Tip of the Month

The ISIS pump unit enables modelling of an open-channel pump by deriving a flow-head relationship from one or more user-entered points on a pump curve. Control of the pump - in operating (ON), stopped or off modes - may be achieved by logical rules, therefore reacting to model conditions, or by preset times. This 'tip of the month' responds to many common questions about pump units, and provides advice on its implementation.

How much pump curve data do I need to enter?

You can enter one set of head-flow-efficiency values or a whole set of them. Remember: The data must be entered in ascending flow order. The optimal point on the curve (maximum efficiency) must also appear in the data entered.

Figure 1: Pump unit input form example

When ISIS runs, it automatically extrapolates using the entered data, to generate a full four-quadrant pump curve based on Suter's method¹. The ISIS model will always use the user-input data in preference to the extrapolated data, so enter as much data as you have available.

What checks can I make?

When a model with a pump unit is run, it always outputs the generated pump rating to the zzu file. It is recommended before running a full simulation to run the model in boundary mode - this will create the data in the zzu file. You can check the flow-head relationship, for example by copying the flow and head data to a spreadsheet and plotting it. Any kinks in the curve, - typically caused when the input data joins with the extrapolated data, may cause model instabilities. If this is the case, the kink can be eliminated by extending the user-input data (for example. by copying new values from the zzu export, but omit the kink).

Figure 2: Displaying generated pump data from the zzu file in a spreadsheet

Tips:

1. Sometimes the flow-head relationship has a tendency to jump to the wrong portion of the curve, especially when the pump starts up. To alleviate this, ensure that the "Steer starting pumps to user rating" option is switched on in the run form (Options tab).
2. Because the sudden switching of a pump on and off causes abrupt changes in flow, you may need to increase maxitr (maximum number of iterations, Advanced Parameters tab), for example to 13, to give the model more time to converge.

1. Fluid Transients. Wylie and Streeter (McGraw-Hill), 1978

New for ISIS  v3.5.1 - Gain simple access to free background GIS data from ISIS Mapper. In line with other GIS packages, ISIS Mapper now provides a means to add freely available maps to your viewport. This can be beneficial both during the model building process to ensure model features are correctly defined and located or in the model post processing stage for the generation of flood maps or animations.

What is a WMS? There are now numerous sources of free GIS data, which are accessible through Web Mapping Services. A Web Map Service (WMS) is a standard protocol for serving georeferenced map images over the Internet or Intranet that are generated by a map server using data from a GIS database. WMS is compatible with many commercial and open source GIS packages.

With the v3.5.1 release of ISIS, ISIS Mapper now includes the capability to add a WMS layer to your map view. The advantage of having this in ISIS Mapper is that you can now provide background maps when visualising ISIS models or viewing flood maps.

ISIS Mapper provides a facility to link through to a map server (as specified by you, through the definition of an appropriate URL). The layer is zoom-dependent so as your view resolution changes the background mapping will automatically update accordingly.

How do I Add a WMS Layer to my View?

The following provides step by step guidance on how to access and use the ISIS Mapper WMS functionality:

• Open ISIS Mapper and select Layer > New > WMS Layer from the main menu, as shown below:

• The create WMS server tool will then be displayed in a new window. If this is the first time you are accessing this function it will have no servers specified initially (as shown below). If a WMS server link has already been established it will be listed in the upper table (though only if it has been saved to the same folder as your 'ISIS Mapper.exe' file).

• The command buttons available on this window are defined as follows;
• Validate - ISIS Mapper checks it can successfully connect to the selected WMS Server and reports back the available layers for you to review prior to adding them to your view
• Add - Prompts you to define a new link to a WMS Server
• Edit - Enables and existing WMS Server link to be edited
• Delete - Removes an existing WMS Server link
• Save - Saves your defined list of WMS Server links as an XML file to be recalled in a subsequent ISIS Mapper session. Note that if the file is named 'WMSservers.xml' it will be automatically loaded into the WMS server links table.
• Load - Enables you to load a previously defined list of WMS server links for you to then select one to load as a new layer in your view.
• OK - After validating a selected WMS Server link this button proceeds to load the associated data as a new layer in your view. You will be prompted to provide a filename for your new layer (which is given a '.wms' file extension)
• Close - Closes the WMS server window without adding any data to your view.

• Click the Add button to define a link to a new WMS server. A new window is displayed as shown below:
  • To link to a WMS server enter its path in the URL field, e.g.:
    http://myServer/MapServer/WMSServer
  • And then enter an appropriate description of the WMS layer in the name field, e.g.:
    Our WMS Server
  • Clicking OK will then add this to the list of available servers (note that if no user name or password is required, these fields should be left blank).

