Section 11 Outputs

Prior to carrying out a simulation the desired output needs to be configured and customised. For viewing outputs, see Chapter 15. There are also a variety of post-processing utilities available, see Chapter 17.

The chapter includes discussion on:

  • Setting the folder location of where outputs are written (Section 11.1).
  • Configuring map output options including 1D results in a 2D form and the powerful output zones feature (Section 11.2).
  • Options for setting up time-series output from 1D and 2D domains at specified locations (Section 11.3).
  • Specialised outputs such as time and depth of inundation along evacuation routes (Section 11.4).
  • Options for customising the check and log file outputs (Section 11.5).
  • List of commands available for customising all the different output options, including output to the simulation console window, check files and folder locations (Section 11.6).

11.1 Output Location

To set the location of where the outputs are written to, the Output Folder command is used. For example “Output Folder == <folder>” redirects all TUFLOW output data except the log and summary files to the specified folder.

Outputs can be written in a different location to where the model sits. It is typically recommended to write outputs to your local C: or D: drive instead of filling up the network drive, or to keep results separate to the input data. A URL can be used (e.g. \myserver), which is useful for running simulations on other computers, but with the output directed to your local drive (your drive will need to be shared) or other location.

The Output Drive command can be used to change the drive letter of any output files with a full path specified. For example “Output Drive == K” changes any output filepaths that have a drive letter to the K drive.

11.2 Map Outputs

TUFLOW has a range of commands that allows the user to control the formats, data types, frequency, sub-areas (output zones) or resolution (SGS high-resolution) of a simulation’s output. The options are highly flexible and can be customised to be different for different output formats and/or output zones.

For example, time-based map output may be produced using one format (e.g. XMDF), and the peak flood level surface for another output (e.g. TIF). Output over a township maybe at a higher output interval frequency than for the whole model so as to produce a higher quality animation of flooding through the township, or if using SGS, at a higher resolution (e.g. HRTIF).

The main commands to control map outputs are:

For example, application of the above commands may look like:

Output Folder == ..\results\ ! Location of the output files
Map Output Format == XMDF TIF ! Produce map output in XMDF and TIF formats
Map Output Data Types == h V d ! Output water level, velocities and depths
Start Map Output == 10 ! Start map output at 10 hours
Map Output Interval (s) == 300 ! Write map output every 5 minutes (300s)
TIF Map Output Interval (s) == 0 ! Output maximums only for TIF format

11.2.1 Output Time Controls

To control when and how frequent outputs are written, time control commands are used. They can be set as a global setting or customised for each output format. A variety of formats can be output for a single simulation, the supported formats are discussed in Section 11.2.2.

To set default settings the followings commands can be used:

To customise the time control commands to be different to the default or global settings, the map output format acronym is included at the start of the command. For example, “XMDF Map Output Interval == 360” will set the output frequency to 360s just for the XMDF format. The format acronym must be identical to that used in Map Output Format. The following commands can be applied in this manner.

To apply the same setting to more than one format, the command needs to be repeated for each format. If a format is not customised the default setting or the setting applied to the whole of the model is used. The order of commands is important, ensure that commands defining the map output settings for all format types are read in prior to any commands specific to a certain format type. For example:

  • The following commands set a Map Output Interval of 120 seconds for the XMDF format and 3600 seconds for the TIF format:
    Map Output Format == XMDF TIF
    Map Output Interval == 120
    TIF Map Output Interval == 3600

  • Swapping the order of the second and third command lines will set a Map Output Interval of 120 seconds for all output formats as the third command overwrites the TIF Map Output Interval command.
    Map Output Format == XMDF TIF
    TIF Map Output Interval == 3600
    Map Output Interval == 120

Of note is that each output is tracked every timestep for its maximum if Maximums and Minimums is set to “ON” or “ON MAXIMUMS ONLY” (the default) to ensure that the peak result (water level, depth, hazard etc.), independent of the time it occurred during a simulation, is recorded.

11.2.2 Map Output Formats

TUFLOW offers a wide range of map output formats, all non-proprietary, to cater for a range of GIS and GUI software. There are no constraints over how many output formats a single simulation can produce. Different formats can have different output settings, and by using Output Zones different regions of the model can output in different formats.

Map output is offered in the following forms:

  • Mesh Based: The output is based on a quadrilateral and/or triangular mesh of the 1D and 2D domains. This output is the closest reproduction of the hydraulic calculations, with minimal interpolation from the 1D and 2D computational points. These formats include: XMDF, DAT, TMO and WRB.
  • Grid Based: The output is over a regular north-south grid in a similar manner to a raster DEM. The hydraulic output at each output grid cell is interpolated from the computational mesh using the Mesh Based output above. These formats include: TIF, GPKG, FLT, ASC, NC, TGO, WRR.
  • High Resolution Grid Based: If using SGS, the retained sampled elevations can be used to produce high-resolution (HR) grid outputs. All supported Grid Based outputs can be high-resolution: HRTIF, HRGPKG, HRFLT, HRASC, HRNC, HRTGO, HRWRR.
  • Combination Mesh and Grid: This format includes: WRC.
  • GIS Based: GIS layers written as the simulation proceeds (similar to using the post-processing TUFLOW_to_GIS utility). The format is set by GIS Format.

The selection of formats is controlled by the TCF command Map Output Format. One or more formats may be specified for the whole model or for an Output Zone (refer to Section 11.2.5).

If no output format has been specified, the results are written by default using the XMDF format. The following sections describe the supported map output formats.

11.2.2.1 Mesh Based Formats

Mesh based map outputs use a mesh of quadrilaterals and/or triangles of the 2D cells and any 1D WLLs to manage and store the model results.

The recommended mesh format is the XMDF.

Additional time outputs are available using Time Output Cutoff Depths or Time Output Cutoff Hazards to output maps of the duration of inundation and the time of first inundation above specified depth(s) or VxD(s).

The options in the table below only apply to the XMDF format. Only one (or none) can be specified.

Note that the Map Output Format SMS HIGH RES option is a legacy feature no longer supported and is, therefore, not included in the table below. For details on the SMS HIGH RES feature search the 2018 TUFLOW Manual.

