We use cookies to make your experience of our website better. You can delete or block cookies, but some parts of this website won't work without them. By using this website you accept our use of cookies.

12/03/2015 12:55:01

One of our users recently asked the question: Is there an easy way of generating hydrographs, in particular peak flow values, for a batch of hydrological boundaries, rather than clicking on the Generated Hydrograph/Calculated Results tab and reading the values off manually?

The answer of course is YES, else this would be a pretty short non-tip of the month.

This is particularly useful if you want to generate and assess a batch of design inflows for a bunch of return periods and/or storm durations.

This tip makes use of the Boundary Mode functionality within Flood Modeller, which processes the boundary data only, as opposed to running a full simulation. Indeed it is this which runs “behind the scenes” when you click on the Generated Hydrograph tab for a hydrological boundary. Since the results are created in the same format as for a full simulation, you can use all the standard tools to postprocess the results.

There is more than one way of skinning this particular cat, but all are essentially variations on a theme. The outline process is to edit the data, run the “simulations” and process the results.

We start by assuming we have a model set up with hydrological boundaries – because we’re running in boundary mode, the model at this stage does not even have to be fully valid (as long as all your boundaries are). It can, if you like, consist of hydrological boundaries and nothing else.

The example shown here consists of two ReFH (Revitalised Flood Hydrograph [UK] method) boundaries, although any type or combination, e.g. US SCS (Soil Conservation Service), FEH (Flood Estimation Handbook [UK]), etc. of hydrological boundary can be used.

We will first demonstrate how to run for a single rainfall duration and return period and then proceed to illustrate how this can be repeated for multiple combinations of these.

Step 1.1 – enter your boundary data

Create or edit a hydrological boundary for each inflow location. Enter the catchment and method details, importantly including your return period and storm duration for your hydrological boundaries. In this instance, I’ve set up a 100-year return period and storm duration of 11 hours.

Then save your network data file.

Step 1.2 – set up and run your model in boundary mode

Create a new simulation. Right click on the Simulations item in the Project Panel and select New 1D Simulation. Ensure that the “Boundary Mode” run type is selected, the finish time of simulation extends at least as far as your quick runoff hydrograph and the timestep is short enough to capture the detail, but not so long as to slow down the calculations – 300s or 900s is usually appropriate.

Then hit the run button to perform the computations – this should take no more than a few seconds.

Step 1.3a – view results

To view the results graphically, you can use the time series functionality (right-click on the boundary and select “Time Series”; ensure “Flow” is selected for the output Variable). NB You can right-click on the graph to allow you to add results from other boundaries or other runs for comparison purposes. 

Note – if you select Stage as the output variable or select output for any of the internal model nodes, you’ll see default null values (-9999.99) – this is because boundary mode simulations do not calculate any values for these.

Step 1.3b – process results in tabulated form

You can also view results in tabular form using the TabularCSV tool (Results tab > Tabular CSV), for either the full hydrograph or the peak flow value.

Select “Flow” as your input variable and choose between Simulation Results (for full hydrograph) and Maxima and Minima as appropriate. The other settings may be left as defaults, although if this is part of a large model, you may wish to select only the boundary units from the Nodes tab to output.

Hint: If you right-click on the tool and select “Save Settings”, you can store these settings in a TabularCSV script file (*.tcs), which will be useful when batch-processing the results (see later).

Then hitting the run button will pull the requested values into a comma separated value (*.csv) file for further analysis.

Note: Running boundary mode simulation also creates *.zzb and *.zzh files, which displays summary input information and computed hyetograph/hydrograph information respectively, which can be viewed in any text editor.

So now we have processed and viewed hydrographs for a single model. Of course, one could copy the boundaries and/or edit these for a different return period or storm duration and repeat this process, but I think it is more logical, as well as more traceable, to create a new data file, or event data file (*.ied) for each distinct design event.

Step 2 – Using Global Edit to change return period / duration

From the Toolbox panel, select Tools > Global Edit > 1D Networks > ReFH Boundary. Hint: You can right-click on this option and select Add to Favourites so it’s accessible by a single click (from Toolbox > Favourites) in future.

Select Return Period or Storm Duration as appropriate (NB If you want change both, you’ll have to repeat this process for each).

Then select “Change to set value” and enter the value of the new Return Period (in years) / Storm Duration. Unless you’re changing specific boundaries only, uncheck both the “where value equals” and “apply only to selection” boxes.

When you press OK, check that the dialogue reports that the expected number of units has changed

Save the data file as a new name, e.g. HydroBoundaries50.dat, using the Home > Networks > Save As ribbon menu item. Care: Check you’re saving the network (*.dat) file not the project (*.fmpx) next door!

Then create a new simulation file to run this – you can either repeat the steps in 1.2 but why not use the copy facility to clone the existing run file? Right click on the simulation file created in Step 1.2 and select Copy. Save as a new name and then double click to open the simulation window. It is important here to remember to change the data file name in the Files tab to that created above and to Save the simulation file again.

Repeat this process for each combination of storm duration and return period you wish to run design events for.

Step 3 – Running and batch processing multiple boundaries

Once all the boundary files and simulation files are set up, we can use the Batch processor to run them together. On launching the batch processor window, simply add all the simulation files (*.ief) created in Step 2 to the table.

If you chose to postprocess the results in TabularCSV and created the tcs file in Step 1.3b, check the “Auto Postprocess” option and enter the name of the *.tcs file in the box below this.

Hit Run to commence the batch, which will then process the boundaries. If the Auto Postprocess option was chosen, then the appropriate the csv files will also be automatically generated at this juncture.

Additionally, you can use Flood Modeller’s inbuilt graphical tools or view the textual summary files as described in Step 1.3 to inspect the results.

Did you know?

We also have bespoke tools in Flood Modeller for:

  • deriving the Optimal Storm Duration (e.g. that which produces the maximum water level or discharge at any location in the model) for a given return period (Hydrology > Optimise Storm Duration)
  • quantifying the combined uncertainty within a number of parameters used in deriving a design ReFH hydrograph (Hydrology > Probabilistic ReFH)

The future

We are already looking into the future for Flood Modeller v4.1 and into better ways of reducing the repetition involved in Stage 2, for both full hydrodynamic and Boundary Mode simulations, and therefore streamlining the process even further.

This tip of the month has been developed based on using Flood Modeller Pro, available from April 2015. However, the concepts still apply to legacy versions of the software. Contact us for further details.

Privacy & Terms

Copyright