• Before the WMS server can be used the connection must be checked by ISIS Mapper. Click the Validate button to perform this check on the selected server. If this is successful the available layers will be displayed in the Layers box, as shown below. Highlighting a layer in the displayed list will display the associated layer details (to the right of the list). Note this tool enables not only selected layers within the server to be loaded, but all available, compatible layers as well.

• To add the selected layer to your ISIS Mapper view, click OK on the Create WMS Layer window. You will be prompted for a name for your layer. This is an XML format file with extension '.wms' (see further description below). ISIS Mapper will then add this file to your ISIS Mapper viewport as a single new layer. This is shown below; however it should be noted that if you have other layers already loaded in your viewport you may initially see nothing in your viewport. In this case change the zoom level or pan a short distance to prompt the WMS server layer to refresh.
• The '.wms' file is an XML format file that is created by ISIS Mapper when the defined server link is successfully validated. This XML file describes the WMS server capabilities (i.e. number of layers, image format, projections etc.). For subsequent sessions this 'wms' file can be loaded directly by selecting Layer > Load from the main menu (and viewing 'All supported files' in the browser).

• The WMS layer displayed is zoom-dependent so as you zoom in the view will be refreshed, as shown below. Each refresh involves passing a request for data across to the server so the operation may take a few seconds to complete.

You can now add your model related data to display over the top of the WMS layer (useful tip: use layer properties to make raster grids partially transparent to still show the WMS data underneath)

Useful tip: To avoid having to go through numerous calls to the server as you zoom to your desired location (e.g. from an initial wide-coverage map) it is recommended to first load your other 'local' data and use the zoom to layer tool to move directly to its location. Alternatively simply load your local data prior to adding your WMS layer (in this case the WMS layer sometime will not initially show any data in your view - this is solved by changing the zoom or pan slightly to initiate a refresh).

ISIS Mapper provides some editable properties for your WMS layer. These are accessed by right clicking on the WMS layer in the table of contents and selecting the Properties option from the displayed menu, as shown below:

The most important property is the Refresh Criteria. This can be adjusted between 1.0001 and 2 and defines how often the WMS data layer is refreshed. If it is set to a low value then the map will refresh with even a small change in zoom level or pan.

Other key properties are:

• Get Map Dynamically - set to false to prevent automatic re-querying of WMS server as you pan or zoom to different areas of your viewport. Switch back to true to start refreshing the WMS data again.
• Time Out - time in seconds that ISIS Mapper will wait while trying to make a connection to the specified WMS server (during refresh operations).

What is it?

One of the greatest sources of uncertainty in hydrodynamic model outputs is associated with estimating hydrological inputs in ungauged locations. With modern drives to apply a probabilistic approach to modelling results, as opposed to the more traditional deterministic approach, a tool has been developed in ISIS v3.5 to quantify uncertainty in ReFH hydrological boundaries. This article gives a quick step-by-step guide to using this new tool.

The new tool uses a stratified uncertainty analysis applied to ReFH model parameters to illustrate how uncertainty in inputs propagates through to uncertainty in model outputs, in this case hydrograph generation. In a staged analysis, a sub-set of ReFH hydrographs are then typically routed through a hydrodynamic model in order to quantify uncertainty in the model results.

How is it applied in ISIS?

Running the ReFH uncertainty analysis is a two-stage process within the ISIS interface. The first consists of the generation of a number of hydrographs (defaulting to 33) which can then be ranked in order of peak flow, time to peak or flow volume in order to determine exceedence.

From this, the modeller may then select a smaller subset, for example based on required exceedence percentiles, for which to run a full hydrodynamic simulation. Uncertainty is therefore estimated by relating a percentage exceedence to model output.

Step by Step Guide

Load your model into the ISIS interface, note your model MUST contain at least one ReFH boundary in order to apply this analysis

Select Run > Probabilistic ReFH analysis



From the ‘Parameters’ tab, select the ReFH boundaries (at least one) in the model at which to apply the distributions.



All parameters defined on this tab can be left as the default settings (changing these is recommended for advanced users only - refer to ISIS Help for more information). This then completes the setup of the uncertainty parameters to be applied to the hydrology. The next stage is to generate the hydrological inputs.

Select the "Hydrology runs" tab and choose a base event file (ief file) - this will define the model run parameters to be used to run the hydraulic simulations. This will ultimately be augmented with the sampled ReFH boundary data when the hydraulic simulations are run.



On this tab, click "Run hydrology". This then generates the hydrological inputs for each of the selected boundary units, generated from different combinations of the defined parameters. Click on the "View detail" button to review generated boundary data.