11.2.2.1.1 Mesh Configurations

Quadrilateral and Triangle Mesh Option

The default mesh used by the XMDF format uses quadrilaterals (squares) for the 2D cells and triangles for representing any 1D WLLs as discussed in Section 11.2.4. Each 2D domain and the 1D WLL triangles are treated as separate meshes, although they will appear as one when viewing the .2dm file.

The advantage of this format is that the amount of data output is optimised keeping file sizes small with little loss of accuracy in translating results from the computational mesh.

Only the XMDF format supports the combination of quadrilaterals and triangles.

Triangular Mesh Option

Except for the XMDF format, all the other mesh-based formats use a triangular mesh to represent the computational 1D and 2D domains. The triangular mesh is also used by the grid-based formats (TIF, FLT, ASC, NC, TGO, WRR) to interpolate from the triangular mesh the values at the grids cells’ centres (this is the cell centres of the output grid, not the 2D domain grid cells).

To utilise this format using the XMDF format, specify SMS TRIANGLES anywhere in the arguments for the Map Output Format command.

The triangular mesh option incorporates output at the cell centres, so the exact water level calculated by TUFLOW at the 2D domain cell centres is used when translating results, thereby providing a slightly more representative surface of the hydraulic calculations. Each 2D cell is represented as four triangles with a common vertex at the cell centre, giving a higher resolution spatial representation than just using the 2D cell corners.

However, the number of nodes in the mesh for the 2D domains increases by 20% and the number of elements by a factor of 4, therefore, the output file sizes will be larger and the viewing and processing times possibly slower. Any triangular elements from 1D WLLs are not affected and remain the same in either mesh arrangement.

2D Cell Corner Interpolation/Extrapolation

Irrespective of the mesh option being used, the Map Output Data Type values at the 2D cell corners need to be interpolated (or extrapolated if at the wet-dry interface). Two methods were developed that can be set using the command Map Output Corner Interpolation == METHOD B or METHOD C. Method C (the default) is recommended as it is simpler and extensive testing has indicated it resolves issues associated with prior methods. METHOD B does largely resolve earlier issues, and has the added advantage that the effect of thin breaklines is better handled when they are dry or flow is upstream controlled.

For model results where the water has risen upwards (e.g. river flooding), the Method B or C approaches should cause no significant changes in results (i.e. fractions of a mm). Where the flow is downwards over steep slopes, some changes in results at cell corners will occur, but usually only slightly. However, maximum hazard values on very steep slopes may experience a more significant change.

11.2.2.2 Grid Based Formats

Grid-based formats output the results over a regular grid in similar formats to DEMs and other raster-based data. The output grid resolution and origin is not necessarily the same as the 2D hydraulic modelling grid, and will also use any 1D WLL triangulations to include map based output of 1D domains. Grid based formats are highly suited for use in GIS, and especially for handling output from very large models (e.g. GPU models).

The available grid formats are listed in Table 11.3. The output grid format is set with the Map Output Format command, for example:

Map Output Format == TIF

Alternatively, instead of including a specific format (e.g. TIF or FLT) in the Map Output Format list, this can also be specified using the Map Output Format == Grid command. In this case the grids are written by default in TIF format. This can be changed to other grid formats by specifying the specific format in the .tcf command Grid Format. If Write Check Files is used, the DEM_Z and DEM_M grid outputs will be of the format specified by the Grid Format command (or TIF by default). Note that not all formats support compression, so those that don’t can produce considerably larger file sizes.

The output grids are placed in a sub-folder called “grids”. For TUFLOW Classic the default output grid resolution is half the smallest 2D cell size (considering that multiple 2D domains may exist). For TUFLOW HPC the default is half the 2D cell size. For HPC models using Quadtree, the output resolution is the smallest 2D cell size. The resolution of the output grid can be controlled by the Grid Output Cell Size command. Only one output grid resolution is possible. If multiple commands exist, the last occurrence of the command will prevail. The origin of the output grid is rounded to the nearest cell size. This can be set to the exact model origin using the Grid Output Origin == Model Origin.

The TUFLOW grid output formats (listed in Table 11.3) support all available Map Output Data Types (listed in Table 11.4 and Table 11.5). The minor exceptions are noted in Table 11.3.

Grids outputs store data on a grid by grid basis and in a north-south direction, therefore interpolation must occur to convert the data from the TUFLOW calculation points. The raster data is interpolated from the values at the cell corners and centres.

For example, the image below shows:

  • A _grd_check file (black lines)
  • A _zpt_check showing the TUFLOW calculation (ZC, ZU, ZV and ZH) points;
  • An output water level TIF grid (underlying grey grid); and
  • Orange dotted lines showing the triangulation between the cell corners and centres.

The raster data (grid centre indicated by red cross) obtains its value from the triangulation between A, B, and E.

To more closely align the grid output with the TUFLOW calculation points:

The <format> Map Output Interval can be set to zero as follows to trigger only outputting the maximums for those formats. For example:

TIF Map Output Interval == 0 ! Only output result maximums for TIF grids
FLT Map Output Interval == 0 ! Only output result maximums for FLT grids

All time outputs are supported by the grid map output formats. The grid file extensions are:

  • _TDur_<cutoff> for duration of inundation;
  • _TExc_<cutoff> for time <cutoff> is exceeded;
  • _TMax_h for time of peak water level; and
  • _TMax_V for time of peak velocity.

11.2.2.3 High-Resolution Grids

When using the Method C sub-grid sampling (SGS) approach the sampled elevations are retained including topography modifiers such as breaklines. At the end of the geometry processing this allows a high-resolution grid to be written and used for high resolution map outputs and also DEM check grids (DEM_Z_HR). All grid formats in Table 11.3 are supported. Currently, high resolution raster outputs are only available for depth (d) and water level (h) output data types.

To produce high-resolution output, add “HR” to the grid format. For example “HRTIF” or “HRFLT” to produce a high resolution TIF or FLT respectively. Note, the Map Output Interval will also have to be set, for example, “HRTIF Map Output Interval == 0” will output just the maximums using the HRTIF format.

