UK EA benchmark 4 (Water Module)

This page reports on the performance of the Tygron Platform's Water Module by means of the UK EA Benchmark Test 4 – Speed of flood propagation over an extended floodplain.

The objective of this test is to assess the package’s ability to simulate the celerity of propagation of a flood wave and predict transient velocities and depths at the leading edge of the advancing flood front. It is relevant to fluvial and coastal inundation resulting from breached embankments.

Description

This test is designed to simulate the rate of flood wave propagation over a 1,000 x 2,000 m floodplain following a defence failure (Fig. (a)). The floodplain surface is horizontal, at datum (= 0 m). One inflow boundary condition will be used, simulating the failure of an embankment by breaching or overtopping, with a peak flow of 20 m³/s and time base of ~6 h. The boundary condition is applied along a 20-m line in the middle of the western side of the floodplain.

Figure (a): Modelled domain and the locations of the 20-m line of inflow, 6 output points, and the aimed for 0.1-m and 0.2-m contour lines at t = 1 h (dashed) and t = 3 h (solid), respectively.

Figure (b): Hydrograph applied as inflow boundary condition.

Boundary and initial condition

• Inflow boundary condition as shown in Fig. (b)
• All other boundaries are closed
• Initial condition: dry bed

Parameter values

• Manning’s n: 0.05 (uniform)
• Model grid resolution (m): 5 (or ~80,000 nodes in the area modelled)
• Simulated time (h): 5

Required output

Point ID	X	Y
1	50	1,000
2	100	1,000
3	200	1,000
4	300	1,000
5	300	1,000
6	300	1,300

• Software package used: version and numerical scheme
• Specification of hardware used to undertake the simulation: processor type and speed, RAM
• Minimum recommended hardware specification for a simulation of this type
• Time increment used, grid resolution (or number of nodes in area modelled) and total simulation time to specified time of end
• Raster grids (or TIN) at the model resolution consisting of:
  • Depths and at t = 30 min, 1 h, 2 h, 3 h and 4 h
  • Velocities (scalar) at t = 30 min, 1 h, 2 h, 3 h and 4 h
• Plots of velocity and water elevation v. time (suggested output frequency: 20 s) at the 6 locations represented in Fig. (a) and provided as part of dataset

Dataset content

• Upstream boundary condition table (inflow v. time). Filename: Test4BC.csv
• Location of output points. Filename: Test4Output.csv

The model geometry is as specified in Section 2. No DEM is provided, as the surface elevation is level at datum (= 0 m).

Technical setup

Figure 1. The relative positions of the measurement points used in this test.

• Flat surface
• Grid-cell size (m): 5
• Area size (m): 1,010 x 2,010 (required domain of 1,000 x 2,000 + 5-m border)
• The measurement points were positioned correctly (see Fig. 1)

In order to regulate the boundary discharge according the hydrograph (Fig. 2), 2 inlets were implemented. Both inlets occupied one grid cell, one of these located above and the other below the green center line (Fig. 3). The inlets were configured as follows:

• External area (m²): 1,000,000,000
• Water level (m): 1
• Threshold (m): none
• Inlet Q (m):

Figure 2. Hydrograph displaying the implemented individual and combined inlet fluxes.

Figure 3. Positions of the inlet cells (red squares) with respect to the center line of measurement (green).

Results

Stats

• Software package used: Tygron Platform
• Numerical scheme: FV (Kurganov, Bollerman, Horvath)
• Specification of hardware used to undertake the simulation:
  • Processor: Intel Xeon @2.10GHz x 8
  • RAM (GB): 62.8
  • GPU: 2x NVidia 1,080
  • Operating system: Linux 4.13
• Time increment used: adaptive
• Grid resolution (m): 5
• Simulation time (s): 29 for 900 timeframes
• Object flow (m³/s): 283,723.8
• Remaining water volume (m³/s): 283,606.9

Raster grids (or TIN) at the model resolution for water level and flow velocity

Contours of inundation extent

• 

  t = 30 min
• 

  t = 1 h
• 

  t = 2 h
• 

  t = 3 h
• 

  t = 4 h


• 

  Results of other packages for t = 1 h and 3 h.

Cross sections

• 

  Water level across center line at t = 1 h.
• 

  Water level across center line at t = 1 h for other packages.


• 

  Flow velocity across center line at t = 1 h.
• 

  Flow velocity across center line at t = 1 h for other packages.

Time series of water level and flow velocity

• 

  Water level at Point 1.
• 

  Water level at Point 1 (others).


• 

  Flow velocity at Point 1.
• 

  Flow velocity at Point 1 (others).


• 

  Water level at Point 2.


• 

  Flow velocity at Point 2.


• 

  Water level at Point 3.
• 

  Water level at Point 3 (others).


• 

  Flow velocity at Point 3.
• 

  Flow velocity at Point 3 (others).


• 

  Water level at Point 4.


• 

  Flow velocity at Point 4.


• 

  Water level at Point 5.
• 

  Water level at Point 5 (others).


• 

  Flow velocity at Point 5.
• 

  Flow velocity at Point 5 (others).


• 

  Water level at Point 6.
• 

  Water level at Point 6 (others).


• 

  Flow velocity at Point 6.
• 

  Flow velocity at Point 6 (others).

Notes

• The stepwise behavior seen in the velocity profiles can be related to the definition of the inlet inflow, which is also stepwise.

Public test project

A copy of this test project will soon be made available to everyone with access to the Preview Server under the name Uk Test 4, providing hands-on insights to those interested. In this project the discussed benchmark test is reproduced, though be it with a cropped version of the map (i.e., 1,000 m x 2,000 m), which may result in tiny deviations of the measurements. The special map dimensions mentioned earlier (i.e., 1,010 m x 2,010 m) are not selectable in the Wizard and must be configured through the API.

The locations of the measurement points are:

Point ID	X	Y
1	50	-1,000
2	100	-1,000
3	200	-1,000
4	300	-1,000
5	300	-1,000
6	300	-700

Note that the coordinate system in this project differs from the one used during the actual benchmark testing as shown in Fig. (a). For the above listed points the center (0,0) is located at the top-left corner instead of the bottom-left, hence will the bottom-right corner have coordinates (1,000;-2,000).