Advisian, together with Wollongong City Council and the NSW Department of Planning, Industry & Environment, has recently completed the updated Fairy and Cabbage Tree Creeks Flood Study, providing an improved understanding of flooding and its potential impacts. 

The draft report is now on public exhibition to allow review and comment by the community.

The Fairy and Cabbage Tree Creeks catchment lies in the central part of the Wollongong Local Government Area and drains from the Illawarra Escarpment to the ocean through Fairy Lagoon.  

The study area also includes the ‘Smith Street’ catchment which drains to Belmore Basin. The study area comprises of about 22 km², including the suburbs of Balgownie, Fairy Meadow, Keiraville, Gwynneville and North Wollongong.

The major watercourses in the study area are Cabbage Tree Creek in the north, Fairy Creek in the south, and Towradgi Arm in the north-east.

Numerous smaller tributaries drain into these watercourses.   

The Fairy and Cabbage Tree Creeks catchment has experienced several significant floods over the past decades.

The August 1998 flood was particularly severe and caused many millions of dollars of damage in the Illawarra Region. The flood resulted in one death, numerous evacuations and flood rescues, and required considerable clean-up operations in its aftermath. 

Council has undertaken several previous flood-related studies for the study area including: 

• The 2010 Flood Study – to define flood behavior
• The 2010 Floodplain Risk management Study to investigate flood mitigation and management options, and
• The 2007 Fairy Lagoon Entrance Management Policy, which describes procedures and responsibilities for the management and opening of the lagoon entrance to help reduce flooding in low lying areas of the catchment.  

Council have also undertaken works to reduce the severity of flooding, including the Brokers Road Basin and Foothills Road Basin, among others.

The updated study was triggered in response to a variety of factors; these incude:

• The availability of new Airborne Laser Survey topographic data and additional detailed ground and structural survey.
• Changes to the catchment have occurred since the 2010 Flood Study
• Release of Council’s updated ‘blockage policy’ which guides the modelling of blockage of bridges and culverts by debris, and
•  Advancements in computational and flood modelling capabilities. These advancements have allowed the catchment to be represented at a finer resolution and for flooding to be defined along additional tributaries and flowpaths which were not captured by the previous study.

The study was also required to assess the potential future impacts of climate change on flooding in the study area.

The findings and models developed by the study will be used to inform the subsequent preparation of an updated Floodplain Risk Management Study and Plan which will assess and recommend measures for better managing flood risk in the area. 

An updated flood model was established using TUFLOW - an Australian developed two-dimensional hydraulic flood modelling software which has become the benchmark here and in the United Kingdom.

One of the most important aspects of the model is the proper representation of the catchment topography.   

Aerial Laser Survey Data captured in 2013 in combination with ground and bathymetric survey data was used to develop a detailed Digital Elevation Model of the catchment, with elevations at 1.5 metre spacings.

This provided an excellent basis for construction of the hydraulic model.

Other important aspects of the TUFLOW hydraulic model include;

• The delineation of the catchment into different surface materials types, which dictate how much resistance there is to flood flows.
• The presence of bridges.
• Culverts and stormwater drainage pipes and other structures which may obstruct flood flows, including handrails and road and sound barriers.

TUFLOW uses these layers of information to build a model of all physical aspects of the catchment which may have an important influence on flood behaviour.

The boundary conditions which apply water into the model include numerous ‘local runoff’ inputs shown here in light blue, and tidal water level boundaries at the ocean, downstream of Fairy Lagoon and at Belmore Basin.

Dynamic breakout of the Fairy Lagoon entrance berm was also incorporated into the model, to represent the scouring away of sand once the berm is overtopped, thus increasing the capacity to discharge floodwaters.

With the model now fully functioning, the next step was to assess and refine its ability to replicate observed flood behavior in the study area, through the process of calibration.

Model calibration is an essential step in the flood modelling process, which confirms that the model can reproduce historical flood events and demonstrates its ability to reliably simulate flood behaviour in the study area.

Recorded rainfall and tidal data are used as inputs to the flood model and the resulting model outputs are compared to recorded and observed flood levels. Model parameters are then adjusted to achieve a better match to recorded data.

The August 1998 flood and February 2012 storm were used in model calibration.  

The model calibration to the August 1998 flood recorded rainfall data from some 9 local gauges was used.  

Consideration of both the timing of the rainfall and the spatial distribution of the rainfall was required.

The heaviest rainfall occurred along the Illawarra Escarpment, while rainfall along the coast was lower.  This is typical for the Illawarra Region.

Tidal data recorded at the Port Hacking ocean level station was adopted to provide the downstream boundary condition.

After an iterative process of simulation, review and refinement, an excellent calibration to recorded water levels was achieved at both the Cabbage Tree Creek and Fairy Creek gauges.