The high resolution grid output uses the following approach:

  • The water level at each HR raster output cell is interpolated from the computed 2D water levels.
  • The depth is the difference between the interpolated water level and the terrain elevation interpolated from the surrounding SGS sampled elevations.

The regular grid output interpolates depth from cell centres/corners with a default output resolution of half the cell size. The difference between the regular and the HR grid output is illustrated in Figure 11.1.

**Regular Grid Depth Output (Left) and HR Grid Depth Output (Right)**

Figure 11.1: Regular Grid Depth Output (Left) and HR Grid Depth Output (Right)

When modelling breaklines in TUFLOW, “thin” breaklines modify the cell face elevations but do not modify the cell storages. When outputting the high-resolution outputs, the user can set whether the cell face elevations are included using the .tcf command (the default is ON, to use face elevations):

HR Grid Output Use Face Elevations == {ON} | OFF

The water level and depth interpolation approaches for the high resolution outputs can be altered depending on the type of model (e.g. hydrology inflows or direct rainfall) and specifically how TUFLOW outputs the high resolution results around thin breaklines. These output options are discussed on the HR Output TUFLOW Wiki page.

HR grid outputs are also compatible with the output zone functionality, see Section 11.2.5.

11.2.2.4 Combined Format

Combination of a mesh and grid-based approach. The only available combined format is the WRC format. The WRC format produces a master (.wrc) file and one or more WRR and WRB files. The approach adopted by TUFLOW is as follows:

  • Each 2D domain is output as a rotated WRR format grid with the cell size equal to the TUFLOW cell size. The WRR format is significantly faster than than the WRB format due to its grid-based formatting. The output values are the cell centre values for each 2D cell.
  • Any 1D WLL triangulations are output as a separate WRB file.
  • The WRC “master” file is output to the specified results folder, while the WRR and WRB files are written to a “waterRIDE” sub folder.

11.2.2.5 GIS Based Format

GIS based map outputs use the format specified by GIS Format to write out map output as a series of GIS layers. Gridded output format as GIS layers can be written directly from TUFLOW during the simulation by including “GIS” in the Map Output Format command. This offers a similar functionality to that using the TUFLOW_to_GIS utility via, for example, the –shp option. For scalar outputs, these are output as a point GIS layer with a separate file for each output time. For vector outputs, this can be either as a point or region GIS layer. Specific commands to this output format are:

The commands above can be applied to all vector outputs or can be specific to the data type by prefixing with a “v”, “q” or “W”, for velocity, unit flow or wind respectively. The example below sets the scale factor to 1 for all outputs except unit flow, which has a smaller factor of 0.1.

GIS Grid Vector SF == 1.0 ! Set scale factor for all GIS vector output to 1.0
q GIS Grid Vector SF == 0.1 ! Set scale factor for unit flow GIS output to 0.1

11.2.3 Map Output Data Types

TUFLOW can output a wide range of output types in map format. Table 11.4 describes all the non-hazard map output types, while Table 11.5 contains all flood hazard category (Z) map output types. The map output types produced by a simulation are controlled using the .tcf command Map Output Data Types.

The map output types’ flags are listed in the first column of the tables and are used to denote the type(s) to be output. They can occur in any combination or order and are not case-sensitive. For example, to output water level, velocity and unit flow, enter the following line in the .tcf file:

Map Output Data Types == h V q

Although optional, it is strongly recommended that spaces are used between each data type for clarity.

The output types are available in a wide range of Map Output Formats and can be varied for different formats (Section 11.2.2) and between output zones (Section 11.2.5). Not all Map Output Data Types are available for all Map Output Formats due to limitations or constraints of the type/format. The supported formats for each type are documented in Table 11.4 and Table 11.5.

It is possible to get different output types for different output formats, as discussed in Section 11.2.2. For example:

TIF Map Output Data Types == h V Z0
XMDF Map Output Data Types == h V q d ci z0

11.2.3.1 Hazard Data Types

Table 11.5 presents the hazard category outputs. Of note is that each hazard is tracked every timestep for its maximum if Maximums and Minimums is set to “ON” or “ON MAXIMUMS ONLY” (the default) to ensure that the peak hazard category is recorded during the simulation. Up to ten (10) different hazard categories per simulation can be specified for map output. Grid map output hazard categories are output as integer grids (i.e. values are rounded to the nearest integer when a grid output cell centre is located at a change in category) when using Map Output Data Types except for output Z0, ZUK0 and ZUK2, which are output as real numbers.

Users have the option to customise hazard outputs based on a .csv file input using the “ZUD1” hazard type. This feature allows users to define a number of depth, velocity and depth-velocity product thresholds to create custom hazard categories. Like other hazards these are assessed at each computational timestep and the maximum is tracked every timestep. To specify a user defined hazard, add output type “ZUD1” in the “Map Output Data Types” command, e.g.:

Map Output Data Types == h d V ZUD1

The thresholds are defined through reading a .csv file using the following command:

Read Hazard File == <file_name>.csv

The csv file should contain three (3) columns defining the thresholds for depth, velocity, and depth-velocity product respectively. The example below produces the same output as for “ZNZ1” described in Table 11.5.

**Example Hazard File for User Defined Hazard**

Figure 11.2: Example Hazard File for User Defined Hazard

  • Category 1: D < 0.1. Note: the velocity, and depth-velocity product thresholds are effectively not applied by using a large threshold value such as 99999.
  • Category 2: D < 0.5 and V < 1.0. Note: the depth-velocity product threshold is effectively not applied by using the large threshold value of 99999.
  • Category 3: D < 1.0 and V < 2.0 and D*V < 1.0
  • Category 4: Otherwise
  • Dry points are assigned Category 0.

A maximum of 10 categories can be specified by the .csv file, but just one User Defined Hazard is currently allowed per simulation. This functionality is compatible with both TUFLOW Classic and HPC solvers.

Please email if you would like to code a unique hazard output that is currently not included as an output type option or possible using the user defined hazard type (ZUD1).

