e
32. ch already has been discussed in section 2 The name of the root node of this file defines the input reader i e the way in which all infor mation is read and interpreted For this manual the root node name is SEDFLOW_StandardInput The root node contains at least the two main nodes overallParameters and riverSystemMethods Fig 3 which will be de scribed in the following An example with potential values is given in Fig 4 For the sake of clarity the main nodes are printed in dark red in all figures while any other node which contains child nodes is printed in blue SEDFLOW _ StandardInput overallParameters lowerDiameterBoundaryForFinestFractionInCM riverSystemMethods sedimentEnergySlopeCalculationMethod L realisation R TTT bedloadTransportEquations realisation Type sese e Figure 3 Minimum xml The overallParameters collect values of typically single parameters which are used throughout the complete simulation space and time It contains at least the node lowerDiameterBoundaryForF inestFractionInCM which de fines the lower boundary for the finest grain size fraction Make sure that its value is smaller than the smallest diameter of the grain size distribution files section 6 The riverSystemMethods typically collect the procedures and algorithms which are used in the course of a simulation The names of the nodes specify the job which is to be fulfilled by the corresponding method The c
33. different time series of sediment flux fed to the simulated channel The code will look for these files in the DischargeAndOtherInputs folder The structure of these time series files Tab 11 and the way in which they are read by the code correspond to the discharge time series as described in section 7 Table 11 Structure of sedigraph spreadsheets ElapsedSeconds SedimentInputInM3PerS 22 22 Please note that the model reads the sediment input rate at the beginning of a time step and keeps this value for the complete time step This means that the temporal discretisation of the sediment input time series should not be finer than the average time step of the simulation Any input event which is shorter than the average time step will not be captured properly by this input mechanism The sedigraph input is usually added to the material coming from up stream To modify this one adds the optional column ReplacingRegularDe positionRateInsteadOfAddingTolt One of the rare applications of this option 15 is the simulation of a retention basin with a defined downstream sediment yield which may be zero as well If the column ReplacingRegularDeposi tionRateInsteadOfAddingTolt is not included in the SedimentInputs tzt its values are set to false by default The sediment inputs are usually considered to include pore volume To modify this one adds the optional column InputIncludingPore Volume If the column is not giv
34. draulicRadius If waterFlowRouting ImplicitKinematicWave maximumFroudeNumber e 0c cece cece een e cece anes 4 0 minimumHydraulicSlope 0 ccc 0 0004 Figure 11 flowResistance including default values 33 13 1 1 FixedPowerLawFlowResistance The FiredPowerLawFlowResistance Fig 12 has four optional nodes which can be derived from the following equation The value of j is defined by the node factor with its default value of 6 5 The value of k is defined by the node grainsFactor with its default value of 1 0 D is the x th percentile grain diameter with the value of x defined by the node grainsPercentile with its default value of 84 0 The value of l is defined by the node exponent with its default value of 0 166666667 flowResistance _ realisationType FixedPowerLawFlowResistance MCI CMM DRESD 6 5 E BIAIDSBAGOE soca iue reu robot e EEUU wanda aera 1 0 gramsPercent le u see rb ere vereor PEPPER 84 0 lt exponent d dd ses aere qat Uo hed aed das 0 166666667 Figure 12 FixedPowerLawFlowResistance including default values If one uses the default value of i for the exponent l this method corresponds to a Manning Strickler flow resistance relation with kst zd TAS 4 n he Dye in which kst is the Strickler roughness coefficient and n is the Manning roughness coefficient If one uses all default values this method corresponds to a Manning Strickler
35. e following four nodes exists referenceCellIDStandard nanunua 777 referencePropertyStandard 0 cee cece eee eee eee 77 L Lhresh ol d ToabBoebxceed ed Rrtandard 777 _secondaryOutputIntervalStandard 2220220 777 regularRiverReachPropertiesForOutputStandard elevation activeLayerPerUnitBedSurfaceD50 activeLayerPerUnitBedSurfaceD84 bedslope maximumWaterdepth accumulatedBedloadTransport notOutputSimulationSetup 0 cece eee eee ee eee false outputSimulationSetup Balle 12 piae a add ets SimulationSetup txt precisionForOutput eee eee eee Standard value SUMO T PET empty L negabat On MEE pea doces era obedire NN N NR ded ad gig empty printStartingTime seeeseseeeees eee true printModelVersion lssseeesesssseee eene true setupPropertiesForOutput CalcBedloadCapacity FlowResistance CalcGradient StrataSorting w Figure 9 outputMethods standard including default values SEDFLOW StandardInput overallParameters riverSystemMethods output Methods regularOutputX or Visual lnterpretartion Standard value regularRiverReachPropertiesForOutput i name name of PropertyFor utput or regular utputX explicitTimesForOutput 22222 standard value soutputinterval u a standard value precisionForOutput eee eee eee standard value reachIDsForOutput Standard value _writeLineEac
36. edloadTransportEquations realisations estimate the dimensionless bed load transport rate defined by the following equation D 2 7 LY 1 R in which q is bedload flux per unit flow width p is fluid density ps is sediment density g is gravity acceleration and D is grain diameter For all bedloadTransportEquations Fig 17 the node maximumFrac tionOfActiveLayer ToBeEroded with its default value of 0 9 defines some sort of vertical Courant Friedrichs Lewy criterion for erosion If a kinematic wave flow routing is selected the code will prevent zero or negative bed slopes The switch preventZeroOrNegativeBedSlopes is used to force this option on or off If zero or negative bed slopes are to be prevented the code will only allow changes of the bed slope which are a fraction of its current value This fraction is defined in the node mazimumRelativeT woCellBedSlopeChange with its default value of 0 9 Please note that the prevention of zero or negative bed slopes will extremely slow down simula tions with bed slopes close to zero 36 bedload Transport Equations maximumFractionOfActiveLayerToBeEroded 0 9 preventZeroOrNegativeBedSlopes if waterFlowRouting some KinematicWave maximumRelativeTwoCellBedSlopeChange 0 9 Figure 17 bedloadTransportEquations including default values 13 3 1 Rickenmann bedload capacity For the bedload transport equation of Rickenmann there exists
37. edrichsLewyNumber eese 0 9 timeStepThresholdForTerminatingSimulation 0 0000000001 L timestepPactor wine nee 1 0 kilometrageOfSimulationOutlet 00 cece eee eee 0 0 __thicknessInputsIncludingPoreVolume 20 22 true riverSystemMethods outputMethods Figure 6 overallParameters including default values 19 0 9 As sedFlow is intended for fast simulations with long time steps the simu lation will stop whenever the time step length drops below the timeStep Thresh oldFor TerminatingSimulation as this points to some problems in the current simulation The default value for this threshold is 0 0000000001 The timeStepFactor may be used to modify the time step lengths which the model determines based on its criteria for numeric stability Its default value is 1 0 and it is not recommended to use the option of changing this value The length of the lowermost river reach is calculated as the difference between its kilometrage from the Branch XProfile trt and the kilometrage OfSimulationOutlet By default the simulations start at a kilometrage of 0 0 The switch thicknessInputsIncludingPore Volume is used to define whether or not thickness inputs such as e g the erodible alluvium thickness are con sidered to include the pore volume The default value of this node is true 12 2 riverSystemMethods The switch upstreamOfSills WedgeShapedInsteadOfParallel Update determines
