Flood Frequency Analysis with Web-based Flike

8 Dec 2015 – Flike is no longer available over the web.  BMT WBM have released a version of Flike that can be downloaded following the payment of a licence fee.

See http://flike.tuflow.com/

See this blog for information on tutorials for this version of flike.

Outdated tutorial on the web version of Flike…

Since 2006, the National Committee on Water Engineering have recommended the use of Flike software for flood frequency analysis.  Details are in the new draft  Chapter 2, Book 3 of Australian Rainfall and Runoff.

There are plans for a commercial release of the software soon but until that is released a free web-version available at:

The following steps are a guide to using this web version. Only try this if you are using Windows.

1. Browse to the Flike website and click on the ‘Use Flike’ tab.  

The screen is shown in Figure 1. I’ve highlighted two items in red boxes.  Create a folder and then click on ‘CLICK HERE’ to download the two executables into that folder – Flike_Preproc.exe and Flike_postproc.exe.

Figure 1. Use Flike screenshot

Figure 1: ‘Use Flike’ screenshot – click to download the executables

2.  Create a csv file with annual flood data

Use excel or a text editor to make a csv file that specifies the annual flood values for the site you are interested in.  Put year in column 1 and flow in column 2.  Don’t include headers for the columns.

An example of a few values for the Tyers River at Browns (226007) is shown below.  Save this file with a csv extension e.g. Tyers@Browns.csv.  Although I’ve put headers in this table, ‘Year’ and ‘Peak flow (cumec)’, don’t include them in your .csv file.  An example file is shown here.

Note that you can set the file up in other ways, for example with headings, but that requires you to change the defaults when the data is read in.  See section 3.7 below for the ‘Import gauged values’ dialogue.

Year Peak flow (cumec)
1962 33.5
1963 35.1
1964 36.2
1965 25.4

3.  Pre-process your data

3.1 Run Flike_Preproc.exe.
3.2 Click File/New (Figure 2)

Figure 2. Pre-Proc.exe

Figure 2: Pre-Proc.exe, File/New

3.3 Browse to a folder and enter the name of a file e.g. Tyers-test.fld (Figure 3) and click Open

Figure 3. Enter a file name

Figure 3: Enter a file name

3.4  Click Yes to ‘Create this file’

Figure 4. Create the .fld file

Figure 4: Create the ‘ .fld’ file when asked

3.5 Complete the TUFLOW FLIKE Editor dialog

A new dialogue will appear; the TUFLOW FLIKE Editor.  In the ‘General’ tab, add a Title e.g. ‘Tyers River at Browns’.  Look at the other options but leave them at their defaults for now.  If you don’t understand the various options, read the draft ARR flood frequency chapter.

Figure 5. Edit the TUFLOW FLIKE Editor dialog 'General' tab

Figure 5: Edit the TUFLOW FLIKE Editor dialog ‘General’ tab

3.6 Add the flood flows

Click on the ‘Gauged Flows’ tab and then click on Import (Figure 6)

Figure 6: Click to import flows

Figure 6: Click to import flows

3.7 Import flow from the csv file

Browse to the csv file you made in step 2.  Have a look at the various options in the dialog but leave everything at its default (Figure 7).  Click View to confirm you have the correct file in the correct format, then Click OK to import the annual flood data.

Figure 7: Browse to the csv file with the annual flood data

Figure 7: Browse to the csv file with the annual flood data

3.8 Confirm the data are as expected

A new dialog will appear.  Check that the data has been imported as you expected (Figure 8). Click Rank, then select the ‘Descending flow’ radio button.  Make sure the largest floods are correct as they are highly influential for flood frequency analysis.  When you are satisfied with the data are as expected, Click OK.

Figure 8: Check the data have been imported correctly

Figure 8: Check the data have been imported correctly

3.9 Save and finalise the file

Save and Finalise as shown in Figures 9 and 10.  You need to complete both these steps in order.

An example of a ‘.fld’ file is available here.