11.2.4 1D Map Outputs

1D domain results can be output in combination with 2D domain(s) by using the 1d_wll GIS layer and the Read GIS WLL 1D command. 1d_wll GIS layer(s) are used to define and control the 1D map output. The layer(s) contain lines (called Water Level Lines or WLLs) that cross or snap to 1D channels and/or nodes. A WLL is essentially a line of horizontal water level, and should be digitised on this basis (i.e. perpendicular to the flow direction).

The direction of WLLs is important. They must be digitised from left to right looking in the positive direction of the digitised channel.

When viewing the results, if the 1D WLLs and 2D domains overlap, the 1D results are displayed on top of the 2D results. However, depending on the viewing platform, when observing the scalar and vector magnitudes as the cursor is moved around, the 2D values maybe given precedence over the 1D where the overlap occurs.

Two WLL options are available. The preferred method can be specified by using the command WLL Approach. Of the two options, Method A and Method B, Method B allows more advanced and accurate mapping of 1D results in map formats and is the default and recommended method, and that described below. For documentation on Method A, refer to the TUFLOW 2010 Manual.

It should be noted that water level lines do NOT change the 1D hydraulic computations, they are purely used in order to display the 1D results in plan (2D) formats.

11.2.4.1 Water Level Lines

Ground elevations and optionally material (Manning’s n) values can be assigned to points along a WLL. A more accurate representation of 1D domain velocity and flood hazard can be mapped using this approach. The velocity at a point on the WLL is estimated by carrying out a parallel channel analysis along the WLL using the flow in the channel the WLL is associated with as explained further below.

The parallel channel analysis assumes the water surface slope is that of the associated channel, and the water level at the WLL is linearly interpolated between the upstream and downstream nodes based on the digitised channel length.

WLLs can have any number of vertices. The association of WLLs to channels and nodes is as follows:

  • At the channel ends, to lock in the water level at a node, use a 3 vertex line with the middle vertex snapped to the 1D node. If you use a 3 vertex line across a channel, the channel “thalweg” is taken at the middle vertex, otherwise, for 2 vertex lines the mid-point of the WLL line is used.
  • If a WLL crosses two or more channels, the channel closest to the middle vertex (3 point line) or half-way point (2 point line) is used.
  • If a WLL middle vertex snaps to a node with two or more channels on the upstream side, the channel that is closest in angle to the WLL’s perpendicular (based on the WLL’s two end points) is used.
  • For 4 or more vertices, one of the middle vertices (i.e. not an end vertex) must snap to a vertex on the channel line.

There is one 1d_wll attribute required as described in 11.6. The attribute, Dist_for_Add_Points, is the minimum distance in metres along which to generate points for that WLL. If Dist_for_Add_Points is zero, only the vertices along the WLL are generated.

Estimates of the elevation, depth, velocity and other hydraulic outputs from the parallel channel analysis are produced at each WLL point including any automatically added ones. The user can subsequently assign elevations and materials to each point to improve the topographic representation if so desired (see Section 11.2.4.2).

An automatically generated triangulation (TIN) of the WLL points is used to display the outputs in map based formats.

Use the 1D command Read GIS WLL to specify the 1d_wll layer and automatically create 1D map output for TUFLOW 1D domains. For Flood Modeller or XPSWMM linked TUFLOW models, use Read GIS ISIS WLL or Read GIS XP WLL in the .tcf file to read the 1d_wll layer(s). The Flood Modeller units or XPSWMM links will also need to be built into a GIS layer and read using Read GIS ISIS Network or Read GIS XP Network from the .tcf file.

Several 1d_wll layers can be specified covering different sections of the 1D domain(s) if required. The TIN of triangular elements is created from the WLL points and included in the .2dm mesh file that can be viewed using the QGIS TUFLOW Viewer or in SMS to check they have been created correctly. In particular, look for overlapping or strange triangulations that arise from WLLs not being digitised correctly.

The default approach uses the processed cross-section data (height–width data) from the channel for setting elevations at each WLL point. For Flood Modeller and XPSWMM, these data are automatically fed through to TUFLOW and is based on the cross-section information entered into the Flood Modeller/XPSWMM model.

If a WLL is snapped to a node, the processed data used for setting any bed elevations is from the higher channel unless it is a bridge in which case it uses the bridge processed data.

WLL Automatic == CULVERTS is particularly useful for automatically generating WLLs for pipe network systems. WLL No Weirs is useful for not assigning WLLs to weir channels where they are in parallel to another channel (this is particularly useful for BW, CW and RW channels).

11.2.4.2 Water Level Line Points

If Write Check Files in the .tcf file is specified, two GIS check layers are created from the WLLs. These are labelled 1d_WLLo and 1d_WLLp. 1d_WLLo (Water Level Line Objects) reproduces the WLLs with attributes containing the channel and node the WLL is allocated to for cross-checking purposes.

1d_WLLp layers (Water Level Line Points) contain all of the elevation points generated based on the Dist_for_Add_Points attribute. This layer can then be used to allocate elevations (first attribute) to each point from a DEM (in the same manner that 2D Zpts can be assigned elevations).

A second attribute, RR, contains the relative resistance of each point (which will have a value of 1 when first generated). The RR attribute can be replaced by the integer material value at each point by using GIS to assign values from material polygons. The material value must exist in the .tmf file or a materials.csv file (see Read Materials File).

The attributes of a 1d_WLLp layer created by Write Check Files and used in Read GIS WLL Points, Read GIS ISIS WLL Points or Read GIS XP WLL Points are listed in Table 11.7.

The parallel channel analysis carried out for each WLL in future simulations will then use the WLL point data to estimate velocities and other outputs across the WLL instead of using the default approach of using processed hydraulic data as previously described. Note, the WLLs do NOT change the 1D hydraulic computations, they are purely used in order to display the 1D results in plan (2D) formats.

For Flood Modeller and XP-SWMM, the layers are essentially the same, but are named using xWLLo and xWLLp.

Note: If using Read GIS WLL Points or Read GIS X1D WLL Points, this layer must be a copy of the 1d_WLLp layer produced by Write Check Files. Points from this layer can be deleted, but not added. At deleted points, the default of estimating an elevation from the channel’s processed data is used. If the 1d_WLL layer is modified or any of the Dist_for_Add_Points attribute values changed, the 1d_WLLp layer needs to be regenerated and repopulated again.