Once the hydrology runs are complete, you can use this information to choose the inputs for the full hydraulic simulations. For instance, one can rank the hydrological inputs by total volume, peak flow or time-to-peak and select hydraulic simulations based on exceedences of these.

Summary data for the generated hydrological boundaries can be seen in the "Hydraulic Runs" tab. To select which of these to be run as full hydraulic simulations, tick the "Include in run" box. Your selection may be guided by required exceedences for particular run parameters as described above. When selection is completed; click the "Run" button to perform all selected simulations in a single batch run.



The run summary then appears on the "Results" tab, detailing where to find the results - csv files of maximum water levels are automatically produced, and further analysis may be performed on the ISIS results as usual, e.g. via the user interface or TabularCSV post-processor.



For further information on applying this method see the associated topic, Probabilistic ReFH modelling in ISIS, in the ISIS Help.

This new feature in ISIS v3.5 has been implemented by popular request, enabling modellers to quickly view differences between two model data files.

This is of particular importance for reviewing and investigating changes to models. The difference tool operates on a number of levels - a visual view of the units which have been added, deleted or changed, details of the parameters which have been modified within each unit and the option to display the text differences between the raw data files.

To use the comparison tool:

1. Open the first data file within ISIS.
2. Select File>Compare and choose the file with which you want to compare the first file.

If no differences are found, then the following dialogue appears:

If differences are found between the two files, the initial Model View is presented as follows:

The green colour indicates units that contain differing data between the two files; a mauve colour indicates units which are present in the first data file but not the second; a pink colour indicates units which are present in the second data file but not the first.

Switching to the "Text View" tab will display the differences in text between the two data files:

Again colour coding exists for individual line modifications (green), and lines not present in one file or the other (pink and mauve).

Lower pane - unit differences

The lower pane shows details of the differences between the individual model units and will pertain to the currently highlighted unit.

Tip: Use the arrow buttons at the top to navigate to the previous or next difference.

Within this pane, the Model Tab displays the individual parameters within the unit which have changed.

One can use the info button to display the individual unit forms, with the modified data highlighted:

Note: if the differences are within tabulated data, then the message "Grid contents differ; click info button for details" is displayed. Again, using the info button displays the differences in the table within the unit form:

The raw data tab shows the differences in the data file text for that particular unit. This may be particularly useful in displaying differences in tabular data:

Note: the summary information on the status bar at the bottom displays the number of lines of the data files which match (Matches), differ (Modifications) and only present in one file (Additions and Deletions) for the current unit.

Following on from the last newsletter's 'Tip of the Month', where Dr. Mark Bailes introduced the concept of batch running ISIS simulations, this month we take this further to show batch post-processing of ISIS simulations.

Batch processing simply enables you to extract the same information (e.g. Peak Water Level at Cross Section 1) from a whole series of ISIS Results Files (.ZZN). Typically you might be interested in Peak Water Level for the 2, 5 and 10 year fluvial flow.

Two approaches to batch processing ISIS simulations are presented below. It is a matter of personal preference which you choose. Approach 1 is slightly more complex to set up - but is more flexible. Approach 2 is easy to set up - but less flexible.

Approach 1 - Running ISIS models and then processing the results afterwards

• Launch Tabular CSV from the Tools Menu and browse to an ISIS Results file (.ZZN)
• Set up Tabular CSV to extract the Information you want e.g. Peak Stage at Cross Section 1
• Then rather than clicking on the run button instead select 'Save Settings' from the File Menu. This will create a Tabular CSV Script File (.TCS). This informs ISIS of the results that you wish to extract.
• In a text editor (e.g. TextPad) create something similar to what is shown below:

Please note that you will need to specify the location of your .ZZN files and the .TCS file created earlier. You can add as many lines as you wish and will need one line for each .ZZN file.

Once complete Save the File as .BAT (rather than .TXT).

In Windows Explorer double click the .BAT file and it will execute. Tabular CSV will then create a .CSV file from each .ZZN file based on the settings you specified in the .TCS file.

Tip: Advanced users might want to set up a whole series of .TCS files (e.g. one to extract peak level and one to extract peak flow). These can then be referenced in the .BAT file as needed.

Approach 2 - Running ISIS models and processing the results at the same time

• Launch Tabular CSV from the Tools Menu and browse to an ISIS Results file (.ZZN)
• Set up Tabular CSV to extract the Information you want e.g. Peak Stage at Cross Section 1
• Next, rather than clicking on the run button instead select 'Save Settings' from the File Menu. This will create a Tabular CSV Script File (.TCS). This informs ISIS of the results that you wish to extract.
• In the ISIS Run Scheduler Form, you need to specify both the .IEF files you wish to use and select 'Auto Postprocess' tick box. You also need to specify the file path of the .TCS file you wish to use to post process the results.
• Once you start the Run, ISIS will write the hydraulic results to the .ZZN file and post process those results to create a .CSV file.