Figure 9: Save the file

Figure 9: Save the file

Figure 10: Finalise the file

Figure 10: Finalise the file

4. Run “Engine on the website”

4.1 Return to the Flike website and the ‘Use Flike’ tab.

4.2 Click ‘Choose File’ and browse to the file you made with the pre-processor (step 3) (see Figure 11).

Figure 11: Choose File and Run Engine

Figure 11: Choose File and Run Engine

4.3 Click ‘Run Engine’

After you run the engine, Flike will make an output file that looks something like the one below (Figure 12). (The complete file is here).  Note that the file name ends in ‘_out.txt’. Save this with the ‘.fld’ file you made with the pre-processor.

The output file contains all the results.  We can now post-process the file to prepare a flood frequency plot.

Figure 12: Flike output

Figure 12: Flike output

5. Post-process the Flike output

5.1 Run the post-processor you downloaded previously – Flike_postproc.exe.

5.2  Click ‘Open and plot’ (Figure 13) and then you will need to browse to the files you made previously.  First the ‘ .fld’ file then the  ‘_out.txt’ file (Figure 14 and 15)

Figure 13: Click 'Open and Plot'

Figure 13: Click ‘Open and Plot’

Figure 14: select '.fld' file

Figure 14: select ‘.fld’ file

Figure 15: select '_out.text' file

Figure 15: select ‘_out.txt’ file

5.4 Click OK and your flood frequency graph will be plotted.  An example is shown in Figure 16.

Figure 16: plot of frequency analysis

Figure 16: plot of frequency analysis

So what is the value of the 1% (100-year ARI) flood?  Its difficult to determine from the graph but we can see that it is between 102.265 and 102.633 i.e. between 184 and 429 m3/s.  Go back to the ‘_out.txt’ file, scroll down, and you’ll see the results (Figure 17). The value we need is the 100-year recurrence interval, expected parameter quantile which is estimated to be 244.78 m3/s with 5% and 95% confidence limits of 167.94 and 429.4 m3/s.

Figure 17: Flike results, 1% (100-year ARI) flood

Figure 17: Flike results, 1% (100-year ARI) flood

Those are the basic options.  Flike has many advanced features including the ability to select different probability models and use censored values.

Check this blog for suggestions on how to make a nicer looking frequency plot.

8 thoughts on “Flood Frequency Analysis with Web-based Flike

  1. Silver Yance

    The Flike program works pretty good and is easy and friendly to use..! However, without the tutorial it was hard to figured out. Thanks for the effort in developing this very useful program.!

    1. Silver Yance

      Hi Tony,
      I’ve tried to run the program for the annual maximum of Bowen River at Myuna GS, but the program somehow it does not give me a result, it does but I’ve got the output file with a unreadable script. Any suggestion would be appreciated. I have tried to attach the file but I do not know how.

  2. tonyladson Post author

    The most common place to have problems is step 3.9. Make sure you save and finalise your ‘fld’ file. Check that your fld file looks like the example.

  3. Karl

    If one year of peak flows (from the annual data peak flow download) has no data, does that year (row) get deleted or do I check sub-year data for that year? Or, do I change the default zero flow for that year to 1 m3/s so that year has a value (this will affect the result of course).

  4. tonyladson Post author

    It is important to understand why the flows are missing. If you have some information e.g. we don’t know the flow but we know it was less than x (or greater than y), then this can be included as a censored value.

    Some missing values may indicate there was no flow in a particular year. Have a look at sections 2.9.13 and 2.9.14 of the draft Book 3 of Australian Rainfall and Runoff for information on how to deal with Potentially Influential Low Flows (PILFs). The link is http://goo.gl/YwVVEi

    There is guidance on missing records in section 2.2.7 of the 1987 version of Australian Rainfall and Runoff and section 2.3.11 of draft Book 3 of ARR (http://goo.gl/YwVVEi).

  5. Pingback: Web-based hydrology tools | tonyladson

  6. Pingback: Better Frequency Plots from Web-based Flike | tonyladson

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s