38. en its values are set to true by default 10 3 Instantaneous sediment pulse input As mentioned before sediment input events which are shorter than the av erage time step cannot be represented by sedigraph inputs as described in section 10 2 To account for such events one adds the file InstantaneousSed imentInputs trt to the DischargeAndOtherInputs folder This file contains the columns BranchIDs KilometrageUpstreamDirected GrainSizeDistribu tion Input VolumesInM3 ElapsedSeconds and InputIncludingPore Volume in which InputIncludingPore Volume can be left out to set all its values to true All columns except InputVolumesInM3 and ElapsedSeconds are treated in the way which is described in section 10 2 The InputVolumesInM3 define the volume of sediment which is to be fed to a certain reach and the Elapsed Seconds define the point in time in which this shall happen Table 12 Structure of InstantaneousSedimentInputs txt part 1 BranchIDs KilometrageUpstreamDirected GrainSizeDistribution 22 22 P Table 13 Structure of InstantaneousSedimentInputs txt part 2 including default values Input VolumesInM3 ElapsedSeconds InputIncludingPoreVolume 22 22 true 11 Sills The large scale effects of sills i e the dissipation of energy can be considered within sedFlow simulations A sill is represented by the fixed elevation of the top edge of the sill and the variable elevat
39. etricMean GrainDiameter depositionRateD20 depositionRateD30 depositionRateD50 depositionRateD84 depositionRateD90 erosionGeometricMean GrainDiameter erosion D20 erosion D30 erosionD50 erosionD84 erosionD90 depositionGeometricMean GrainDiameter depositionD20 depositionD30 depositionD50 depositionD64 depositionD90 IncludingPore Volume erosionPer UnitBedSurface Overall Volume e erosionPer UnitBedSurface GeometricMeanGrainDiameter e erosionPerUnitBedSurface Overall VolumeIncluding Pore Volume STE na EN e erosionPerUnitBedSurfaceD30 e erosionPerUnitBedSurface MedianGraimbaneter e erosionPerUnitBedSurfaceD50 e erosionPerUnitBedSurfaceD84 e erosionPerUnitBedSurface ArithmeticMeanGrainDiameter e erosionPerUnitBedSurfaceD90 e depositionPer UnitBedSurface e depositionPer UnitBedSurface ArithmeticMeanGrainDiameter e depositionPer UnitBedSurface IncludingPore Volume e depositionPer UnitBedSurface GeometricMeanGrainDiameter depositionPer UnitBedSurface e depositionPer UnitBedSurfaceD20 Overall Volume e depositionPer UnitBedSurfaceD30 e depositionPer UnitBedSurface Overall VolumeIncludingPore Volume depositionPer UnitBedSurfaceD50 E e depositionPerUnitBedSurfaceD84 e depositionPer UnitBedSurface MedianGrainDiameter e depositionPerUnitBedSurfaceD90 The erosionRate properties display the potential local transport capac ity in m not considering supply limitations The erosion proper
40. factor fo base level flow resistance ftot total flow resistance F proportion of sand fraction Fr Froude number g gravitational acceleration h hydraulic head hq thickness of moving sediment layer is grain size influence from the subsurface alluvium j k l empirical constants kst Strickler roughness coefficient L process break point m empiric hiding exponent Mwe hiding exponent according to Wilcock and Crowe 16 n Manning roughness coefficient q discharge per unit flow width q dimensionless unit discharge qb bedload flux per unit flow width dban Corrected for gravel abra sion qe discharge per unit flow width threshold for initiation of bedload motion QcA qe corrected for the effect of bed armouring Q discharge Qiat lateral water influx rj hydraulic radius The rn for 050 theta S slope Se virtual slope for the correction of 0 based on rj Sp bed slope S friction slope Sk slope corrected for bedload weight at steep bed slopes Sreq slope reduced for form roughness t time U shear velocity v flow velocity vo virtual v corresponding to base level flow resistance ty bedload velocity Utot Virtual v corresponding to total flow resistance Vpore pore volume fraction w flow width x distance in flow direction z elevation of channel bed 57 List COND OF We 10 11 12 13 14
41. fference between the thresholds default Sna te with Amin minimum DifferenceBetween Thresholds For Active Layer Thickness low valueOrFactorForLowerThresholdForActiveLayer Thickness and up valueOrFactorFor Upper Threshold ForActiveLayer Thickness 43 strataSorting realisationType StratigraphyWithThresholdBasedUpdate LHcTrerment Laver Thcmnegg ne dynamicThresholds 22 2 0 RR N R N R N eee 200er uselnitialGrainSizesForConstantLayerThicknessonly needed if not dynamicThresholds referenceGrainSizePercentile uns qe pad qaod axe daa dd only needed if either dynamicThresholds or uselnitialGrainSizesForConstantLayerThickness valueOrFactorForLowerThresholdFor ActiveLayerThickness valueOrFactorForUpperThresholdFor ActiveLayerThickness _minimumDifferenceBetweenThresholdsFor Active aver Thickness 1 1 times incrementLayerThickness minimumLowerThresholdForActiveLayerThickness see text for default numberOfLayers 0 c cece eee ee 77 Figure 26 StratigraphyWithThresholdBasedUpdate including default val ues 13 5 2 TwoLayerWithContinuous Update Within TwoLayerWithContinuous Update Fig 27 the thickness of the ac tive surface layer stays constant for the complete duration of the simulation or is dynamically adjusted according to the current local grain size distribu tion If it is constant in time either a single value can be selected for all
42. hTimeStep standard value _ output TimeStepLength standard value outputInitial Values 000000 000s standard value printUpstreamMargins standard value _ printDownstreamMargin 2 2 2 standard value Secondary OutputInterval referenceCelllD 2 22222200 standard value referenceProperty 0 ceee eee eee standard value threshold ToBeExceeded Standard value secondaryOutputInterval standard value outputAccumulatedBedloadTransport X 2 NAME Er empty L explicitTimesForOutput 2 2 standard value L OULput Tnterval essen standard value precisionForOutput K 2c eee eee eee standard value reachIDsForOutput 0 e eee eee standard value _writeLineEachTimeStep standard value outputIncludingPoreVolume e eee eee ee true outputDetailedFractional 00 0000 cece ences false L backupXML alternativePathForXMLBackupOutputs empty L explicitTimesForOutput 2 2 2 standard value outputInterval 22222222cee nennen standard value precisionForOutput eee eee eee standard value overwriteFiles 0 00 ccc cece eee eere true numberOfFileIDDigits sseseseeee II 4 Figure 10 outputMethods regular including default values 32 13 Realisations 13 1 flowResistance realisation
43. hild nodes specify the way in which the method fulfills its job Typically there is always one child node called realisation to select the general algorithm of the current method Depending on the realisation there may be other child nodes defining further parameters which are specific for the current reali sation Different realisations and their various parameters are described in section 13 The riverSystemMethods contain at least the nodes sedimentEn ergySlopeCalculationMethod and bedloadTransportEquations The node sedi mentEnergySlopeCalculationMethod is mainly used to determine whether or not the energy slope for the determination of bedload transport capacity should be reduced in order to account for flow resistance partitioning The node bedloadTransportEquations defines the way in which the bedload trans 10 port capacity is estimated The recommended combination of realisations is WithFlowResistancePartitioning and RickenmannBedloadCapacityBasedOn Theta Fig 4 SEDFLOW _ StandardInput overallParameters lowerDiameterBoundaryForFinestFractionInCM 0 2 riverSystemMethods sediment EnergySlopeCalculationMethod L reahsation Type WithFlowResistancePartitioning bedloadTransportEquations L realisationType RickenmannBedloadCapacityBased nTheta Figure 4 Minimum xml with recommended values Please consider the cautionary notes given in section 13 2 related to the water flow routing method UniformDischarge
