Geographic Wizard for catchments

Geographic Wizard

The Geographic Wizard is a structured sequence of windows that guides you through the process of creating a catchments scenario. On completion of the wizard, you will be presented with a node-link network. Figure 1 shows the sequence of steps in the wizard.

Figure 1. Catchments scenario setup process

If some data or parameter values are unavailable whilst creating the scenario, skip the step by clicking Next. Once you have created a scenario, you can make changes using the choices on the Edit menu.

In some situations, the wizard will skip over several steps if certain values have not been specified. For example, if you do not define any constituents (step 4), you cannot assign inputs or parameterise any constituent-related steps. Therefore, steps 8-13 will appear greyed out in the wizard progress indicator (Figure 2). This figure also shows the action buttons that are visible at the bottom of every screen in the wizard.

Figure 2. Network definition via a Digital Elevation Model

On completion of a step, you can move to the next step by clicking Next and backwards by selecting Back. Note that you cannot go back to the network definition step once you have completed it and moved to the next step. Cancel allows you to quit the wizard at any time. Note that you will lose all the data entered so far. You can also complete the scenario if the Finish button is active (from step 7 onwards). Source will create a scenario when all mandatory steps have been completed, even if all the steps in the wizard have not been completed.

Note: Most of the windows appearing in the wizard are identical to the related Edit menu choices. For steps that are related to rainfall runoff, constituent generation and filter models and have screen output similar to the wizard, refer to the Constituents and Rainfall runoff models chapters.

Welcome screen (Step 1)

The Welcome screen gives some information about the steps within the wizard. These include:

  • Specifying the catchment network;

  • Creating a list of the constituents and functional units that will be used in the scenario;

  • Assigning functional unit areas and models to rainfall runoff, constituent generation, filters, node and links;

  • Selecting input files for the rainfall runoff models; and

  • Setting parameter values for models assigned in the node-link network.

Name the scenario (Step 2)

Enter a scenario name and an appropriate description, or accept the defaults.

Define the network (Step 3)

You must define the sub-catchments and stream network within the catchment, using either a catchment map or Digital Elevation Model (DEM).

The DEM based network generation method requires a DEM, and divides the entire node-link network into sub-catchments based on a user-specified measure of upstream area (sometimes called stream threshold), usually for first-order streams. You can define points on the stream network where confluences between sub-catchment will be formed. The minimum sub-catchment area can also be modified. For example, in a node-link network spanning 1000 km2, entering a stream threshold of 10 km2 will produce approximately 100 sub-catchments.

The minimum sub-catchment area defines the drainage area at which headwater catchments can be formed (also known as drainage threshold or stream threshold). As a result, any link in the generated network will have a total upstream catchment area of at least this amount. Individual sub-catchment sizes will vary and some will be significantly smaller. Notably, where two connected nodes are very close together, the link between them will be short and the corresponding catchment can be very small.

DEMs that have been derived only from contour maps or from other remote sensing technologies such as light detection and ranging (LIDAR) or shuttle terrain radar mapping (STRM) are usually not hydrologically correct. Such DEMs need to be adjusted to represent the flow of streams in the right direction. This is because they create a sense that there are spurious pits in the landscape when actually, none are present.

The Draw Network method allows you to define the networks manually, by loading a sub-catchment map (a raster with cell values set according to sub-catchment number) and then connecting sub-catchments by clicking and dragging using the mouse. This specifies how those sub-catchments are linked together by the node-link. You should use this method when you have previously defined the boundaries of your sub-catchments, either by using Source or an external program.

You can create a draft of your sub-catchment boundaries using the DEM based network generation method, export those boundaries from Source, edit them in an external GIS program (for example combining or splitting sub-catchments), and then import your edited sub-catchment boundaries back into Source to create a revised sub-catchment node-link network using the Draw Network method.

Note: Any spatial layers displayed in the Geographic Editor must have the same projection as the sub-catchment map or DEM that was used to create the node-link network.