For culvert channels (R and C channel type), only the end and mid vertices are used along the WLL, and the elevations are set to the culvert invert irrespective of the number of points along the WLL or the Dist_for_Add_Points value.

Elevation values along WLLs for bridge channels are always based on the processed data (i.e. any WLLp Z values are overridden) to ensure that the bridge deck underside is correctly represented. This has benefits when using the post-processing utility TUFLOW_to_GIS.exe (see Section 17.3) when extracting obverts of structures for longitudinal profiles.

A useful tip at a junction of 1D channels is to use a connector for any side channels (Type = “X” – see Section 5.9.3). Separate WLLs can then be allocated to the side channel and main channel removing the confusion that sometimes occurs in generating the triangulation between WLLs at junctions.

11.2.4.3 Adding Triangles into the 1d_WLL Layer

When using Method B, triangular regions can also be included in the 1d_WLL layer (as a separate region or polygon layer if using the .gpkg or .shp formats). The triangles are inserted into the WLL triangulation. This is particularly useful at junctions or between parallel channels to enhance the WLL TIN by facilitating a smooth transition of the water level surface at, for example, junctions or across meanders to infill areas and avoid steps in the output surfaces.

The triangles must snap to the ends of WLL line objects. In the example below, the WLLs are shown in red and the triangles are shown as yellow. These are connected (snapped) to the ends of the WLLs. The map output surfaces will interpolate over the triangle based on the hydraulic output values at the snapped WLLs.

If the region object is not correctly snapped an ERROR 1311 message occurs, pointing to the vertex on the triangle that is not snapped.

**Adding Triangles into 1d_WLL Layer to Infill Areas**

Figure 11.3: Adding Triangles into 1d_WLL Layer to Infill Areas

11.2.5 Output Zones

Output Zones are a powerful feature that allows users to generate map and check file outputs for sub-regions of a model. The region is defined using a polygon feature using the 2d_oz_empty template file. No attributes are associated with the polygon. One polygon per output zone and GIS layer is recommended. Note, multiple polygons within a single layer are not supported.

Different Output Zones can have different output formats, start/end times, frequency of output, and output data types. Any number of Output Zones can be defined for a model, and all, a selection of, or none of the Output Zones can be activated when the model is run.

Output zones are very useful, particularly for large models, where higher frequency map output is required for a portion of a model, for example to create an animationfor an urban region using a shorter map output interval. Another useful purpose is to create an Output Zone for a problematic section of model that requires closer examination, by generating output on a fine time interval during the period in question. Another benefit is simply reducing the size of the output files by only outputting where information is required, and disabling the output for the whole model using Map Output Entire Model == OFF. Output zones currently only apply to TUFLOW’s map and GIS check file outputs (i.e. does not apply to time series and other plot outputs).

Note: the Model Output Zones command is used to select which Output Zones are to be used (activated). If this command is omitted, no output from the Output Zones is written. Separate multiple Output Zones using a “|” (vertical bar or pipe character). For example, to output from zones ZoneA and ZoneC specify:

Model Output Zones == ZoneA | ZoneC

If there are numerous Output Zones, it is recommended that the definitions are placed in one or more separate files, and use the Read File command to reference these file(s). This will keep the size of the .tcf file to a minimum. This, however, is not a requirement.

Each Output Zone is defined using a definition block as follows:

Define Output Zone == <ZoneA>

…..

End Define

The following commands can be used within an Output Zone definition. With the exception of Read GIS Output Zone, all commands are optional. The optional commands can be used to change the default setting or the setting applied outside the Output Zone definitions.

Read GIS Output Zone
Map Output Format
Start Map Output
End Map Output
Map Output Interval
Map Output Data Types
Output Folder
Maximums and Minimums
Maximums and Minimums Only for Grid
Write Check Files

The Read GIS Output Zone reads a GIS layer containing one or more polygons that define the regions to be output. The attributes of the layer are not used.

Note: If one of the commands above is not specified within the Output Zone’s definition, the latest setting of that command prior to the Output Zone’s definition, or the global default if the command has not been used prior to that location in the .tcf file, will be used. If, for example, all output is to be in the .xmdf format, only specify Map Output Format == XMDF once and prior to any Output Zone definitions.

Map Output Entire Model can be used outside an Output Zone definition block to turn on or off map output for the entire model (the default is ON). If set to OFF only map output for any Output Zones is written. Map output commands that occur outside Output Zone definitions apply to the entire model output.

Example: Defining an Output Zone

The example below defines Output Zone “ZoneA”. In the example, a DEM_Z check grid will be the only check layer written for the Output Zone, and Grid, WRB and XMDF outputs will be produced. The Grid output will consist of water level (h) and VxD (Z0) results, starting at time 0.5 hours and ending at 4 hours at an interval of 0.5 hours. The WRB and XMDF output files will contain water levels (h), velocities (V) and the ZMBRC hazard categories starting at the simulation beginning and finishing at 6 hours at an interval of 6 minutes.

! Entire Model settings
Map Output Format == TIF XMDF
Map Output Interval == 3600 ! Default output interval is hourly
TIF Map Output Interval == 0 ! Maximums only for .tif output
Map Output Data Types == h v q d z0 ! Default output types

Model Output Zones == ZoneA | ZoneB ! Output both of the zones below

Define Output Zone == ZoneA

  Read GIS Output Zone == ..\2d_oz_ZoneA.shp
  Write Check Files Include == DEM_Z
  Map Output Formats == TIF XMDF WRB
  End Map Output == 6.0
  Map Output Interval == 360
  TIF Start Map Output == 0.5
  TIF End Map Output == 4.0
  TIF Map Output Interval == 1800
  XMDF Map Output Data Types == h V ZMBRC
  TIF Map Output Data Types == h Z0

End Define

Define Output Zone == ZoneB

  Read GIS Output Zone == ..\2d_oz_ZoneB.shp
  Map Output Formats == XMDF
  Map Output Interval == 360
  XMDF Map Output Data Types == h V ZMBRC

End Define

11.2.6 Gauge Level Output

The 2d_glo GIS layer uses the .tcf command Read GIS GLO and writes map-based output data when the water level at the gauge reaches user defined levels.