44. ields stress 7 see equation below If not the code will determine the sand fraction F from the local grain size distribu tion However it is recommended to have only grain size fractions larger than 2 mm as they will be transported as bed load For such a grain size distribution the sand fraction is always 0 Therefore it is recommended to select a constantSandFraction between 0 0 and 0 2 Trm 0 021 0 015 exp 20 F 16 The second optional node useConstantSandFraction can be used to force the determination of the sand fraction based on the constant value or based on the grain size distribution Its default value is true if the node con stantSandFraction is given and false if not 40 bedload Transport Equations realisationType WilcockCroweBedloadCapacity constantSandFraction 00 ccc eee een ee useConstantSandFraction if constantSandFraction is given Figure 22 WilcockCroweBedloadCapacity including default values 13 3 3 ReckingBedloadCapacityNonFractional The ReckingBedloadCapacityNonFractional Fig 23 has no specific nodes It calculates the non fractional transport rates local grain size distributions constant within simulation according to the following equations from Reck ing 11 D 0 93 Bega 1 32 Sy 0 037 38 17a D 4445 v5 L 12 53 28 04605 17b S g4 ERES 17c DES D 18 S 6 5 0 0005 28 Pa for 6g
45. ified by the node name 12 3 2 backupXML For very long simulations it is possible to create backups of the current state of a running simulation To do so one adds the node backup X ML to the outputMethods The child nodes ezplicitTimesForOutput outputInterval and precisionForOutput are used as described for standard and regular outputs in section 12 3 1 By default any new backup will replace the previous one To change this one sets overwriteFiles to false In this case an ID will be appended to the file name The number of digits of this ID can be changed using the numberOfFileIDDigits node For multiple backups it may be worth to redirect these files from the standard Output folder where they are stored by default to some other location which is given in the node alternativePathForX MLBackupOutputs To restart a simulation from a backup one simply starts sedFlow and imports the backup file instead of 29 the normal main input xml section 8 In general it is not recommended to create backup files as this considerably slows down a simulation 12 3 3 outputSimulationSetup By default the model creates an easy to read summary of the current sim ulation setup To supress this output one sets the node notOutputSimu lationSetup to true The format of the output can be defined within the outputSimulationSetup node The child node precisionForOutput is used as described for standard and regular outputs in section 12 3 1 The selectio
46. ion of the river bed just downstream 16 of the sill disregarding local effects such as the stilling pool The slope upstream of the sill is calculated using the elevation difference between the next upstream cross section and the maximum of the fixed sill top edge and the variable bed just downstream of the sill The maximum function allows for the consideration of the burial of sills The slope downstream of the sill is calculated using the elevation difference between the variable bed just downstream of the sill and the next downstream cross section To simulate a series of sills with distances between them which are smaller than the ideal distance of cross sections it is recommended to substitute the series of sills by one single sill with an elevation drop equal to the sum of elevation drops of the complete series of sills As only the large scale effects of sills are considered the results will be the same and with one single sill it is possible to keep the ideal distance of cross sections All outputs for a cross section with a sill refer to the variable bed just downstream of the sill The evolution at the top edge of the sill is not displayed as it can be trivially derived To introduce sills one adds the optional Sills txt to the folder Longitudi nalProfile The columns BranchID and KilometrageUpstreamDirected define the reach at which a sill should be introduced If no existing reach can be found to match the branch ID and kilometrage
47. ionTypeRickenmannBedloadCapacityBased nThetaNonFractional useOnePointOneAsExponentForFroudeNumber false simplifiedEquation 0 c cece cence nen true takeArmourLayerIntoAccount 0 cee eee eee e eee thresholdCalculationMethod realisationType 2220 220 LambEtAlCriticalTheta thetaCriticalBasedOnConstantSred 22222220220 false Figure 19 RickenmannBedloadCapacityBasedOnThetaNonFractional in cluding default values 39 bedload Transport Equations realisationType RickenmannBedloadCapacityBasedOng correctionForBedloadWeight AtSteepSlopes true hidingFactorsCalculationMethod Bu S EE a ala RET et PowerLawHidingFunction Figure 20 RickenmannBedloadCapacityBasedOnq including default values bedload Transport Equations realisationTypeRickenmannBedloadCapacityBasedOnqNonFractional correctionForBedloadWeight AtSteepSlopes true takeArmourLayerIntoAccount 0 cece e eee ee eee Figure 21 RickenmannBedloadCapacityBasedOnqNonFractional including default values 13 3 2 WilcockCroweBedloadCapacity The WilcockCroweBedloadCapacity Fig 22 calculates bedload transport capacity according to Wilcock and Crowe 16 Its main optional node is constantSandFraction the value of which should range between 0 0 and 1 0 If this node is given the code will use its value for the determination of the reference dimensionless Sh
48. iven discharge Dif ferent realisations are described in section 13 1 Please note that for the ImlicitKinematic Wave flow routing section 13 2 3 only the F redPower LawFlowResistance flow resistance section 13 1 1 can be used However 20 the VariablePowerLawFlowResistance section 13 1 2 is assumed to better represent the processes in steep mountain channels If the flowResistance node is not given its realisation type is set to FiredPowerLawFlowResis tance if the waterFlowRouting realisation is ImplicitKinematic Wave or to VariablePowerLawFlowResistance in any other case by default The node bedload VelocityCalculationMethod defines the way to estimate grain velocities which are only needed and used to check for the bedload flux the Courant Friedrichs Lewy criterion of numeric stability Different realisations are described in section 13 4 If the node bedload VelocityCalcu lationMethod is not given its realisation is set to VelocityAsTranpsportRate PerUnitCrossSectionalArea section 13 4 1 by default The node tauCalculationMethod defines the way to estimate the bed shear stress Different realisations are described in section 13 7 If the node is not given its realisation is set to Energyslope Tau section 13 7 by default The node active WidthCalculationMethod defines the way to estimate the width in which sediment transport takes place At the moment SetAc tive WidthEqualFlow Width is the only realisation for this node
49. ll be evaluated in a by hand postprocessing The precisionForOutput defines the precision for the output of floating point numbers Please note that floating point numbers are output in scien tific format convention For long simulation times make sure that the output precision is sufficient to discriminate the different ElapsedSeconds values By outputTimeStepLength an additional column with the current time step length in seconds may be added to the output file By outputInitial Values an output row with the values at the start of the simulation is added This makes sense for some state properties like elevation or grain diameter percentiles For other properties like e g the transport rate which are calculated in the course of a simulation dummy values will be printed As described in section 10 1 the code internally creates virtual margin reaches at the upstream ends and at the downstream end of the simulated system These margin reaches are used to feed inputs to the system and to recieve the outflux at the downstream end The printUpstreamMargins and the printDownstreamMargin switches may be used to add these virtual mar gin reaches to the output file This option is mainly intended for debugging 24 Output timing The ezplicitTimesForOutput may contain a series of child nodes which define certain points in time expressed as elapsed seconds at which an output row shall be written The names of the child nodes are ignored The o