To use the DEM-based Network Generation method:

  • Choose DEM based network generation from the Available Methods drop-down menu;
  • Click Load DEM;
  • Navigate to the catchment DEM and open it. Source generates a default network based on a minimum sub-catchment area of 50 km2. It is recommended that you specify the minimum sub-catchment area prior to loading a high resolution DEM to reduce computational time;
  • Specify a different minimum sub-catchment area in km2, if necessary, and click Recompute Streams. The drainage density depends on the minimum sub-catchment area; and
  • Select a desired catchment outlet (a point in the network where all streams in the desired area converge) for investigation by clicking inside the map. This generates a network area broken into sub-catchments (shaded blue), where two sub-catchments converge at a confluence.

Figure 2 shows an example of a sub-catchment network.

Once a DEM has been loaded, you can alter it by right-clicking and choosing the various options available in the contextual menu:

  • To add a new catchment outlet, choose Select Cell from the contextual menu, and select another point on the catchment map;
  • Pan allows you to move the map up, down, left or right;
  • To zoom in, right-click, and choose Zoom In. Drag the mouse to form a zoom box (top right to lower left). To zoom out, double-click anywhere on the map or click Undo Zoom;
  • To remove a catchment outlet point, choose Remove Cell;
  • Choose Properties to change the attributes of various elements, such as the colour or symbol of nodes; or
  • Save allows you to save the map as a .asc file.

Adding nodes

Ordinarily, nodes are placed at stream confluences identified in a DEM analysis. It is possible to supplement these nodes and place additional nodes at different locations, typically to identify some point of interest, such as a gauging station of a dam wall. This will also cause additional sub-catchments to be created.

Additional nodes are added from a file of geographic points (eg .MIF, .SHP, .tsd). The points are automatically added to the streams and become nodes and catchment boundaries in the network. There is the potential for this automated process to give erroneous results, so it is important to understand the process used to ‘snap’ to the stream.

When added, each additional point is compared to the stream map to identify the closest point on the stream. If that point is within a threshold distance of 1000m, a node is added on the stream. This process is necessary because in many cases, the recorded location of gauging stations (or other features of interest) doesn’t correspond to the location of the stream, as determined from a DEM.

The snapping process works well in many cases, but it can result in errors, particularly where there are two or more distinct stream reaches, all within the distance threshold. The system does not have a way to determine the correct location and instead chooses the closest. This issue typically arises where a gauging station is just upstream or downstream of a confluence (when two or three stream reaches can be within the threshold) and in cases where a small minimum catchment area has been selected relative to the areas drained by the gauge. In the latter case, the higher density of sub-catchments and links can lead to a point snapping to a small, lower order stream.

You can use a list of nodes to define the sub-catchments as well:

  • Set the stream threshold to a large value (so that only a few sub-catchments are defined); and
  • Load a list of nodes in one of the pre-defined formats.

You can set Source to recalculate networks manually by enabling Multiple Outlets. Then, select multiple points/cells on the DEM as required and click Generate Network.

Additional buttons under Multiple Outlets allow you to work with outlet nodes:

  • Add all outlets - detects all outlet edge cells with the lowest elevation and adds nodes at these points;
  • Add largest outlet - detects the largest outlet in a DEM and adds a node for the outlet;
  • Import outlets - allows you to import a file of outlet nodes. This is different from Load Nodes, which creates confluence nodes; and
  • Clear outlets - clears all specified outlet nodes.

Note that the total area covered by the catchment can be viewed in the Recording Manager once the scenario has been run (Figure 3).

Figure 3. Recording Manager, View Catchment Areal

To use the Draw Network method:

  • Choose Draw Network from the Available Methods drop-down menu;
  • Click Load Sub-catchment Map and open the desired sub-catchment raster file;
  • Click and drag on the map to create a stream network by specifying the direction of flow between sub-catchments. Ensure a connection is also made from the lower-most sub-catchment to a point outside the sub-catchment.This is the catchment outlet, highlighted with a circle in Figure 4.
  • Click Next once the network includes every sub-catchment and a corresponding outlet. You can rename the sub-catchments in the table under the Sub-catchment button.

If you need to delete a link, right-click on the map and choose Select Link/Node from the contextual menu. Click on the link you wish to delete; it will turn red. Press the Delete key on the keyboard or click Delete Selected elements on the right of the map. Ensure that you delete any unattached nodes that were connected to the deleted link. To re-draw links, right-click on the map and choose Draw Network. Continue drawing links as before.