Note: Map output based on reaching gauge levels replaces the conventional approach of using a Start Map Output time and a Map Output Interval.

The 2d_glo GIS layer is used to define the location of the gauge within the modelled extent. The gauge is digitised as a point object within a 2d_glo layer with the attributes as described in Table 11.8 and referenced within the .tcf using the command Read GIS GLO.

For TUFLOW Classic only, the water level at the gauge location is tracked by TUFLOW throughout the simulation and output to the Console DOS Window (see Section 14.2) preceded by the letters “GL”.

When the water level reaches a user specified value, the map output results at that simulation time will be written. The results are therefore useful for mapping the predicted area of inundation for specified gauge heights. The range of water levels at the gauge for which the results are written may be defined using one of two methods:

  • By referencing a .csv file within the 2d_glo GIS layer using the first attribute. The .csv file contains a single column of levels, and comment lines are allowed using the “!” or “#” character. If a .csv file is specified in the first attribute, the remaining attributes are not used, but dummy or empty values for these attributes must exist.
  • By populating the remaining attributes of the 2d_glo GIS layer as described below.

When using the attributes of the 2d_glo GIS layer to define the gauge heights, the map output results are first written when the water level at the gauge reaches the specified “Bottom_elevation”. Subsequent results are written as the water level at the gauge rises based on the value of the “increment” attribute. The map output ceases once the water level at the gauge reaches the “Top_elevation”.

For example, if the 2d_glo GIS layer has been defined as shown below, map output results will be written when the water level at the gauge reaches 41m, 41.5m, 42m, 42.5m and 43m.

Only one gauge location may be specified per model simulation. If more than one object exists within the 2d_glo layer, the gauge that is monitored by TUFLOW will be the last digitised point object. Similarly, if the Read GIS GLO command is used more than once, only the last occurrence of the command will be used.

11.3 Time-Series Outputs

Time-series results (sometimes referred to as Plot Output) produces output for graphing in charts and profiles. The data output are specified using the Map Output Data Types command, at an interval specified by the Time Series Output Interval command.

Time-Series data can be output for 1D domains (Section 11.3.1) and 2D domains (Section 11.3.2). If the total flow across the floodplain is required for a 1D/2D model where the river is in 1D and the floodplain in 2D, flows can be combined using Reporting Locations (Section 11.3.3). The Structure Reporting feature outputs time-series and summary data for structures (Section 11.3.4).

There are two formats available for time-series outputs: comma-separated values (.csv) (the default) and NetCDF (.nc). One of the advantages of NetCDF is all the timeseries output is in a single compressed .nc file, rather than multiple uncompressed .csv files. To change the format use Time Series Output Format command.

It is possible to write 1D and 2D time-series outputs as the simulation progresses, using the Write PO Online command.

11.3.1 1D Time-Series Output

Time-series data output from 1D domains is available for a range of hydraulic parameters. The Output Data Types ECF command or the 1D Output Data Types TCF command controls the types to output. The options are:

  • A: flow area at channels (m2);
  • E: energy at nodes (m);
  • H: water level at nodes (m);
  • Q: flow at channels (m3/s);
  • QI: integral flow at channels (m3);
  • S: structure and grouped structure output (see Section 11.3.4);
  • V: velocity at channels (m/s); and
  • Vol: volume at nodes (m3).

To review the 1D time-series data see 15.3.

11.3.2 2D Time-Series Output

Time-series data output from 2D domains is available for a range of hydraulic parameters (as listed in Table 11.9). Output takes the form of time-series hydrographs (referred to as PO – Plot Output) or longitudinal profiles (LP) over time.

This is carried out by creating one or more GIS layers containing points, lines and regions that define the locations of PO and LP output. Figure 11.4 illustrates how 2d_po objects are interpreted.

The start time for PO and LP output and the output interval is set using Start Time Series Output and Time Series Output Interval. If no start time is specified the simulation start time is used. If no output interval is specified the simulation will stop with ERROR 0046 to prevent excessive amounts of memory and disk space from being used.

The output is written to a .csv file and also to the _TS layer (refer to Section 15.3.4). 2D domain time-series (PO) output is synchronised with 1D domain output by default. This allows both 1D and 2D time-series to be placed in the _TS layer. Set Output Times Same as 2D to OFF in the .ecf file if 1D and 2D time-series data is not to be synchronised. In this case, no 2D PO is written to the _TS layer.

Maximums and minimums are output to four additional rows near the top of the _PO.csv file, and columns in the _LP.csv files, containing the Maximum, Time of Maximum, Minimum, and Time of Minimum values. The _TS GIS layer also contains the tracked values. For TUFLOW Classic this information is tracked every computational timestep. For TUFLOW HPC the maximum/minimum values are post-processed at the end of the simulation based on the Time Series Output Interval, not every computational timestep. Tracking of maximums and minimums can be disabled by setting the Maximums and Minimums Time Series command to OFF in the .tcf file.

**Interpretation of PO Objects and Map Output**

Figure 11.4: Interpretation of PO Objects and Map Output

11.3.2.1 Plot Output

The data types available for PO are as listed in Table 11.9, noting that only supported object types (i.e. Point, Line and Region) for the data type are documented in the table. PO data is read into a simulation using the Read GIS PO .tcf command. This is carried out by creating one or more GIS layers containing points, lines and regions that define the locations of PO and LP output. Figure 11.4 illustrates how 2d_po objects are interpreted.

Table 11.10 describes the GIS attributes of the 2d_po layer.

11.3.2.2 Long Profile Output

H_ (water level) and V_ (velocity) are the only data types available for long profile (LP) outputs. LP locations and data types are initiated for a simulation using the Read GIS LP .tcf command.

Table 11.11 describes the GIS attributes of the 2d_lp layer. The 2d_lp layer(s) contain lines defining where the profile data are to be generated. Each line is given a label to uniquely define the profile in the output. The starting vertex of the line will set the distance origin for the profile.