50. marimumFroudeNumber The default for this threshold is the value which is used for the flowResistance section 13 1 If the child node minimumInputSlope is given the code will use this value as unreduced slope for the calculation if the original unreduced slope S is smaller than this value The switch calculationBasedOnqInsteadOfh may be used to correct the water energy slope according to the following equations if a BasedOng ver sion of the Rickenmann transport equation is used in which q is the di 48 mensionless unit discharge The default value of this node is false xk q r 23a vg S Di ENT 0 2435 viot d vH S Da 1 443 q7999 11 2 x 23b vo q Vg S Dai 3 074 q 9 23c 1 5 Vis e and Srea 5 uS u 236 The companion realisation ReducedWaterEnergyslopeNot Using WaterEn ergyslope Variable Fig 33 can be used in the same way as Reduced WaterEn ergyslope However it does not use the water energy slope as unreduced slope for the calculation but offers the node simple WaterEnergyslopeCalculation Method to define a new gradient calculation method to create the unreduced slope for the calculation bed or waterEnergy or sedimentEnergySlopeCalculation Method _realisationType 222222200 ReducedWaterEnergyslope stressPartitioningExponent 0 cece eee eee eee 1 5 calculationBasedOngInsteadOfh 0 0 0 0 cc ccc eee false m
51. n SizesForConstantLayer Thickness is true the thresholds are defined as a mul tiple of a reference percentile grain diameter D of the initial local grain size distribution If the switch dynamic Thresholds is true the thresholds do not stay constant in the course of a simulation but are updated according to the changing local grain size distribution If the thresholds are defined as a function of the grain size distribution the node reference GrainSizePer centile defines the value of x of the reference percentile diameter D and the nodes valueOrFactorForLowerThresholdForActiveLayerThickness and value OrFactorForUpperThresholdForActiveLayerThickness are not used as abso lut values but as factors which are multiplied with D in order to calcu late the threshold values When D becomes small the thresholds become small as well However very thin active layer thicknesses can slow down the simulations To prevent this the node minimumLower Threshold ForAc tiveLayer Thickness defines a minimum for the lower threshold When the thresholds become small the difference between them will become small as well If this difference becomes smaller than the incrementLayer Thickness it is not possible for the code to produce an active thickness which has a value between the thresholds Therefore the node minimumDifferenceBe tweenThresholdsForActiveLayerThickness with its default value of 1 1 times the incrementLayerThickness defines a minimum for the di
52. n and order of the displayed setup properties is defined in the setupProperties ForOutput The default values are CalcBedloadCapacity FlowResistance CalcGradient and StrataSorting Other potential values are CalcTau and CalcActiveWidth The file name can be changed using the node name The simulationID and simulationName can be used to include any user defined text to the output file The switch printStarting Time defines whether the starting time of the simulation shall be included to the file and the switch printModelVersion selects whether the compilation date of the used model binary shall be included to the file as well By default the value of both switches is true 30 SEDFLOW StandardInput overallParameters riverSystemMethods outputMethods createStandardOutputs 0 e cece eee eee eee eee true forVisualInterpretationStandard cece eee false explicitTimesForOutputStandard wc empty outputIntervalStandard 0 060 c cece eee ee 3600 0 precisionForOutputStandard 0 0 0 c cece neces 4 writeLineEachTimeStepStandard false outputTimeStepLengthStandard eee eee false outputInitial ValuesStandard 0 0000 cece true _printUpstreamMarginsStandard 2 22222220 false printDownstreamMarginStandard 0 0 cee eee eee false reachIDsForOutputStandard all reach IDs useSecondaryOutputInterval if one of th
53. n diameter fractions fractional transport iv consideration of the effects of adverse slopes in terms of pondages e g due to sudden sediment deposition by debris flow inputs v fast calculations for modelling entire catchments and for automated calculation of many scenarios exploring a range of parameter space vi flexibility in model development featuring an object oriented code de sign vii flexibility in model application featuring an easy and straightforward pre and postprocessing of simulation data Thus the model sedFlow fills a gap in the range of existing sediment transport models for mountain streams and the mentioned aims have led to the implementation described in the following sections This implementation represents a current state and may be easily extended and adjusted in the future sedFlow is a console program without graphical user interface This format was chosen to enable automated simulation starts within batch or shell scripts and especially to allow for short calculation times The input files are prepared in other programs sedFlow reads these files and creates output files which are continuously updated in the course of a simulation To check the progress of a simulation one can take a look at the ElapsedSeconds column of any output file As some programs block files for writing it is recommended to open copies of an output file as long as the code is still running As sedFlow is intended for fast simul
54. name Additional extra nodes are simply ignored and may be used for commenting Any typing errors in the names of the files columns or nodes will lead to error messages or even worse that the model will ignore the corresponding element without giving a warning at all Blanks are usually ignored by the code In the tables and figures of this manual three question marks in dicate a required user input for which no default exists If such a required input is not given usually an error message will request it 2 Starting the model The starting point for each simulation is the main input xml file The loca tion of this file can be given as first argument of the model call e g when starting simulations within a script If the location is not given or the code is started by a simple doubleclick the model will ask for the location of the file The location can be given either by an absolute path or by a relative path starting at the location of the model It is recommended to copy the model into the same folder as the main input xml file and then simply input the name of the xml file At the model start the user is requested to hit enter to accept the terms of the license under which sedFlow is distributed Alternatively the user may type acceptLicense as second argument of the model call e g when starting simulations within a script 3 The simulation folder The location of the main input xml file usually defines the simulation folder