The timing, shape and peaks of the recorded flood level hydrographs were well replicated by the model.

Simulated peak flood levels were also compared to a historic flood level database which includes 162 flood levels survey by Council after the August 1998 flood. 

An excellent result was also achieved with simulated and observed peak flood levels within 15 cm at 55% of locations, within 30 cm at 83% of locations, and a mean difference of 8 cm.

To confirm that the flood model’s ability to replicate the August 1998 flood was not just by chance, calibration to the February 2012 event was also undertaken.  

Rainfall during this event was not typical for the Illawarra Region. The storm moved in a south-easterly direction across the catchment with the heaviest rainfall occurring along the southern boundary close to the coast, while rainfall along the Illawarra Escarpment was considerably lower.

A good calibration to recorded water levels was also achieved for the February 2012 flood, including the Balgownie Road, Cabbage Tree Creek and Fairy Creek gauges. However, no surveyed flood level database was available for this event.

After a thorough model build and calibration process, the study moved on to the ‘Design Flood Estimation’ phase with confidence in the model’s ability to reliably simulate flood behaviour in the study area.

From the design flood model results, a detailed understanding of flood behaviour and potential impacts can be developed that will later be used to determine appropriate courses of action to manage flood risk in the area.

Design flood conditions are estimated from theoretical rainfall events that have a particular statistical probability of occurrence, which is expressed in terms of percentage Annual Exceedance Probability, or AEP. 

Flood conditions for the 20%, 10%, 5%, 2% and 1% AEP design events and the Probable Maximum Flood, or PMF, have been investigated in this study.

In the mid to upper catchment flooding is generally ‘flashy’ in nature. Flooding occurs in response to relatively short durations of intense rainfall and flood levels quickly rise and fall over the course of a few hours. 

Lower in the catchment, from about Memorial Drive and the Princes Highway downstream, flooding is more sensitive to the total volume of rainfall, and floodwaters may rise somewhat slower and remain elevated for several hours. Flooding in these areas can also be influenced by the condition of the lagoon entrance berm and coinciding ocean levels.

During the 1% AEP flood event, sometimes known as the ‘1 in 100 year flood’, areas of high hazard that may pose a significant threat to life and property are generally constrained to defined water courses, open channels and flood flow paths. However, there are various exceptions including Exeter Avenue in North Wollongong, properties along the western edge of Memorial Drive in Fairy Meadow, and College Place in Gwynneville.

Flood depths and velocities also become significant along various roads in the study area and would affect vehicle safety and pose constraints for evacuation and emergency response. This includes major roads such as the Princes Highway, M1 Motorway, Memorial Drive and Mt Ousley Road.

During the PMF (or extreme) flood event, the extent and degree of hazard posed to life and property would increase significantly. Evacuation and emergency response constraints would also be significantly worse, with some areas requiring early evacuation to avoid rapid isolation and inundation.

Model Sensitivity and Climate Change analyses were also undertaken.

The investigated sea level rise scenarios of up to 0.9 m would be expected to cause little change to existing flood conditions and impacts, with the exception of flooding of Towradgi Arm, Fairy Lagoon, Fairy Creek downstream of the Princes Highway, and Cabbage Tree Creek downstream of Puckey Avenue. 

The greatest impacts on flood affectation of properties would occur at Achilles and Exeter Avenues, Ajax Avenue, Ralph Black Drive and Montague Street. 

Increases in the intensity of heavy rainfall events that may occur due to climate change would be expected to have a more significant impact on flooding. Flood model results indicate that a 20% increase in rainfall intensity for the 1% AEP event would lead to increases in peak flood level of 0.1 m or more along most tributaries, including significant areas with increases of more than 0.25 m and localised increases of up to about 0.5 m. 

It was found that the revised 2016 blockage policy ‘design’ factors result in a decrease in 1% AEP peak flood levels upstream of many structures relative to the 2002 Blockage Policy, including widespread decreases along the western edge of the M1 Motorway, Memorial Drive and the Illawarra Railway.

The revised 2016 blockage policy ‘risk management’ factors were found to result in higher 1% AEP peak flood levels upstream of many structures relative to the ‘design’ blockage factors.  The magnitude of these increases was often less than 0.1 m, though localised differences of up to 0.45 m were observed.  

Model sensitivity to a 25% change in hydraulic roughness was also tested and found to be low, with changes in 1% AEP peak flood levels less than 50 mm across most of the floodplain. 

Draft versions of the Fairy & Cabbage Tree Creeks Flood Study main report and flood mapping are now undergoing public exhibition and are available for download from Council’s website.

Volume 1 details the project scope, methods and results, while Volume 2 provides a comprehensive set of flood maps, including mapping of flood extents, depths, velocities and hazard.

Following the public exhibition period, submissions from the public will be collated and reviewed to assist in the preparation of the final flood study report.