The advantage of having TUFLOW generate the profiles directly rather than post-processing them is the outputs will be slightly more accurate due to no post-processing interpolation rounding, plus if the profile(s) are repeatedly being plotted, the plotting process can be streamlined via python scripts that use the LP outputs.

11.3.3 Reporting Locations

Read GIS Reporting Location allows for plotting of time-series results that automatically combines 1D and 2D outputs. For example, it is possible to digitise a reporting location line that extends across 1D and 2D domains, including multiple 2D domains, and TUFLOW will sum the flow across any 1D channels intersected by the line and all the 2D cells. This will save the need to post-process time-series output from 1D and 2D domains to accumulate the flow.

Reporting location lines are digitised into a 0d_RL GIS layer containing only a single attribute, the name of the reporting location, as outlined in Table 11.12. Points, lines and regions can be used. Points will be treated as water level output, lines as flow output and regions as volume output. For a point snapped to a 1D node, the 1D water level is used, if no 1D node is snapped a 2D water level is output.

The flow line can cross 1D and 2D sections of the model, for the 1D channels it does not have to snap to any vertices on the channels, it just needs to intersect them.

The RL outputs are written to the “plot\csv\” folder. The following files are produced:

  • _RLL_Q.csv - flow time-series;
  • _RLL_Qmx.csv - maximum flow information;
  • _RLP_H.csv – water level time series;
  • _RLP_Hmx.csv – maximum water level information;
  • _RLR_Vol.csv - volume time-series; and
  • _RLR_Volmx.csv - maximum volume information.

As well as the maximum water level and flow information, the time that these occur, the water level at maximum flow and vice versa, and the maximum change between timesteps are also output to the mx.csv files.

The RLs are also output to the plot\gis PLOT GIS layers and can be viewed and their time-series data displayed using the TUFLOW Viewer using the QGIS TUFLOW Plugin as illustrated in Figure 11.5.

An example model using reporting locations is available in the Example Models on the TUFLOW Wiki. In addition, see Section 15.4.3 for information on plotting of Reporting Location results.

**Example of the QGIS TUFLOW Plugin for a Reporting Location**

Figure 11.5: Example of the QGIS TUFLOW Plugin for a Reporting Location

11.3.4 Structure Output

The Structure Reporting feature outputs time-series and summary data for single and grouped structures. 1D and 2D structures are all output together to give a complete set of results. The summary output at the flood peak also produces the flow split between below and above deck, along with other information such as the head drop. The structure output is particularly useful in the reporting of hydraulic structure flows and afflux.

Structures are classified according to the following logic:

1D Structures:

  • 1D structures that are in parallel (i.e. two structures that link to the same upstream and downstream nodes) are automatically grouped together and treated as a single structure for this output. The ID assigned to the group is the structure with the lowest bed elevation. Note that the directions of the digitised channels is important, that is, to form a group all channels must be digitised in the same direction.
  • For 1D structures that have no parallel 1D structures, these are also included in the output so as to provide a complete set of results for all 1D structures.
  • The flow split between below and above deck is based on the structure geometry, except weirs contributing to the below deck flow and/or above deck flow, depending on the configuration.

2D and 1D/2D Structures:

  • To create a structure output that includes 2D flow, and optionally any 1D structures, the “QS” PO line is digitised in a 2d_po layer (see Section 11.3.2.1). All 2D flow across this line and any 1D structures that intersect this line are grouped together. The 1D structure’s 1d_nwk line does not have to snap with the QS line; they only have to cross over each other. The ID assigned to the structure output group is the 2d_po QS Label (see Table 11.10).
  • If the 2d_po QS line selects cell sides that are modified by the Layered Flow Constriction or 2D Bridge feature, the summary output will split the flow into a below and above “deck” component based on Layers 1 to 2 being below “deck” and Layer 3 and 4 above.
  • 2d_po HU and HD lines or points can be used to define the upstream and downstream water levels of the structure. HU objects should be located upstream of the structure and HD downstream. The average 2D water level along a line object will be used to populate the upstream and downstream water level data in the output. Lines can have more than two vertices (i.e. polylines are accepted). To associate the HU and HD objects with the QS line, all three (QS, HU and HD) must have the same ID for the 2d_po Label attribute Label (see Table 11.10). If a QS line has no HU and/or HD objects associated with it, output that cannot be produced, such as the water level drop across the structure, is given a -99999 value in the _SHmx.csv output file described below. HU and HD inputs are necessary for 2d_lfcsh and 2d_bg bridges.

An example model of the structure output for a bridge represented by a 2d_lfcsh is available in the Example Models on the TUFLOW Wiki.

The structure output is located in the plot/csv folder and includes:

  • _SQ.csv file that contains time-series data of the flow through the structure. This file is similar to other time-series .csv output, but it only contains 1D and/or 2D structures as described above.
  • _SHmx.csv file that contains a summary of each structure when the upstream water level reached its maximum. To generate this output the flow and other information is tracked every timestep for grouped structures. The output columns include: flow, area and average velocity for below and above deck; total flow, area and average velocity for the whole structure; upstream and downstream water levels; the head drop across the structure (i.e. upstream minus downstream water level); and the time these data were recorded (e.g. the time the upstream water level peaked).
  • Time-series output for the upstream and downstream water levels are available through the 1d_H.csv, 2d_HD.csv and 2d_HU.csv files. Note that for the 2d_HD.csv and 2d_HU.csv files, if all 2D cells are dry a -99999 is output.

See Section 15.4.3 for further information on plotting of grouped structure results.

Two structure group check files are output if a model contains any structure groups (either automatically created, or via a “QS” type line in a 2d_po layer). The check files are both .csv files as follows:

  • <simulation name>_Str_Grp_All.csv, contains information for all structure groups, including single 1D structures.
  • <simulation name>_Str_Grp_Multi.csv, contains information for structure groups that are comprised of more than one 1D channel or are generated from a 2d_po “QS” line.

For more information on the structure group check files, please see the Check File Page on the TUFLOW Wiki.