55. ncluding default values SillTopEdgeElevationInM BranchIDs KilometrageUpstreamDirected or SillDsopHeiehtinM PoleniFactor 227 227 22 Ji potential nodes and their default values which are inserted if a node is not given The summarising illustrations Figs 6 to 10 just give one possible example of a complete input xml structure including all default values 12 1 overallParameters inputUpperBoundaryInsteadOfMean GrainDiameter is used to switch the way in which the GrainDiameterInCM column of the grain size distribution files is interpreted If it is true default value the diameters are interpreted as upper fraction boundaries and the representative fraction diameter is calcu lated as the mean of the boundaries If it is false the diameters given in the grain size distribution files are directly used as representative fraction diameters The lowerDiameterBoundaryForF inestFractionInCM which has been al ready described in section 8 is only needed if inputUpperBoundaryInstead OfMeanGrainDiameter is true If inputUpperBoundaryInsteadOfMeanGrainDiameter is true the use ArithmeticMeanInsteadOf Geometric MeanForFractionGrainDiameters is used to switch between the geometric default and arithmetic mean for the rep resentative fraction diameter The density Water and densitySediment define the liquid and solid den sities in 4g with their default values of 1000 0 and 2650 0 respectively The pore Vol
56. o estimate the thickness of the moving sediment layer estimateThicknessOfMovingSedimentLayer realisationType MultipleDiameter0fCoarsestGrainMoved minimum TransportRateForFractionToBeConsideredMoving 0 0001 ico E 1 75 Figure 45 MultipleDiameterOfCoarsestGrainMoved including default val ues 13 10 3 MultipleReferenceGrainDiameter The MultipleReferenceGrainDiameter Fig 46 multiplies some reference di ameter percentile Dy of the local grain size distribution with the value de fined by the node factor with its default value of 1 25 in order to estimate the thickness of the moving sediment layer The value of z is defined by the node referenceDiameterPercentile with its default value of 84 0 In cases with no alluvium cover or input from upstream it is not possible to calculate any di ameter percentile In these cases the thickness of the moving sediment layer is assumed to be equal to the value in metres defined in the child node default ThicknessForCasesWithNoAlluviumOrInputFrom Upstream with its default value of 0 7 54 estimateThicknessOfMovingSedimentLayer realisation Type MultipleReferenceGrainDiameter fI DOM RE EEEE EEEE E E E E E OE EE 1 25 referenceDiameterPercentile 00 c ccc e cece nee 84 0 default ThicknessFor Cases WithNoAlluvium OrInputFrom Upstream 0 7 Figure 46 MultipleReferenceGrainDiameter including default values 13 11 additionalMethods with Sternberg abr
57. ope J Geophys Res 113 F02008 2008 doi 10 1029 2007JF000831 Z Liu and E Todini Towards a comprehensive physically based rainfall runoff model Hydrology and Earth System Sciences 6 5 859 881 2002 doi 10 5194 hess 6 859 2002 M Nitsche D Rickenmann J M Turowski A Badoux and J W Kirchner Evaluation of bedload transport predictions using flow re sistance equations to account for macro roughness in steep moun tain streams Water Resour Res 47 W08513 2011 doi 10 1029 2011WR010645 G Parker Transport of gravel and sediment mixtures In Sedimentation Engineering Theories Measurements Modeling and Practice volume 110 of ASCE Manuals and Reports on Engineering Practice chapter 3 pages 165 252 American Society of Civil Engineers ASCE 2008 60 11 12 13 14 15 16 A Recking A comparison between flume and field bed load transport data and consequences for surface based bed load transport prediction Water Resour Res 46 W03518 2010 doi 10 1029 2009WR008007 D Rickenmann Comparison of bed load transport in torrent and gravel bed streams Water Resour Res 37 12 3295 3305 2001 doi 10 1029 2001WR000319 D Rickenmann Geschiebetransport bei steilen Gef llen In VAW 75 JAHRE Festkolloquium 7 Oktober 2005 pages 107 119 Versuch sanstalt f r Wasserbau Hydrologie und Glaziologie Zurich Switzer land 2005 D Rickenmann and A Recking Evaluation of flow
58. or y correction factor for 6 At temporal discretisation i e time step duration Ax spatial discretisation i e reach length AX travel distance of grains 0 dimensionless bed shear stress g4 0 for Desa 0 dimensionless bed shear stress threshold for initiation of bedload motion Oca a representative 0 for the complete active layer Oci 6 for i th grain size fraction Oek 0 corrected for bedload weight at steep adverse bed slopes Oer Oe corrected for form rough ness cs a representative 0 for the complete subsurface alluvium des 0c for Da 0 0 for the roughness length according to Julien and Bounvilay 6 A empiric abrasion coefficient u weir coefficient p fluid density ps sediment density T bed shear stress Tm reference dimensionless Shields stress for mean size of bed surface r angle of repose of bedload material dimensionless bedload flux a b empiric constants D grain diameter Dj representative grain diameter for i th grain size fraction Dm geometric mean for grain diameters DmArith arithmetic mean for grain diameters DmAritha DmArith for the active layer DmArith DmAritn for the subsurface alluvium D x th percentile for grain diameters Des x th percentile for grain diameters of bed surface Dan median grain diameter 56 C empiric exponent ranging from 1 to 2 f Darcy Weisbach friction
59. quation from Wilcock and Crowe 16 mye _ 0 67 bci 0e with Mwe 1 29 Dm 1 exp 1 5 f In this Dm is the geometric mean diameter of the local grain size distribution and Mwe the hiding exponent according to Wilcock and Crowe 16 hidingFactorsCalculationMethod realisation Type ha un Rc ne WilcockCroweHidingFunction Figure 43 WilcockCroweHidingFunction 13 10 estimateThickness OfMovingSedimentLayer realisations 13 10 1 Constant Thickness Of MovingSedimentLayer The Constant ThicknessOfMovingSedimentLayer Fig 44 assumes the thick ness of the moving sediment layer to be equal to the value in metres defined in the child node constantThickness with its default value of 0 7 This 53 rather large default has been chosen to avoid unnecessary slowing down of simulations estimateThicknessOfMovingSedimentLayer B realisationType ConstantThicknessOfMovingSedimentLayer constantThickness e cece ccc e 0 7 Figure 44 ConstantThicknessOfMovingSedimentLayer including default val ues 13 10 2 MultipleDiameterOfCoarsestGrain Moved The MultipleDiameterOfCoarsestGrainMoved Fig 45 takes the diameter of the coarsest fraction for which the local transport rate exceeds the thresh old defined by the node minimum TransportRateForFraction ToBe Considered Moving with its default value of 0 0001 and multiplies it with the value defined by the node factor with its default value of 1 75 in order t
60. s flowResistance methods are intended to determine the 3 in the following equation U 2 f VI rn Sp In this f is the Darcy Weisbach friction factor v is flow velocity g is grav itational acceleration ra is hydraulic radius and 8 is friction slope To deal with numeric issues all flowResistance realisations Fig 11 have the following optional nodes starting ValueForlteration accuracyFor Termi natinglteration maximumNumberOflterations and typeOfNumericRoot Finder Potential values for typeOfNumericRootFinder are BisectionMethod Secant Method FalsePositionMethod and RiddersMethod among which the Ridder sMethod is recommended The switch useApproximationsForHydraulic Radius defines whether the hydraulic radius should be approximated by flow depth By default it is true if ImplicitKinematic Wave is selected as waterFlowRouting and false in any other case If UniformDischarge is selected as waterFlowRouting the code will ensure that the Froude number will not exceed a maximum and that the hydraulic slope will not fall below a minimum The respective values are defined in the nodes mazimumFroudeNumber and minimumHydraulicSlope flowResistance L startingValueForlteration sseesseeseesese esee 0 4 accuracyForTerminatinglteration 0 02 0 e ee ee eee 0 001 maximumNumberOflterations 00 e eee eee eens 400 typeOfNumericRootFinder eee eee RiddersMethod useApproximationsForHy