Figure 4. Network definition by drawing

The other options in the list (Pan, Zoom, Copy graph and Properties) have the same function as that described for using the DEM based network generation method.

Additional buttons under Load Sub-catchment Map assist you in adding further detail to the network map. Load Gauge Map and Load Background Maps allow you to upload a shape file of gauges or an image like a stream flow map to assist in drawing the node-link network. Add Links from Shp file allows you to load a shape file of links that is then mapped to the sub-catchment map automatically.

If all the sub-catchments have not been connected, they will be shaded in blue. All sub-catchments must be connected by links before you can click Next (Figure 5).
Figure 5. Incomplete connection of sub-catchments

Define constituents (Step 4)

Enter a list of constituents that are of interest and which you wish to model. If you do not enter any constituents, the wizard does not display any steps that are constituent-related, ie. the constituent generation, filter assignment and filter parameterisation steps appear greyed out. Examples of constituents are total nitrogen (often abbreviated as TN), total suspended sediment (TSS), heat, dissolved oxygen, radioisotope tracers, etc.

Constituents cannot be deleted once they have been added and the Configure... dialog has been closed.

You can also add constituents after a scenario is set up using Edit » Constituents, which opens the Configure Constiuents dialog (Configure Constituents). Refer to Defining constituents for more detail. To add a constituent, type its abbreviation in the Name field and either click Add or press the carriage return. To remove a constituent, highlight it and click Delete.

Specify functional units (Step 5)

You must specify all possible functional units present in the area of interest, or that you want to represent in the scenario. There are three ways of defining FUs: manually, from a text file, or from a shape file (Figure 6).

Figure 6. Manually adding functional units

The default method of defining FUs is manually. To add a FU, type it in the Name field, and either click Add or press carriage return. To remove a FU, highlight it and click Delete.

You can import FU names from a text file by choosing Import from a text file from the Available Methods drop-down menu. The format of the text file is shown in Figure 7.

Figure 7. Adding functional units via a text file

You can also import FU names from a shape file:

  • Choose Specify from a Shapefile from the Available Methods drop-down menu.
  • Click Load to upload a shape file that has been configured with a field for FU types.
  • Select the layer from the list and click Create. A list of FU names is generated in the right hand pane. The shape file used to define FU types will also be used to assign the areas to each FU in each sub-catchment.
Note: Once a scenario is complete it is possible to add or remove FUs. However, the FU areas must be adjusted to incorporate this change. Additionally, rainfall runoff, constituent, filtering models and the corresponding input data must be re-assigned and parameterised.

Specify functional unit areas (Step 6)

You must assign areas to every functional unit in each sub-catchment. The functional units that appear depend on those you defined previously. You can assign areas manually (using either the Table or Map tabs), or from a raster such as a land use map. Alternatively, choose Edit » Functional Units » Assign Area... on completion of the wizard.

To assign FU areas manually using the Table tab:

  • Choose Manual Area Allocation from the Available Methods drop-down menu;
  • For each sub-catchment, enter the area that is allocated to each FU. You can specify this as an absolute value in the Area (ha) column, or as a percentage in the Area (%) column; and
  • You may enter the areas individually, or use the Apply-to options to fill multiple cells at once. This is explained in detail in the Using the Apply-to options section. You can also copy values from one cell and paste into other cells. Right-click on the source cell (the one you want to copy) and choose Copy. Then, right-click on the target cell, and choose Paste. The resulting table is shown in Figure 8.
Figure 8. Assign areas to functional units

When assigning FU areas manually using the Map tab, you can view the entire catchment and its outlines. FU areas can be assigned as follows:

  • Click on a sub-catchment that you want to assign a model to, or highlight several using the mouse (drag and select). Alternatively, use Ctrl to select several sub-catchments. If you hover the mouse over a selected sub-catchment you will see the its number or name;
  • To assign the area covered by each FU to the selected sub-catchments, select a cell in the table you want to assign the area to; and
  • Use the Apply-to options to assign the same area to the selected sub-catchment/FU combinations.