See Figure below:



Tip: It is only possible to specify one .TCS file using Approach 2 - so you can only extract one set of results (e.g. Peak Stage). If you wish to extract a series of variables, Approach 1 is likely to be more appropriate.

Many of you will be familiar with the concept of running a batch of models one after another. This is done using the Batch Simulation option on the Run menu. A series of ISIS Control Files (.IEF) are loading into the ISIS Run Scheduler Form and when instructed ISIS will then run one after another. In this month's Tip of the Month Halcrow's Mark Bailes introduces the concept of ISIS Event Files (.IED) which can assist running the same model with different boundary conditions.  This is typically used when running the same model with lots of different return periods.

The Problem

If you construct a hydraulic model (e.g. Model1.DAT) representing your site and you wish to run various return period flows through that model (e.g. 2 year, 5 year, 10 year etc) you could set up a series of .DAT files, each with the different return period flows e.g. Model1_2year.DAT, Model1_5year.DAT, Model1_10year.DAT etc) You then need to run each one of these models with the corresponding ISIS Control File (.IEF) e.g. Model1_2year.IEF, Model1_5year.IEF, Model1_10year.IEF.

A complete set of files would therefore look like:

• Model1_2year.IEF which would reference Model1_2year.DAT
• Model1_5year.IEF which would reference Model1_5year.DAT
• Model1_10year.IEF which would reference Model1_10year.DAT

Whilst it is perfectly acceptable to adopt this approach, problems arise if you wish to change the hydraulic model. You have to make the same change to a whole series of hydraulic models. Not only is this inconvenient and time consuming, it also increases the likelihood of inconsistencies and errors. The recommended approach is presented below:

The Solution

A single hydraulic model is created (including dummy boundary conditions) and saved as a ISIS.dat file in the usual way.

Separately each set of boundary conditions is then saved as an ISIS Event File (.IED), i.e. one for each return period. So you would have a 2year.IED file, a 5year.IED file and a 10year.IED file. Event files are created in the same way as an ISIS Model is created the only difference being when you come to save the file you "Save As" .IED rather than .DAT. It is important to note that the Event file should only contain boundary conditions and the ISIS nodes in the Event file must have the same name as the dummy ISIS nodes representing the boundary conditions in the hydraulic model. This is so ISIS knows to replace those in the ISIS.DAT file with those in the ISIS.IED File.



Having set up your ISIS.IED files the next step is to set up a series of .IEF files in the usual way. The only differences are that on the Files Tab you not only need to specify the ISIS.DAT file you wish to use, you also need to specify the Boundary Conditions you wish to use and a different results file name to which the model results are saved. You do the former by referencing the IED file in the Event Box at the bottom of the ISIS Run Form, and the latter in the "Results File" box.

A complete set of files would therefore look like:

• Model1_2year.IEF which would reference both Model1.DAT and 2year.IED
• Model1_5year.IEF which would reference both Model1.DAT and 5year.IED
• Model1_10year.IEF which would reference both Model1.DAT and 10year.IED

With this approach there is only one hydraulic model to update should you need to make any changes.

Tip: Advanced users might want to store the Fluvial flows in one event file (.IED) and Tidal Boundaries in another event file (.IED). This gives you complete flexibility when combining events together e.g. You may wish to use the 2 year Fluvial flow with the 100 year Tidal Boundary. Having set up your IED Files as before, you simply add both .IED files to the Event Data window (as described above).

ISIS Mapper v3.3 can be used to georeference non-georeferenced cross sections. A fully geo-referenced cross-section is essential to produce a flood map since this is required to create the TIN, from which the flood map is derived.

You can fully georeference cross-sections which are either already partially georeferenced, or those without georeferencing, although the latter also requires a node location shapefile to be available.

Georeferencing non-georeferenced cross sections

Non-georeferenced cross sections are those which do not have the easting and northing coordinates specified for all the points of all or some of the cross sections in the ISIS model file. Please note that a gxy node locations point shapefile should be available for this model in order to use this functionality.


An example of a non-georeferenced cross section shown in the
River Section data entry form in ISIS

Step-by-step guide for georeferencing non-georeferenced cross sections

Step 1 - Import non-georeferenced ISIS data file. Go to Layer > ISIS Data Import > ISIS Model as Shapefile.