61. scribed in section 9 and summarised in the Tables 7 amp 8 Predefined sediment inputs may be included in a simulation by additional files in the DischargeAn dOtherInputs folder as described in section 10 The influence of sills may be considered in a simulation by including an additional file in the Longitudi 12 nalProfile folder as described in section 11 The complete structure for the input xml file is described in the section 12 and an example of the complete xml structure is summarised in the Figures 6 to 10 9 Complete structure for BranchXProfile txt The complete structure of Branch X Profile trt is summarised in the Tables 7 amp 8 The columns KilometrageUpstreamDirected ElevationInM ChannelWidthInM and StrataGrainSizeDistribution have been described in section 5 The AlluviumThicknessInM defines the amount of erodible material in a reach If this column is not given the model will set its values to 4000 0 by default As described in section 5 the StrataGrainSizeDistribution defines the grain size distributions of the alluvium to be used at the start of a simulation It is possible to define an initial grain size distribution for the surface or active layer which may differ from the rest of the alluvium To do so one adds the column SurfaceLayerGrainSizeDistribution which defines the surface layer grain size distribution file names without the txt extension If this column is not given the code will use the file names of
62. sedFlow User manual Published by the Swiss Federal Research Institute WSL Written by Florian U M Heimann June 2014 Version 1 00 The manual is subdivided in two parts The Quick start guide from page 4 to page 11 gives the minimum information to run the model The Further specifications from page 12 to page 55 complement the short introduction of the Quick start guide and gives an in depth description of the different ways to adjust simulation parameters and output formats The manual provides an introduction to the use of the model sedFlow For an in depth description of the model please refer to Heimann et al 3 Examples of the application of sedFlow are given in Heimann et al 4 Copyright 2014 Swiss Federal Research Institute WSL Permission is granted to copy distribute and or modify this document under the terms of the GNU Free Documentation License Version 1 3 pub lished by the Free Software Foundation See http www gnu org licenses for further information Contents I Quick start guide 1 General remarks 2 Starting the model 3 The simulation folder 4 BranchTopology txt 5 BranchXProfile txt 6 Grain size distributions 7 BranchXDischarge tat 8 The main input xml II Further specifications 9 Complete Branch XProfile txt 10 Sediment input 10 1 Transport capacity input on nn nn 10 2 Sedigraph input x3 d es oer RR 10 3 Instantaneous sediment pulse input 11 Sills 12 Comple
63. simply sets the switch usePredefinedBreak Up Conditions to true If usePredefinedBreakUpConditions is true the node referenceMe dianDiameter defines the representative diameter for the calculation of 0 thetaCriticalForActiveLayer defines the value of Oca and thetaCriticalFor Sublayer defines the value of bes 45 strataSorting realisationType TwoLayerWithShearStressBasedUpdate dynamicLayerThickness 0 0 0 ccc eee eee eee ee nee false uselnitialGrainSizesForConstantLayerThickness true layerThicknessFactor 00 cece eee eee e eee ee 1 75 referenceGrainSizePercentile 0 cece eee eee 84 0 layerThickness only needed if neither dynamicLayerThickness nor useInitialGrainSizesForConstantLayerThickness dynamicBreakUpConditions 000 0 c cece eee eens false usePredefinedBreakUpConditions eee eee eee false L reference Median Diameter une only needed if usePredefinedBreakUpConditions thetaCriticalForActiveLayer cece eee ee only needed if usePredefinedBreakUpConditions thetaGriticalForsublayer u a only needed if usePredefinedBreakUpConditions thresholdCalculationMethod method from bedloadTransportEquations or default displayed there Figure 28 TwoLayerWithShearStressBasedUpdate including default values 13 5 4 SingleLayerNoSorting The only mandatory node of SingleLayerNoSorting Fig 29 is layerThick ness Of
64. slope minimumSlope nenn esl RKR RKR RR aa a aa a Figure 34 ReturnBedslope bed or waterEnergy or sedimentEnergySlopeCalculation Method realisation Ype ReturnWaterEnergyslope minimumSlop wii s cessive K K 9 b beh d RR a RRR KA Figure 35 ReturnWaterEnergyslope 13 7 tauCalculationMethod realisations The recommended realisation EnergyslopeTau Fig 36 calculates bed shear stress T according to the following equation P 6 9 Th S 24 50 The switch correctionForBedload WeightAtSteepSlopes selects whether the following correction of energy slopes according to Rickenmann 13 should be applied default or not sin dr y 25 5 sin dr Sy 25 tauCalculationMethod realisation Type aa aad hehe ta b RU Oo OCA Re arte an EnergyslopeTau correctionForBedloadWeight AtSteepSlopes true Figure 36 EnergyslopeTau including default values The companion realisation Energyslope TauBasedOnFlowDepth Fig 37 uses flow depth instead of the hydraulic radius r for the determination of T It can be used in the same way as Energyslope Tau tauCalculationMethod realisation Type iius rada EnergyslopeTauBasedOnFlowDepth correctionForBedloadWeight AtSteepSlopes true Figure 37 EnergyslopeTauBasedOnFlowDepth including default values 13 8 thresholdCalculationMethod realisations For all thresholdCalculationMethod realisations Fig 38 the
65. switch correc tion For Bedload Weight At Steep Counter Slopes selects whether the following correction of 0 according to Rickenmann 13 should be applied default or not sin br Sp sin In this Oek is the 0 corrected for the effects of bedload weight at steep adverse bed slopes Oek be 26 thresholdCalculationMethod E correctionForBedloadWeight AtSteepCounterSlopes true Figure 38 thresholdCalculationMethod including default values 51 13 8 1 ConstantThresholdForInitiationOfBedloadMotion The Constant ThresholdForInitiationOfBedloadMotion Fig 39 has one manda tory child node constantThreshold which defines the threshold which should be used in this simulation thresholdCalculationMethod realisationType ConstantThresholdForInitiation0fBedloadMotion constant Threshold ue rere Ee teh nett Figure 39 ConstantThresholdForInitiationOfBedload Motion 13 8 2 LambEtAlCriticalTheta The LambEtAlCriticalTheta Fig 40 calculates the threshold for the ini tiation of bedload motion according to the following empiric relation from Lamb et al 7 0 0 15 S975 27 In this equation gentle slopes may result in too low values of 0e Therefore the optional child node minimum Critical Dimensionless Shear Stress whith its default value of 0 03 defines a minimum value for e thresholdCalculationMethod realisationType 2 22222c22000 LambEtAlCriticalTheta minimumCriti
66. t 2 including default values 23 outputMethods standard including default values 31 outputMethods regular including default values 32 flowResistance including default values 33 FixedPowerLawFlowResistance including default values 34 VariablePowerLawFlowResistance 35 UniformDischarge including default values 35 ExplicitKinematicWave 2 2 2 nn nn 35 ImplicitKinematicWave including default values 36 bedloadTransportEquations including default values 37 RickenmannBedloadCapacityBasedOnTheta including default Wallies amp 2 2 mw Er Den ee em Jnd ran 5 39 58 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Al 42 43 44 45 46 47 48 RickenmannBedloadCapacityBasedOnThetaNonFractional in cluding default values len RickenmannBedloadCapacityBasedOnq including default val lese Sy rodent ar ahah doe Rt bue qe Hee moe Mul eo ec A ded ey RickenmannBedloadCapacityBasedOnqNonFractional includ ing default values lll WilcockCroweBedloadCapacity including default values ReckingBedloadCapacityNonFractional Velocity AsTranpsportRatePerUnitCrossSectionalArea includ ing default values lll JulienBounvilayRollingParticlesVelocity including default val DM te aby Ye Ae Pr Stratigraphy WithThresholdBasedUpdate incl