If the areal percentages per sub-catchment, do not sum to 100%, the percentage cell turns pale red (to indicate an error) for that sub-catchment. In Figure 9, SC # 1 (Urban FU), the area percentage should be 50, not 5. The error applies only to SC#1, so the cell for SC#2 is not pale red.

Figure 9. Example of incorrect area values (SC#1)

Figure 10 shows the correct areal percentages for SC#1.

Figure 10. Example of correct area values (SC#1)

You can assign areas to multiple FUs simultaneously, using the Apply-to options on the right of the screen. This is useful when you have a large number of sub-catchments, each containing several FUs.

  • In the Area (ha) or Area (%) column, click in any cell;
  • Enter a value or percentage;
  • Specify the appropriate values in the Apply selected cell to: options (to the right of the table); and
  • Click Apply, and the value will propagate through the specified sub-catchments and/or FUs.

To assign FU areas using a raster:

  • Click Load to upload a raster file;
  • If the land use map does not include a mapping of grid code to land use, click Assign Mapping;
  • In the resulting window, assign a FU to each grid code using the drop-down menu in the FU column (Figure 11); and
  • Click OK to close the Match FU Definitions to Raster window.

For each sub-catchment, the FU areas MUST add to 100% (+/- 0.1%). If areas do not sum to 100%, cells in the Area column will be shaded pale red. An exclamation mark will also appear on the left side of the first column.

Figure 11. Assigning FU areas using a land use raster

Rainfall runoff models (Steps 7-9)

The next 3 steps involve configuring the rainfall runoff models for every FU/sub-catchment combination. Refer to Assign rainfall runoff models for details on completing these steps. Click Next or Back to navigate between them.

On completion of step 7, the Finish tab at the bottom of the wizard becomes active, indicating that the information provided up to this point is the minimum data required to complete the wizard. Click Finish if you wish to complete the scenario.

Constituent models (Steps 10-11)

Here you configure the constituent generation models for each FU/sub-catchment//constituent combination. Configuring constituent models provides details on how to do this.

Filter models (Steps 12-13)

You now assign a filter model for each FU/sub-catchment//constituent combination. Filtering models represent for example, the effects of riparian filter strips, artificial wetlands, farm dams and similar management treatments. Fluxes from each FU can be passed through separate "filters". Refer to Configuring filter models for more details.

Designate a meaningful parameter set name to make it easier to select appropriate parameter sets when testing alternate scenarios.

Assign catchment models (Step 14).

This step allows you to assign models to an entire catchments. It generally applies to catchments that model, for example, groundwater (for the GWLag Plugin).

The default node type in Source is a confluence node, where the inputs to the node are the upstream sub-catchment outlets, and the output from the node is the current sub-catchment outlet. Similarly, straight through routing is the default link type in Source. To change the node model or link type, right click on the confluence node in the Geographic Editor, then choose Change Node Model or the link type respectively, from the contextual menu. For nodes, this lists all the model types available. Choose the required model.

Using the Apply-to options

This technique can be replicated for other Apply-to options available in the wizard. It can be used to:

  • Assign a single value to all FUs across all sub-catchments;
  • Assign a single value to all FUs in a particular sub-catchment (eg 50% of the total area is assigned to each FUs in sub-catchment #9);
  • Assign a single value to all FUs of a certain type (across all sub-catchments); or
  • Re-allocate FU areas by value or percentage. For example, if you have already specified that 50% of each sub-catchment is of FU type "Forest", and 50% is of type "Urban", you could model increases in urban encroachment by specifying Auto-increase by 5%, all FUs of type "Urban" at the expense of all FUs of type "Forest". See Figure 12.

Ensure that the percentage values for each sub-catchment sum to 100% prior to re-allocating FU areas.

Figure 12. Automatically re-allocating FU area by percentage

Editing scenarios

Once a scenario has been created it can be edited using the Edit menu.

Note: After a scenario has been created using the wizard, you cannot change the network, ie. you cannot merge or remove sub-catchcment.

The Edit menu gives you access to all the elements of a scenario that can be modified. Refer to the Constituents and Rainfall runoff models chapters for further details.

Note: If the FU list and areas are modified, the models, parameters and climate input data associated with them (ie. rainfall runoff, constituent generation and filter models) will be modified as well. Check that any of these models, parameters and input data are assigned correctly.