Step 2 - Import a GXY node locations file. Go to Layer > ISIS Data Import > ISIS Model as Shapefile. The GXY node locations file should be for the non–georeferenced ISIS data file loaded in Step 1.

Step 3 - Start editing the layer with the cross sections shapefile created at Step 1. Right-click on the name of the cross sections shapefile file in the Table of Contents. Select the menu item Start Edit. 

Step 4 - Click on the menu item 'Georeference Cross Sections from Node Locations'. Go to the shapes details drop-down menu (see screenshot below), then 'Cross Section Operations', then 'Geo-reference Cross Sections From Node Locations'.  The points of each cross section will be placed on a line perpendicular to the river centreline (averaged if the centreline is kinked at this node).



Notes:

• If no match between the node name in the GXY node locations file and a node label in the cross sections file was found, then this particular cross section will not be geo-referenced.
• If there are at least two geo-referenced points in a cross section, then the GXY file node location for this section will not be taken into account (if available) and the georeferencing of this section will be done as for the case of partially georeferenced cross sections (see next section).

Step 5 - Save the shapefile or convert it into a file in ISIS format. In order to save the shapefile, click on the 'Save Shapefile' button (second button from the left) on the 'Shapefile Editor Toolbar'. In order to convert the shapefile into a file in ISIS format (.dat file), go to the 'Shapes Details' drop-down menu, then 'Cross Section Operations', then 'Save as ISIS File'.  After saving the file in one of the formats mentioned above, you can exit the editing mode of the shapefile by clicking on the 'Stop Editing Shapefile' button on the 'Shapefile Editor'.

Georeferencing partially georeferenced cross sections

Partially georeferenced cross-sections are those cross sections which have easting and northing coordinates specified for only some of the points (for all or some of the cross sections in the ISIS model file).


An example of a partially georeferenced cross section shown in the
River Section data entry form in ISIS

In order to georeference partially georeferenced cross sections, the user has to import an ISIS Model file into a shapefile (by selecting the menu items 'Layer > ISIS Data Import > ISIS Model as Shapefile'), enter the 'Edit' mode for this file (right-click on the file in the table of contents and select the menu item 'Start Edit'), and then select the menu item 'Geo-reference Partially Geo-referenced Cross Sections' from the 'Shapes Details' dropdown menu.

In this case georeferencing is based on the coordinates of the two furthest available geo-referenced points in the cross section.  A line is drawn between these points and other points of the cross section are placed along this line taking into account the information from the model file about this cross section.   

For more details about this process please refer to the page 'How do I georeference non-georeferenced cross sections' in the web version of the ISIS Mapper manual.

This tool will find the optimal storm duration (OSD) for one or more FEH or ReFH boundaries, i.e. that storm duration which will give you a maximum stage (or flow) at a particular node in your model. All other parameters in the boundary will not change.

Where can I find it?

On the Run menu in the main ISIS interface.

What do I need?

A valid ISIS model containing FEH and/or ReFH boundaries and an ISIS Event (run parameter) file (*.ief).

How do I use it?

After loading the data file into ISIS and selecting the OSD option, you will be presented with the OSD form.

• On the left, select the FEH/ReFH boundary(s) for which you require the storm duration to be optimised. (Use shift-click to select multiple boundaries).
• From the next table, select the [single] node at which you require the optimal value to be found.
• At the top, select whether to optimise Flow or Stage (default) at this node, and to find the maximum (default) or minimum of this value.
• For Optimising method, "Brent" performs an intelligent search; "Exhaustive" tries each possible duration in turn.
• Select the bounds for the storm duration within which the optimal duration should lie; it is also advised to select an Initial duration - one can select the "Use current" box to read the initial storm duration from the data file.
• By default, the simulation duration used will be that in the ief file; one can change this to be relative to the storm duration, but ticking the "Set simulation period as a multiple of storm duration" box. This guards against setting a simulation duration that is too short for a long storm duration. A suggested value for this is 1.2, but should be greater if the point of interest is a long way downstream of the inflow(s).
• Enter the name of the ief file in the "Base ISIS event file" box. All parameters within this will be used, excepting the simulation time (if the above box is checked). 
• Hit the "Optimise" button to run.

Output

When the optimal duration is found, the results (optimised variable against duration) are displayed in graphical and tabular form, on separate tabs of the OSD form.



More details

The process involves running simulations of varying storm durations until the optimum duration is found. Hence, it could be time-consuming if the model run-time is long. To this end, the Brent method is employed (if selected), which converges to a solution within few simulations. Simulations containing ISIS event data files (*.ied) are permitted.

Constraints

• The storm durations used are discrete, and must be valid storm durations (i.e. odd integer multiple of the data interval).
• User input rainfall profiles can not be used.