2D Simulations - Boundary Conditions Subtab

A 2D Solver model requires input data to apply water into (or out of) the modelled region. This may be one or more of the following: link from a 1D model, hydrology/rainfall source or flow/elevation data at an external or internal boundary. The following parameters can all be defined as boundary data:

• Water elevation (ADI/TVD solvers only)
• Inflow
• Inflow per unit width
• Depth
• Normal depth
• Outflow (recommended for GPU solver as alternative to normal depth)
• Weir - FAST solver only (recommended alternative to water elevation)
• Water level - FAST solver only (equivalent to water elevation)

Hydrological inputs (e.g. rainfall) are covered in the next section.

The Flood Modeller 2D Solver allows the user to locate boundary datasets within your model by associating data to a shapefile feature, which can be a point, line or polygon. In this case if you are associating shapefile features to Flood Modeller node names in a separate text file (see below for details), then your shapefile must have an attribute field entitled ‘node’ containing the cross referencing to the node names in your text file. At runtime different data types associated to a feature will be treated differently:

• Flow data will be divided equally between the cells ‘crossed’ by your shapefile feature, e.g. a 10m3s-1 inflow assigned to a shapefile line crossing 10 grid cells will result in a 1 m3s-1 into each cell
• Depth and level data values will be assigned to all cells ‘crossed’ by your shapefile feature, e.g. 2m water depth assigned to a shapefile line crossing 10 grid cells will result in all 10 cells being assigned a 2m depth of water.

Different options are also provided to define the actual data assigned to each boundary:

• You can specify a single constant value, assigned to a shapefile feature.
• You can enter a time series directly into the Flood Modeller interface. A blank table for times and values is provided for each boundary input defined. This can be typed into one value at a time or a complete time series can be defined in a spreadsheet and then copied and pasted into the Flood Modeller user interface. Again this can be assigned to a shapefile feature. Note that a time unit must also be defined in the Flood Modeller interface for time series data entered here.
• You can cross reference boundary input locations with time series defined in a separate text file. This option is only available to link data to shapefile features. Node names must be cross referenced between your text file (in the headers of time series) and the associated shapefile (in an attribute field entitled ‘node’). The time series text files can take either of the following formats:
  • A Flood Modeller 1D event file (text file with extension ‘.ied’) – this is a fixed format text file, i.e. each entry must occupy a set space, which can contain one or many time series datasets. Editing a 1D Network in the user interface will ensure event files are created in the correct format. Furthermore the Flood Modeller user guide provides a definition of this file format. An example of the format is provided below:
    
  • A comma separated text file (with extension ‘.csv’) – this is a standard text format in which each entry is separated by a comma. Each row must contain a time/date plus one or more values (applying to one or more nodes). The first row contains column headers which are ‘time’ and then the node name references for each subsequent dataset. You can prepare your time series in a spreadsheet application (for example, Microsoft Excel) and then save directly to csv format. An example of the format is provided below:
         Time,     Node1,   Node2,   Node n
           T1,        N1v1,     N2v1,      Nnv1
           T2,        N1v2,     N2v2,      Nnv2
           Tn,        N1vn,     N2vn,       Nnvn
• You can cross reference boundary input locations with a 1D Solver hydrological boundary file. This option is only available to link data to shapefile features. Node names must be cross referenced between your 1D Solver file (in the headers of each hydrological boundary node) and the associated shapefile (in an attribute field entitled ‘node’).

The 1D Solver hydrological boundary will contain hydrological parameters rather than time series data. When you start your simulation 2D Solver will use these parameters to calculate a time series input to be applied over the areas of your model defined by the associated shape file.

Setting Boundary Inputs in the 2D Simulation window

For each domain in your model, you can specify one or more flow, depth or water level boundaries. These data can be specified as single values or as time series. They are located within your domain by associating your boundary data to a shapefile feature. This section details how these data are specified within the 2D Simulation window.

All boundary data are specified on the Boundary Conditions tab (a sub-tab within the Domains tab). When first accessed this looks like the example below, displaying one blank table:

Defining a new boundary input

To add a new boundary to your model, click the Add button on the Boundary Conditions tab. A window will open. This window is shown below:

This window provides a series of options to define your new boundary input:

1. Boundary condition – here you can select your boundary parameter. It is selected from a dropdown list of variations of flow and water depth (or level) available for inputs:
   
   Boundary conditions data types are all variations of flow, water depth or water level. The data types that can be utilised as 2D Solver boundary inputs are summarised as follows:
   • Water elevation – water level data, elevations above a datum specified in metres or feet (depending on your system setup, i.e. SI or US units). This boundary type is only available when solver type is set to ADI or TVD.
   • Inflow – flows to represent sources coming from underground, e.g. sewer overflows (at manholes) or overflows from flooded underground structures (tunnels, basements, etc). These types of flow inputs will be treated slightly differently in the 2D Solver calculation engine and so may yield different results. Units are the same as total flow.
   • Inflow per unit width – vertical flow converted into a per unit width value. 2D Solver will multiply values by the width of your boundary to determine actual flows entering each cell of your model domain.
   • Depth – water level data converted to actual depth (by subtraction of underlying ground level). Values must be specified in metres or feet.
   • Normal Depth – calculates the normal depth-flow relationship at a boundary and is usually used as an outflow boundary. The location can be specified by a shapefile (point or polyline) and user-defined slope, or if a shapefile is not specified, it may apply anywhere around the domain boundary. The latter (free location) case is useful when the outflow location is not known a priori, but results should be interpreted with care if there exist multiple outflow locations since the relationship is calculated for the entire domain boundary; in the free location case, the slope is not entered, but calculated as the water surface slope. In either case, the boundary is only activated when water reaches the boundary and the relationship is only calculated using those boundary cells, which are wet.
   • Outflow - The outflow boundary condition can be set to establish a line where the flow can exit the active area. Two modes can be applied, the free exit and the weir-flow control. This option is recommended if the Normal Depth boundary is proving unstable for your setup.
   • Water level - water level data are the same as water elevation boundary type, but this applies to FAST simulations only (not available for ADI/TVD). However, it is not recommended to use a water level boundary with the FAST. Instead the preferred equivalent option to water elevation in FAST is the weir type boundary.
   • Weir - water level data applied to the 2D domain in conjunction with a standard weir equation (to calculate an inflow). This option is only available for FAST simulations. It is recommended to use this boundary type as an equivalent to a water elevation boundary (if switching solver type from ADI or TVD to FAST). 
     Note