67. tains child nodes with the possible names reachID and branchID The branch IDs are used as defined in the branch Topology tat The reach IDs are used as defined in the reachIDExplanations tzt which is written to the Output folder by the model If the node reachIDsForOutput is not given the model will produce outputs for all reaches by default The node regularRiverReachPropertiesForOutput defines for which prop erties the values shall be written to the output files The node contains child nodes the names of which define the output properties With the standard outputs the code will create one file for each property if the switch create StandardOutputs is true default For the individual outputs all properties will be written into one file Potential values for the regularRiverReachProp ertiesForOutput are the following e elevation e sillOccurence e discharge e length e sillTopEdgeElevation e flowVelocity 25 mazimumWaterdepth e strataPerUnit e bedslope BedSurface waterLevel e waterEnergyslope f e strataPerUnit hydraulicHead BedSurface e sedimentEnergyslope kineticComponent IncludingPoreVolume unreduced OfHydraulicHead l SedimentEnergyslope e alluvium Thickness bedShearStress IncludingPoreVolume e waterVolumeChangeRate active Width e bedrockElevation e water VolumeChange activeLayerPer UnitBedSurface activeLayerPer UnitBedSurface IncludingPore Volume active LayerPer UnitBedSurface Overall Volume acti
68. tant descriptive variable of the accumulated bedload transport is defined as the temporal integral of the local transport rates Therefore the regularOutputX nodes are comple mented by the outputAccumulatedBedloadTransportX nodes which may be used to output the temporal integral of the bedload transport rates Within the regularRiverReachPropertiesForOutputStandard the pseudo property ac cumulatedBedloadTransport may be used to create an output file correspond ing to outputAccumulatedBedloadTransportX The outputAccumulatedBed loadTransportX contains the following child nodes which are used as de scribed for regularOutputX explicitTimesForOutput outputInterval preci sionForOutput reachIDsForOutput and writeLineEach TimeStep The switch outputincludingPore Volume defines whether the displayed volumes shall in clude pore volume or just represent the solid material The default for outputincludingPore Volume is true as any volumetric data recorded in the field commonly includes the pore volume as well By default the outputAc cumulatedBedload Transport will create a file with the overall transported volumes The switch outputDetailedFractional is used to create an addi tional file in which the transported volumes are specified for each grain size fraction separately To discriminate these two types of files the suffixes ABT OverallVolume or ABT DetailedFractional are appended to the by default empty file name which is spec
69. te structure of the main input xml file 12 1 coverallParameters s kso odes debo opos 12 2 riverSystemMethods 2e 12 3 outputMethods ls 12 3 1 Standard and regular outputs 12 3 2 ba kupX MD su goo eu RARO RR 12 3 3 outputSimulationSetup llis 13 Realisations 13 1 flowResistance realisations 22e 13 1 1 FiredPowerLawFlowResistance 13 1 2 VariablePowerLawFlowResistance 13 2 waterFlowRouting realisations 2 22 nn lle 10 12 13 14 14 14 16 16 17 18 20 24 24 29 30 13 2 1 UniformDischarge ee 13 2 2 ExplicitKinematicWave 22r 13 2 3 ImplieitKinematicWave 2l 13 3 bedloadTransportEquations realisations 13 3 1 Rickenmann bedload capacity ss 13 3 2 WilcockCroweBedloadCapacity 13 3 3 ReckingBedloadCapacityNonFractional 13 4 bedloadVelocityCalculationMethod realisations 13 4 1 VelocityAsTranpsportRatePer UnitCrossSectionalArea 13 4 2 JulienBounvilayRollingParticles Velocity 13 5 strataSorting realisations a aooo 13 5 1 StratigraphyWithThresholdBasedUpdate 13 5 2 TwoLayerWithContinuousUpdate 13 5 3 TwoLayerWithShearStressBasedUpdate 13 5 4 SingleLayerNoSorting lll 13 6 Gradient calculation method realisations 13 6 1 SimpleDownstreamTwoCellGradient 13 6 2 SimpleThreeCellGradient
70. ties display the actual local transport in m considering supply limitations For the ac tual local transport rate the erosion properties have to be divided by the current time step length The erosionPerUnitBedSurface properties display the erosion properties normalised to a column of 1m base area The de positionRate and deposition properties are usually equal to the erosionRate and erosion properties of the upstream reach If there are several upstream neighbours the depositionRate and deposition properties display the sum of the upstream values Local depositionRate or deposition may differ from the upstream erosionRate or erosion if the effects of gravel abrasion sec tion 13 11 are simulated Output file naming If an output file contains only one property the file will have the name of this property by default If a file contains several properties it will be named regularOutput or regularOutputX by default 28 corresponding to the name of its node These default names may be over ridden by the node name Please note that if a file name default or user defined is used several times several output routines will write to the same file creating cryptic artefacts outputAccumulatedBedloadTransport Within the regular and standard output files all values represent a snapshot of the property at a specified point in time That is the values are not integrated over an output interval or the complete simulation time However the impor
71. udeNumber 200 e cece ence eee eee 4 0 minimumHydraulicSlope 00 cee eee eee eee 0 0004 realisationType VariablePowerLawFlowResistance bedloadVelocity CalculationMethod realisationTypeVelocityAsTransportRatePerUnitCrossSectionalArea estimateThicknessOfMovingSediment Layer k realisationTypeConstantThicknessOfMovingSedimentLayer constant Thickness 000 00sec cece eee eee eens 0 7 tauCalculationMethod realisation TYDpe sd t APR Kahala EnergyslopeTau correctionForBedloadWeight AtSteepSlopes true activeWidthCalculationMethod L realisationType SetActiveWidthEqualFlowWidth outputMethods Figure 7 riverSystemMethods part 1 including default values 22 SEDFLOW_ StandardInput overallParameters riverSystemMethods waterFlowRouting LL realisationType lesse UniformDischarge maximumTimeStep 6 cece cece eee eee eens 900 bedloadTransportEquations maximumFractionOfActiveLayerToBeEroded 0 9 preventZeroOrNegativeBedSlopes if waterFlowRouting some KinematicWave maximumRelativeTwoCellBedSlopeChange 0 9 realisationType RickenmannBedloadCapacityBased nTheta useOnePointOneAsExponentForFroudeNumber false simplifedEquation sssssesseesseeesess eese true _thetaCriticalBasedOnConstantSred false thresholdCalculationMethod
72. uding default val les Sods Ma cea he Sent rd UE eise hse ae Mh TwoLayerWithContinuousUpdate including default values TwoLayerWithShearStressBasedUpdate including default val Hes ure E Eee au Boa ener eee Rut e FORO dre ER DAR SingleLayerNoSorting including default values SimpleDownstreamTwoCellGradient SimpleThreeCellGradient 2 ln ReducedWaterEnergyslope including default values ReducedWaterEnergyslopeNotUsingWaterEnergyslopeVariable including default values lll ReturnBedslope nennen ReturnWaterEnergyslope 222 nn EnergyslopeTau including default values EnergyslopeTauBasedOnFlowDepth including default values thresholdCalculationMethod including default values Constant ThresholdForInitiationOfBedloadMotion LambEtAlCriticalTheta including default values PowerLawHidingFunction including default values NoHiding oer wa la A ee SE WilcockCroweHidingFunction lees Constant ThicknessOfMovingSedimentLayer including default values osos oh dH Di hh A Ae EE MultipleDiameter OfCoarsest Grain Moved including default val ESLa Serle use d ats eate Ree dne OM eoe le Bol OT eR d MultipleReferenceGrainDiameter including default values SternbergAbrasionWithoutFining SternbergAbrasionIncludingFining 59 39 40 40 Al Al 42 42 44 45 46 46 47
