Advanced parameters
    • 07 Aug 2022
    • 4 Minutes to read

    Advanced parameters


    Article summary

    This section summarises the advanced parameters available to the 2D solver user, which have not been previously discussed. It is strongly recommended that users do not adjust these from their default values unless they understand the function of the parameter and its likely effect on model results.

    Parameter

    Solver

    Description

    See section

    Beta

    ADI/TVD

    Multiplier for the advection term to allow for velocity varying with depth. For typical floodplain flows this should be in the range 1-1.03.

    Can also be set to zero to switch off the advection term if this is causing problems and only an approximate solution is required.

     

     

    Calculation depth

    ADI/TVD

    Water above this depth uses the simplest approach to calculating depths at cell edges: average of the water levels either side minus the ground height. Below this a more sophisticated approach is used. Users may need to increase this parameter for modelling over steep topography, typically when the ground slope is big enough to make elevation differences between cells > 0.5m.

    The ADI solver

    Depth threshold

    ADI/TVD

    The depth below which friction is increased to reduce negative depth problems on steep drying slopes

    Wetting and drying

    Eddy viscosity

    ADI/TVD

    Factor controlling horizontal turbulent and viscous momentum transport. Typically ranges between 0 and 1.

    Shallow Water Hydraulics

    Nadvit

    ADI

    Controls the number of iterations used to deal with non-linearity in the Shallow-Water equations. Increasing this may improve stability; reducing it below the default value of 2 may result in an unstable model.

    The ADI solver

    -ve depth threshold

    ADI/TVD

    Depths below this negative number are corrected by balancing with water from neighbouring cells. Increase this value if negative depths are acceptable; this may improve stability.

    Wetting and drying

    New flooding and drying

    ADI/TVD

    Switches on the more sophisticated approach for calculating depths at cell edges, as parameterised by Calculation depth.

    The ADI solver

    Rainfall acc. Depth

    ADI/TVD

    Depth to which rain must fall before a "pulse" is added to a cell in direct rainfall models. This may need to be increased for models with very high topography; should only be adjusted if mass balance errors are encountered.

    Direct rainfall

    Slip wall flag

    ADI/TVD

    Specifies whether a slip or no slip condition is used at walls and dry cells. This will only have an effect on flow if a non-zero eddy viscosity is used.

     

    Upwind flag

    ADI/TVD

    Switches on upwinding in the advection term calculation. Switching this off may result in a slightly more accurate solution, but may produce an unstable model when the advection term is significant.

     

    Velocity head threshold

    ADI

    The velocity head above which advection capping is used. Choose a very high value (>=5) to switch off advection capping; choose a lower value where accelerating flows are causing instabilities. However, note that a lower value may reduce the accuracy of the model output  in the areas of your model with accelerating flows. If this area is an important part of your model then you could consider using the TVD solver instead (which tends to perform better for accelerating flows).

    Modifications to the advection term

    Water depth (dry)

    ADI/TVD

    Depth threshold at which a cell becomes wet. A smaller value may improve accuracy for very shallow flows (e.g. direct rainfall), at the expense of stability. A larger value may improve stability; flows below this depth will be routed using the kinematic wave approximation.

    Direct rainfall

    Water depth (friction)

    ADI/TVD

    The water depth below which the ADI scheme switches to the semi-implicit treatment of the non-linear friction terms. The default value is high to ensure the semi-implicit method is used nearly all the time. Reducing this value may help model stability for coastal flows.

    The ADI solver

    Convergence criteria

    FAST solver

    Controls convergence of the 2D FAST Solver algorithm. Reduce this for a more accurate, but slower solution.

     

    Discretisation

     

    FAST solver

    Water level – volume curves for each depression are stored at heights separated by the Discretisation parameter. Increasing this will reduce file size, and reduce accuracy.

    DTM pre processing

    Filter

    FAST solver

    Controls a preprocessing stage to filter identical ground heights in neighbouring cells.

    DTM pre processing

    Merge factor

    FAST solver

    Topographic depressions are merged if the “lip” between them is below this height. Increasing this value will lead to fewer depressions, and fewer, larger flooded areas.

    DTM pre processing

    Padding height

    FAST solver

    Depth to be filled in along flow path

    2D FAST Solver

    Padding width

    FAST solver

    Width in cells of flow path

    2D FAST Solver

    Pre-process DTM

    FAST solver

    Switch this on to force 2D FAST Solver to recalculate topographic depressions; switching off will cause 2D FAST Solver to use previously calculated depressions, meaning a shorter computation. Switch this to off after the first run of 2D FAST Solver, unless you change the topographic processing parameters.

    DTM pre processing

    Probable path mark-up

    FAST solver

    Switch on marking of drainage paths in model output

    2D FAST Solver


    Was this article helpful?

    Changing your password will log you out immediately. Use the new password to log back in.
    First name must have atleast 2 characters. Numbers and special characters are not allowed.
    Last name must have atleast 1 characters. Numbers and special characters are not allowed.
    Enter a valid email
    Enter a valid password
    Your profile has been successfully updated.
    ESC

    Eddy AI, facilitating knowledge discovery through conversational intelligence