     FAST simulations will run if specified with a water elevation boundary. The solver in this case will assume the boundary is a weir type. However, the 2D model interface does not offer this option if the solver type is set to FAST (so this scenario would only occur with older 2D models or models edited outside of Flood Modeller using a text editor).
2. Boundary location – Here you need to specify a boundary shapefile, either by typing a filename (with full path) into the space provided or by using the adjacent button to browse to your shapefile.
   
3. Boundary data – the default setting assumes that your boundary input will be constant flow in cumecs. If this is the case you can type a single constant value for your boundary input. This value will then be applied continuously throughout your subsequent simulation.
   The other options for assigning data to your boundary are as per the radio buttons on the "Add/Edit Boundary Condition" form shown earlier:
   • 1D boundary unit file (dat/ied) – this option allows you to enter an 1D Solver hydrology file into the space provided (ied or dat file extension). This has to be entered with full path. A browse button is also provided to help you browse to and specify the required file.
   • Time series file (csv/ied) - this option allows to enter a timeseries data file. This can either be typed into the space provided or you can use the adjacent button to browse to your text file, which must be either a Flood Modeller 1D event file ('.ied') or a comma separated values file ('.csv').
   • User entered time series – this option enables you to define a time series in the data table provided on the form under this radio button (see below for details). You can then either type in your data or paste it in from another application, e.g. MS Excel spreadsheet.
   • Constant inflow (cumecs) - this option enables a box into which you can then type a single constant value for your boundary input. This value will then be applied continuously throughout your subsequent simulation.
   • Outflow discharge coefficient - this option is only available when using the outflow boundary condition. Radio buttons allow for selection between free-exit and weir-control modes. For both options, the discharge coefficient can range between 0 and 2.
     The free-exit mode allows the flow to exit the active area assuming the downstream end is always dry. A discharge coefficient of 1.2 (the default) allows the flow to exit completely freely. A coefficient between 0 and 1.2 allows flow at the rate of   where  is the discharge coefficient, and  is the water depth.
     The weir-control mode mimics a weir flow control point. The boundary condition regulates the exit flow assuming the downstream end has the water elevation calculated from the previous time step. In weir-control mode, the discharge coefficient can be determined based on the Bernoulli equation.
   • When a boundary is defined with a boundary data option set to ‘User entered time series’ then the time series table will be enabled on the Add/Edit Boundary Condition form. To enter a value in here, simply click a cell and type in your coordinate value. Then press the Enter or Tab key to commit a value. When an X and Y coordinate pair are completed in the table, a new blank row will automatically be created ready for your next entry.
     
     This table also provides a menu accessed by right-clicking on the table.
     
     This provides the following functions:
     • Copy All - copy all rows from the table to the clipboard. Copied rows can then be pasted into another application, e.g. MS Excel, to edit further.
     • Copy Selected – copy the highlighted rows in the time series table (to enable pasting elsewhere). Use the Shift and Ctrl keys in conjunction with mouse to select multiple rows.
     • Paste & Append – paste a time series into your table, e.g. a time series created and copied from an Excel spreadsheet. If your table already contains data, then such an operation will append the new data to the existing data.
     • Paste & Replace all – paste a time series into your table, e.g. a time series created and copied from an Excel spreadsheet. If your table already contains data, then such an operation will replace the existing data with the new data.
     • Paste & Replace Selected – paste a time series into your table replacing the existing data in selected rows only. Use the Shift and Ctrl keys in conjunction with mouse to select multiple rows. This operation will append the new data to existing data that has not been selected.
     • Remove Selected Row(s) – removes all selected rows from the table. Use the Shift and Ctrl keys in conjunction with mouse to select multiple rows. You will be prompted to confirm before deletion.
     When a boundary is defined with a boundary data option set to ‘User entered time series’ or 'Time series file (csv/ied)' then a menu "Units of time are:" is enabled for your data series. This is set by selecting from the provided options in a drop-down list located above the chart (this chart is not greyed out only when the time series table is also enabled).
4. When your new boundary settings are defined click the OK button (or click Cancel to discard the boundary). The new boundary will be added to the Domain boundaries table on the Boundary Conditions Tab in the 2D Simulation window.
   
   When the new boundary is highlighted in the boundaries table, it is possible to edit this boundary by either double-clicking on the line or clicking on the Edit/View button at the bottom of the screen. Then you can modify the boundary by editing the values on the Add/Edit Boundary Condition form and clicking OK and subsequently confirming that you would like to Save the changes on the window that opens.