11.4 Specialised Outputs

Specialised outputs include:

  • Receptors: records the flood level and time at one or more gauge(s) when the receptor is inundated above the specified trigger level, see Section 11.4.1.
  • Evacuation Routes: provides additional information (e.g. first point of closure) along routes, see Section 11.4.2.

11.4.1 Receptors

Read GIS Objects RECORD GAUGE DATA records the flood level and simulation time at one or more gauge(s) when receptors are first inundated above their trigger inundation levels (e.g. floor levels). Read GIS Receptors can be used as an alias to Read GIS Objects. Level output associates a flood level at one or more gauges with the time of first inundation at properties, buildings, or other areas of interest within the modelled extent.

When the water level at the property reaches a user-defined trigger inundation level, the gauge height and simulation time are recorded and tagged to the receptor. This is particularly useful for flood warning and forecasting studies where property specific information on the likelihood of a property being inundated for a given gauge height can be generated. An example of translating gauge data information to catchment receptors can be found in the TUFLOW Wiki Example Models.

Gauges are defined as a point within a 2d_po GIS layer with type “G_” (see Section 11.3.2 and Table 11.9). The levels from all gauges are recorded at each receptor once inundated.

Receptors must be GIS point and/or polygon objects located in one or more GIS layers nominally prefixed by 2d_obj or 2d_rec. Each object within the layer represents a receptor, property or other object of interest. For information on the attributes of the GIS layer see the Read GIS Objects .tgc command. The command also includes options to set the Zpt elevations to the receptor level or the first attribute in the layer (for example, to set the Zpts to the floor level of the buildings), or to alternatively use the existing ZC elevations.

Once simulated, a GIS layer is written to the location as specified using the .tcf command Output Folder, and has a _GDO extension standing for Gauge Data Output. The layer contains point objects (for regions the centroid is used). The attribute data for the layer are described in Table 11.13. Figure 11.6 shows an example of how the output GIS layers may be used to illustrate the flooding of properties in relation to the water level at a gauge.

As well as outputting the water levels at gauge objects (as described above), the water levels at all point reporting locations and flows for all line reporting locations are also recorded. The reporting locations can be 1D and/or 2D locations. See Section 11.3.3 for a full description of the reporting location functionality. For example, flows past a gauge can be recorded, including combined flows from 1D channels and the 2D domain(s).

**Example Use of Gauge Data Output Layer**

Figure 11.6: Example Use of Gauge Data Output Layer

11.4.2 Evacuation Routes

Evacuation routes can be specified to output information relating to:

  • Evacuation route suitability;
  • Warning times;
  • Risks;
  • Degree of route inundation; and
  • Duration of inundation.

The .tgc Read GIS Z Shape Route is used to define the routes and, by default, also adjust the Zpt elevations along the route using the standard Z Shape options. The 2d_zshr layer is the same as a 2d_zsh layer, but with three additional attributes, as shown in Table 11.14. Note, if using the TUFLOW HPC SGS functionality (Section 7.4.3, the cells/faces selected by this command are assumed flat (e.g. the traditional sampling approach using a single cell centre/face elevation).

The commands Set Route Cut Off Values and Set Route Cut Off Type can be used in the .tcf and/or .tgc files. If used in the .tcf file, this sets the default values for the 2d_zshr Cut_Off_Type and Cut_Off_Values attributes if these attributes are left blank. The default values can be changed between different Read GIS Z Shape Route commands in the .tgc file by repeat usage of the same commands.

The _RCP output layer is a layer of points showing where a route’s cut off value(s) were first exceeded (e.g. first point of closure). The layer contains the attributes as shown in Table 11.15.

The _RC.xmdf output file contains the Route Categories over time and is written when adding RC to the .tcf command Map Output Data Types. This file can be used to view and animate the route category values.

A range of evacuation route examples are available in the Example Models on the TUFLOW Wiki.

11.5 Check and Log Files

The options to control check and log file outputs are discussed in Chapter 14, with common commands to control these outputs listed in Table 11.16.

11.6 Output Control Commands

A wide range and variety of commands allow the user to configure TUFLOW output to be different to the default settings. Table 11.16 lists these commands in different categories followed by a brief description.

References

Australian Emergency Management Institute. (2014). Australian Emergency Management Handbook 7: Managing Floodplain Best Practice in Flood Risk Management in Australia. https://knowledge.aidr.org.au/media/3521/adr-handbook-7.pdf
Cox, R., Shard, T., & Blacka, M. (2010). Australian Rainfall and Runoff Revision Project 10: Appropriate Safety Criteria for People. Austalian Institute of Engineers. https://arr.ga.gov.au/__data/assets/pdf_file/0006/40578/ARR_Project_10_Stage1_report_Final.pdf
CSIRO. (2000). Floodplain Management in Australia Best Practice Principles and Guidelines. CSIRO Publishing.
Environment Agency. (2006). UK FD2321 Technical Report - Flood Risks to People: Phase 2 Project Record. UK Government; https://assets.publishing.service.gov.uk/media/602bbb768fa8f50386a7f8aa/Flood_risks_to_people_-_Phase_2_Project_Record.pdf.
Environment Agency. (2008). Explanatory Note for FD2320 and FD2321 Project Record. https://assets.publishing.service.gov.uk/media/602bbdcfe90e070561b31432/Explanatory_note_for_FD2320_and_FD2321_project_record.pdf.
GHD. (2011). MBRC Storm Tide Management Study: Stage 1 – Scoping Report.
Melbourne Water. (2016). Flood Mapping Projects Guidelines and Technical Specifications (Version 1).
NSW Goverment. (2005b). Floodplain Management Manual. NSW Department of Land; Water Conservation (NSW Government).
Port Macquarie-Hastings Council. (2018). PORT MACQUARIE-HASTINGS COUNCIL FLOOD POLICY. https://www.pmhc.nsw.gov.au/files/assets/public/v/1/document-files/your-council/publications/policies/port-macquarie-hastings-council-flood-adopted-2018-12-12.pdf.
Queensland Reconstruction Authority. (2012). Planning for Stronger, More Resilient Fl Oodplains: Part 2. https://www.qra.qld.gov.au/sites/default/files/2018-10/resilient-floodplains-part2-full_0.pdf.