This Report details the results of a hydraulic model investigation
into flooding along the Georges River at East Hills. The
investigation was undertaken by Manly Hydraulics Laboratory as part
of a study by the Rivers and Ports Branch for Bankstown City Council.
A physical model of horizontal scale 200 and vertical scale 50 was
constructed using recent hydrosurvey and ground survey information.
The model study was conducted in two stages separated by a gap of
three years. For the second stage, the design floods had been reassessed. An additional flood event in August 1986 was available
tor verification.
The first stage involved:
- Tuning the model using flood data from the 1956 and 1978
floods, as well as tidal data;
- ModeI testing a range of flood discharges and tailwater levels
encompassing an early estimate of 1 in 100 year flood;
- Obtaining river channel water levels and velocities for the
full range of tests;
- Determining overbank velocities and detailed flow paths for a
discharge of 2100m3/s and a 4.7m tailwater at Deadmans Creek;
- Testing the effects of four flood mitigation options (i.e.
Options I to 4) under the central design flow conditions.
The second stage was prompted by requirements to test additional
management options for flood discharge and peak flood levels. This
stage involved:
- Testing the accuracy of the model with respect to Stage I using
the previous I in 100 year flood condition of 2100m3/s
discharge and 4.7m tailwater.
- The verification of the model against the recent 1986 flood.
- Testing a more recently derived 1 in 100 year flood with a
discharge of 2300m3/s and corresponding tailwaters of 4.Sm and
4.Im at Deadmans Creek. Overbank velocities and flow paths
were obtained in addition to water level measurements.
- Testing the effects of several new flood mitigation options
(i.e. Options 5 to 8) using the new I in 100 year flood
discharge and both tailwater levels.
The model satisfactorily reproduced the 1986 floods and showed that,
under existing conditions, approximately 80 properties in the study
area experience inundation. Properties near the East Hills Footbridge experience depths of up to 4m. Flow velocities of up to 2.5m/s were measured adjacent to several houses.
By using several combinations of finger levees (Options 2, 3, 4, 6A
and 6B) various degrees of flow velocity reductions were achieved
without changing the level of inundation. The two options involving
ring levees (1 and 5) prevented inundation up to 1 in 100 year flood
conditions in the selected areas.
All of the levee options caused minor upstream afflux but this had
attenuated to within the accuracy of measurement at Milperra Bridge.
Options 7 and 8 were purely non-structural, involving house removal and house relocation.