73. ughness length 7 is the bed shear stress rp is the hydraulic radius D is the representative roughness diameter percentile with x defined by the node roughnessDiameterPercentile with its default value 84 0 and a is an empiric factor defined by the node roughnessDiameterFactor with its default value 3 5 bedload VelocityCalculationMethod realisationType JulienBounvilayRollingParticlesVelocity roug hness Diameter Percentile 0 0 cece eee een eee 84 0 roughnessDiameterFactor L 3 5 Figure 25 JulienBounvilayRollingParticlesVelocity including default values 13 5 strataSorting realisations 13 5 1 Stratigraphy WithThresholdBased Update The StratigraphyWith ThresholdBasedUpdate Fig 26 represents the river alluvium by a pile of sediment layers The number of layers is defined by the node numberOfLayers The thickness of the topmost and lowermost layer may vary All other layers have the same thickness which is defined by the 42 incrementLayer Thickness Whenever the variable thickness of the topmost or active layer exceeds an upper threshold or falls below a lower threshold sediment is sorted down or upward until the thickness of the active layer again has a value in the middle of the thresholds In the simplest case these thresholds have constant values which are defined by the nodes val ueOrFactorForLowerThresholdForActiveLayer Thickness and valueOrFactor ForUpperThresholdFor ActiveLayerThickness If the switch uselnitialGrai
74. umeFraction defines the volumetric portion of pores and fine suspension load sediment with its default value of 0 3 The gravityAcceleration given in 23 has a default value of 9 80665 The angleOfReposeInDegree has a default value of 36 0 The elapsedSeconds and finishSeconds define the start and end of the simulation By default they are set to the minimum and maximum of the ElapsedSeconds column of the Branch X Discharge trt with X equal to the smallest used branch ID The courantFriedrichsLewyNumber is used to determine numerically sta ble time step lengths It represents the maximum potential value for the ratio Ue in which v is velocity and Ax amp At are spatial amp temporal dis cretisation The default value for the Courant Friedrichs Lewy number is 18 SEDFLOW StandardInput overallParameters inputUpperBoundaryInsteadOfMeanGrainDiameter true lowerDiameterBoundaryForFinestFractionInCM useArithmeticMeanInsteadOfGeometricMeanForFractionGrainDiameters false density Water ee eee desee tbe t UE a na 1000 0 denstySediment iiec eer 4b Ee eR EUR HO IERI LAC 2650 0 poreVolumeFraction 0 00 c cece eee ence een 0 3 praviby Aocaleradiona aan icesrnccepandteiwddewededdeadedads 9 80665 angleOfReposeInDegree seesesseese e 36 0 L elapsedSeconds Min of first discharge file finishSeconds da eene odd Max of first discharge file courant Fri
75. ure of SedimentInputs txt part 1 BranchIDs KilometrageUpstreamDirected GrainSizeDistribution 22 22 22 14 Table 10 Structure of SedimentInputs txt part 2 including default values ReplacingRegularDepositionRate SedimentInput TimeSeries InsteadOfAddingTolt InputIncludingPoreVolume 22 false true The BranchIDs and KilometrageUpstreamDirected define the crosssec tion at which the sediment is fed to the system So if sediment enters the river channel between two crosssections one would simply input the branch ID and kilometrage of the more downstream profile If the code cannot match branch ID and kilometrage to an existing profile it will simply ignore the corresponding row in the SedimentInputs tat The GrainSize Distribution defines the file name without the txt extension for the grain size distri bution of the sediment input The code will look for the specified files in the GrainSizeDistributions folder Please note that the grain size distribu tion of the sediment input stays the same for the complete simulation It will not become coarser at high input rates and it will not become finer at low input rates To mimic such behaviour one may either use the transport capacity input as described in section 10 1 or one may feed multiple sedi ment inputs with different grain size distributions to the same reach The SedimentInputTimeSeries define the file names of the
76. utputInterval defines the interval in seconds at which output rows shall be written The writeLineEachTimeStep may be used to enforce the output of every simulated time step This option is intended for debugging and not recom mended for daily use In sedFlow it is possible to use a secondary output interval which is applied when some property like e g discharge exceeds a certain thresh old This may be used to e g increase output frequency during floods For the standard the secondary interval is used when one of the four nodes referenceCelllDStandard referencePropertyStandard threshold To BeExceed edStandard or secondaryOutputIntervalStandard is given For individual out puts the secondary interval is used when the node SecondaryOutputInterval is given which contains the four mentioned nodes The node referenceProp erty defines the property for which it is checked whether it exceeds some threshold The node reference CellID defines the ID of the reach at which it is checked whether the reference property exceeds a threshold The node threshold ToBeExceeded defines the threshold which needs to be exceeded for the application of the secondary output interval Finally the secondary OutputInterval defines the output interval which is applied whenever the threshold is exceeded Selecting output reaches and properties The node reachIDsForOutput is used to select the reaches for which output should be written to files This node con
77. ve LayerPer UnitBedSurface Overall VolumeIncluding Pore Volume activeLayerPer UnitBedSurface MedianGrainDiameter erosionRate erosionRateIncludingPore Volume erosionRate Overall Volume erosionRate Overall Volume IncludingPore Volume erosionRateMedianGrainDiameter erosionRateArithmeticMean GrainDiameter 26 activeLayerPerUnitBedSurface ArithmeticMeanGrainDiameter activeLayerPerUnitBedSurface GeometricMeanGrainDiameter activeLayerPerUnitBedSurfaceD20 activeLayerPerUnitBedSurfaceD30 e activeLayerPerUnitBedSurfaceD50 activeLayerPerUnitBedSurfaceD64 activeLayerPerUnitBedSurfaceD90 erosion Rate GeometricMean GrainDiameter erosionRateD20 erosionRateD30 erosionRateD50 erosionRateD64 erosionRateD90 depositionRate e depositionRateIncluding Pore Volume depositionRate Overall Volume K depositionRate Overall Volume IncludingPore Volume e depositionRateMedian GrainDiameter e depositionRateArithmeticMean GrainDiameter e erosion e erosionIncludingPore Volume erosionOverall Volume erosionOverall Volume IncludingPore Volume erosionMedianGrainDiameter erosionArithmeticMean GrainDiameter e deposition P depositionIncludingPore Volume depositionOverall Volume be depositionOverall Volume e IncludingPore Volume e depositionMedianGrainDiameter e depositionArithmeticMean GrainDiameter e erosionPerUnitBedSurface erosionPerUnitBedSurface e 27 depositionRate Geom
78. woLayerWithShearStressBased Update section 13 5 3 by default The optional node additionalMethods may be added to include further methods which are not covered by the aforementioned riverSystemM ethods nodes Up to now the only additionalMethods are concerned with the effects of gravel abrasion section 13 11 21 SEDFLOW StandardInput overallParameters riverSystemMethods upstream OfSillsWedgeShapedInsteadOfParallelUpdate false bedSlopeCalculationMethod realisationType SimpleDownstreamTwoCellGradient propertyOflnterest 0 cece eee eee eens elevation waterEnergySlopeCalculationMethod L realisationType 0c cece eee eee ReturnBedslope minimumSlope nun ee eee Rh aan inf sedimentEnergySlopeCalculationMethod L realisation Type sues WithFlowResistancePartitioning _stressPartitioningExponent K 1 5 I calculationBasedOnqInsteadOfh false maximumFroudeNumber value from flowResistance minimumInputSlope 6 cece eee eee eens ensureMinimumInputSlope if minimumInputSlope node exists flowResistance startingValueForlteration 0 ee eee eee eens 0 4 accuracyForTerminatinglteration lt e 0 001 maximumNumberOflterations cece eee eee 400 _typeOfNumericRootFinder RiddersMethod useApproximationsForHydraulicRadius If waterFlowRouting ImplicitKinematicWave maximumFro

sedFlow - User manual (Version 1.00)

Contents

Download Pdf Manuals

Related Search

Related Contents